National Academies Press: OpenBook

Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2 (2014)

Chapter: 4 Measures Reviewed for Each Candidate Domain

« Previous: 3 Identified Candidate Domains
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

4

Measures Reviewed for Each Candidate Domain

The chapters that follow move beyond addressing the Phase 1 objectives to its Phase 2 objectives, as per its Statement of Task, Box 1-4. This chapter’s primary focus is on measurement tools for the committee’s candidate domains—essential ingredients in electronic health records (EHRs) that must be consistently defined and used in order for our health system to assess and to achieve quality health outcomes. Having identified relevant domains in relation to their importance to health and the usefulness of having information on the domain for improving health, the committee turned to reviewing the availability of appropriate measures for each domain. Even if a domain is strongly linked to health and would inform individual or population health or research on health and health care, it could be problematic to include in the EHR without measures that meet the four criteria set by the committee in relation to the measure (see Box 4-1 for the criteria used in selecting measures). The proliferation of measurement sets and reporting requirements can place a burden on both patients and clinical teams. The logistical challenges for routine, harmonized measurement tools are significant, but as described in earlier chapters, opportunities exist toward increasing standardization. Toward meeting this goal, the committee applied criteria 3 to 6 to the 17 candidate domains, along with their subdomains, in reviewing and evaluating measures of those domains. The criteria are identified in Chapter 2 and listed below in Box 4-1.

In this chapter the committee provides definitions for measures and metrics for these domains and discusses the measures it reviewed. The committee did not have time or resources to provide extensive descriptions for all of the measures that exist relevant to each domain. In several instances,

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

BOX 4-1
Criteria Used for Selecting Core Domains and Their Measures as Part of Phase 2

3.   Availability and standard representation of a reliable and valid measure(s) of the domain.

4.   Feasibility, that is, whether a burden is placed on the patient and the clinician and the administrative time and cost of interfaces and storage.

5.   Sensitivity, that is, if patient discomfort regarding revealing personal information is high and there are increased legal or privacy risks.

6.   Accessibility of data from another source (i.e., information from external sources may be accessible to meet the needs of patient care, population health, and research; if so, the domains would have less priority for inclusion in the EHR).

there was a single accepted measure, which had been tested for its reliability, validity, and scoring. In others, multiple measures of a domain were considered if no single measure stood out.

The committee conducted literature searches to collect measurement tools and questions used domestically and internationally for the candidate domains. It relied on peer-reviewed literature in these efforts as well as for identifying a clear and consistent purpose for each measure. The committee considered the usefulness and feasibility of collecting data, the needed frequency of collecting this data, and privacy concerns or other sensitive issues that may exist for collection of those data in EHRs. The committee used its criteria to judge whether patients would consider a question or instrument to be sensitive or if it requested personal information that they may be reluctant to disclose. Privacy issues in collecting and sharing health data are discussed in more detail in a commissioned paper the committee used to inform their decisions. The paper is located in Appendix B of this report.

Finally, the committee examined the accessibility of the data from other sources. Data that are consistently collected externally would have a lower priority for being collected in the EHR. However, linkages between data collected in surveys, such as in the U.S. Census, currently cannot be smoothly integrated into an individual EHR.

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

MEASUREMENT TERMS

For this report, domains refer to behavioral or social determinants of current or future health outcomes. In the social and behavioral sciences, domains are often called constructs, because they are attributes that cannot be measured directly. The operational definition of a domain or construct describes the operations that are used to assess it, which generally involve measurement tools that enable assigning numbers to (social or behavioral) attributes according to rules (Stevens, 1946). Subdomains are dimensions within a domain, each with its own operational definition and measurement tools.

Some constructs can be directly measured. For example, body mass is a function of height and weight. Both are observable variables that can be directly measured using instruments, such as a stadiometer or scale, respectively. Operational definitions of other domains involve asking individuals to respond to structured questions that can yield a numerical value that indicates the presence or absence or extent of the construct. Depressive symptoms or social support are examples of constructs that are assessed through responses to questions. See Box 4-2 for information on instruments and scales.

If measurements of social and behavioral attributes are to be collected in EHRs, they should be recorded in ways that enable interoperability across time and between electronic health record systems. For interoperability to succeed, the data that are intended to measure a given domain must have the same meaning across data sources and be able to be combined. For example, weight is conventionally represented on the kilogram scale. If, however, one dataset represented weight as pounds and the other as kilograms, the data could not be combined without transformation from one scale to the other, because of nonequivalence of the meaning of the numerals used to represent the domain of weight. A 2011 Institute of Medicine report For the Public’s Health: The Role of Measurement in Action and Accountability noted that because data elements are not standardized, individual decision makers base their choices on different information (IOM, 2011b).

By metric the committee is referring to the underlying data structure and scale for any measurement, including social or behavioral variables. A metric is agnostic to the specific instrument used to obtain the data. However, it does specify the properties of the scale on which the measurements of the variable that represents the domain will be expressed. The metric clearly defines the structure of the variable in the dataset and the meaning of the numerical representations of different categories or levels of the variable. A single metric in theory could have numerous instruments all of which provide measurements that are represented on the same standard

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

BOX 4-2
Use of Instruments and Scales

Instruments are tools or procedures used to obtain measurements of a domain. The data produced by an instrument can be represented on four types of scales that differ by their mathematical properties. These include nominal, ordinal, interval, and ratio scales. A nominal scale assigns numbers to categories that have no logical ordering. There is a one-to-one correspondence between a category and its numeric representation. For example, gender and race/ethnicity are measured on a nominal scale, and a single numeral is used to represent each gender and race/ethnicity category. An ordinal scale rank-orders categories such that increasing numbers mean more of some attribute, and vice versa. An example would be a zero to ten pain scale with increasing ratings connoting higher levels of pain. In an ordinal scale the difference between a score of 6 versus 5 is not necessarily equivalent to 10 versus 9. The numbers of an ordinal scale merely indicate more or less of an attribute. An interval scale reflects rank-ordered categories like an ordinal scale, but the differences between numerals are known to be equal. Two interval scales of the same attribute would be linear transformations of one another, such as the Celsius and Fahrenheit scales of temperature. (Most behavioral and social domains are measured on a scale somewhere between ordinal and interval. In such scales, the differences between values may be equivalent across most levels of the variable, but equality of differences may not hold for the full range of values, such as at the tails of the distributions.) A ratio scale refers to equality of ratios, and has a natural zero point. Doses of a given medicine can be used as an example that would be measured on a ratio scale. There is a logical zero point (e.g., no medication) and an equivalent ratio scale of two doses have the same interpretation (e.g., a 10-milligram dose is twice the amount of a 5-milligram dose).

scale. For example, if the metric for weight is the kilogram scale, the instrument used to obtain weight could be patient self-report or observed and reported on by the clinical team based on a physical weighing scale. In both cases, the measurement can be recorded on the same kilogram scale, but the measurement error associated with self-reports may be higher.

The relationship between the domain concept, its subdomains, the metrics, and instruments that provide measurements on the common metric is shown in Figure 4-1. For example, the domain of physical activity has several subdomains (mobility/motion, flexibility, and strength) and various ways of expressing it. The committee chose a metric of metabolic equivalent of task (MET)-minutes to represent the mobility/motion subdomain of physical activity. MET-minutes may be obtained by asking individuals

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

to recall the frequency, intensity, and type of physical activities they have engaged in over a specified prior interval; a multitude of self-reported instruments could be used for this purpose. Alternatively, MET-minutes could be derived from a device called an accelerometer. (See section below on physical activity for more information on accelerometers.) Interoperability is achieved if all the instruments can be represented electronically as the same metric, that is, MET-minutes. This example also illustrates the importance of specifying the metadata associated with how the data were collected and which instrument was used.

The committee used the following process to evaluate the suitability of the measures of each of the candidate domains for inclusion in all EHRs or for special populations. First, a committee member with expertise in the domain identified available measures for the domain or subdomains. She or he summarized the extent to which a measure met the additional considerations set by the committee for adequacy of measures: availability of standard measures and instruments free from intellectual property restrictions; reliability and validity of those measures; feasibility of collecting the data required by the measure; sensitive information or patient discomfort in the information reported; and the potential benefit and risk of including the measure in EHRs. In some instances, the committee chose to use one or two questions from a full set of validated questions. They did this in order to make the domain measures feasible in a clinical setting. In

images

FIGURE 4-1 Examples of domains, subdomains, metrics, and instruments.

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

general, individual questions have already been tested prior to using them as a question set.

The committee combined information regarding the six considerations described above into a rating on four criteria: standard measure, usefulness, feasibility, and if the measure includes sensitive information or causes patient discomfort. Usefulness was defined as the usefulness of systematic incorporation of the standard measure in all EHRs, which includes broad applicability and utility in the clinical setting. Committee members individually assigned from one to three stars to each measure for each criterion, where three was best. Following committee discussion, a consensus judgment was reached on these ratings, and a rating of the overall committee judgment of the priority of including the measure in all EHRs was determined. The small table for each measure summarizes these ratings.

In the sections that follow, the committee describes examples of possible measures for each of the candidate domains listed in Chapter 3 and repeated below. As in Chapter 3, the domains are not listed in order of priority, but follow the committee’s initial classification and outline of the domains into five levels (sociodemographic, psychological, behavioral, individual-level social relationships and living conditions, and neighborhoods and communities) in Chapter 2.

Sociodemographic Domain Measures

Sexual orientation

Race and ethnicity

Country of origin/U.S. born or non-U.S. born

Education

Employment

Financial resource strain: Food and housing insecurity

Psychological Domain Measures

Health literacy

Stress

Negative mood and affect: Depression and anxiety

Psychological assets: Conscientiousness, patient engagement/activation, optimism, and self-efficacy

Behavioral Domain Measures

Dietary patterns

Physical activity

Tobacco use and exposure

Alcohol use

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

Individual-Level Social Relationships and Living Conditions Domain Measures

Social connections and social isolation

Exposure to violence

Neighborhoods and Communities Domain Measures

Neighborhood and community compositional characteristics

SOCIODEMOGRAPHIC DOMAIN MEASURES

Candidate sociodemographic domains include domains described in the Kaplan and colleagues’ (2000) epidemiological model and the Ansari and colleagues’ (2003) public health model for describing the interactions, pathways, and causalities between social and behavioral determinants of health and health outcomes, such as individual factors (e.g., country of origin, sexual orientation, education, and employment), social factors (e.g., race, ethnicity, sexual orientation), and living conditions (e.g., financial resource strain). In the Ansari model, most of these domains would fall under socioeconomic determinants. The committee notes, however, that some domains may span across levels. For example, race, ethnicity, and sexual orientation are individual factors that are affected by social and cultural factors.

Sexual Orientation

Sexual orientation is characterized by two separate but related subdomains: self-identification and choice of partner for having sex. Self-identification—how people define their sexual orientation—can differ from their actual behavior. For example, an individual may define her- or himself as heterosexual (or gay or bisexual), yet not actually have sex with anyone. Another individual might define her- or himself as heterosexual yet have sex with others of the same gender.

Identification and Description of Measures

Two standard questions with standard responses have been used in multiple surveys to measure sexual orientation: one question for self-identification and another for sexual behavior. For example, the California Health Interview Survey (CHIS) has asked California residents about their sexual orientation by asking two questions (CHIS, no date). The question for sexual behavior asks the respondents about the gender of their sexual partner(s):

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

In the past 12 months, have your sexual partners been male, female, or both male and female?

The question for self-identification is:

Do you think of yourself as straight or heterosexual, as gay (lesbian) or homosexual, or bisexual?

During the CHIS interview, interviewers are prompted to further explain to the respondent what is meant by the self-identification question by saying (CHIS, no date):

Straight or heterosexual people have sex with, or are primarily attracted to people of the opposite sex, Gay (and Lesbian) people have sex with or are primarily attractive to people of the same sex, and Bisexuals have sex with or are attracted to people of both sexes.

The committee believes that more detailed questions about sexual practices are better asked as part of the clinical interview.

Common Metric

There is not a commonly accepted metric for these measures at this time.

Ratings of Measures by Committee

The two questions listed in the above section are both publicly available with no licensing restrictions for use. Knowledge of a person’s self-identified sexual orientation and sexual behavior can be useful for diagnosing and treating conditions that may be related to sexual orientation—for example, African American gay, bisexual, and men who have sex with men represented an estimated 72 percent (10,600) of new HIV infections among all African American men and 36 percent of an estimated 29,800 new HIV infections among all gay and bisexual men (CDC, 2014b). For most conditions, however, knowing this would not change the clinical approach. Because of this, the committee judged that it is not highly useful to systematically include these measures in all electronic health records.

Both questions could be feasibly asked without putting much burden on the patient or clinical workflow. The CHIS reports that the refusals to answer these questions are no greater than that of other questions on the survey (CHIS, 2014). However, asking questions about sexual orientation can be highly sensitive, including for people who think that their sexual

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

orientation or sexual behaviors are private matters and should not be asked unless they are directly relevant for their current medical care. Based on these considerations, the committee rated the measure as follows in Table 4-1.

Limitations of Measures

Some individuals may not want to answer questions related to sexual orientation because they feel it is not relevant to their medical care. The questions do not enable a clinician to determine what potential screening a patient needs for sexually transmitted diseases without asking additional questions concerning specific sexual behaviors.

Specific Populations

Adolescence can be a particularly challenging time for teens who think they may be gay or lesbian. Gay and lesbian teens have a higher rate of mental distress and suicidality than their “straight” peers. Their sexual preferences are still being formed and, no matter what their self-identification, they may have not had sex with anyone, or they may have only had sex with persons of the other sex. They may not want their parents or their friends to know about their sexual feelings or activities. For teenagers, the committee believes questions about sexual orientation should be asked starting at 13 years of age, which is the age at which pediatricians start to separate children from their parents during the examination. The questions should be asked by the clinician in the examination or consultation room, not on a paper form to be handed to a registration clerk.

Other Measures Reviewed

There are other commonly used measures (to whom a person is attracted regardless of how they identify themselves or who they have sex with), but these questions appear to be less critical for medical care. A panel, Sexual Minority Assessment Research Team (SMART), was formed by the Ford Foundation to determine the “best scientific approaches to gathering data on sexual orientation” (SMART, 2009). SMART recommends three questions: one for self-identification, one for sexual behavior, and one for sexual attraction. (See the SMART, 2009, report for more information.) The self-identification and sexual behavior questions are the same as the ones identified by the committee. The committee viewed the sexual attraction question to be less clinically relevant to health than the other two questions. Fenway Health, a group practice in Boston with a large population of lesbian, gay, bisexual, and transgender individuals has begun measuring

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

TABLE 4-1 Ratings of the Measures on Sexual Orientation

Domain Measure Standard Measure and Freely Available (*** = standard, * = no standard) Usefulness (*** = most useful, * = least useful) Feasible (*** = most feasible, * = least feasible) Lack of Sensitive Information or Patient Discomfort (*** = least sensitive, * = most sensitive) Committee Judgment of the Measure (*** = highest rating, * = lowest rating)
Self-Identification *** * *** ** **
Sexual Behavior *** ** *** * **
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

sexual orientation and gender identity using a paper registration form. A study conducted by the Fenway Institute on four community health centers found acceptability and feasibility of asking three questions—one on sexual orientation and two on gender identity—as part of patient registration (Fenway Institute and Center for American Progress, 2013). The question on sexual orientation is similar to the question reviewed by the committee on sexual identity. The other two questions ask about gender identity, which is a domain the committee considered and reviewed but did not select as a candidate domain.

Race and Ethnicity

Measures of race and ethnicity are commonly included in EHRs; however, the method of ascertainment (i.e., patient self-report versus clinical staff determination based solely on patient appearance) and the metrics (i.e., the specific racial categories available for selection) vary considerably.

Identification and Description of Measures

The Office of Management and Budget (OMB) defines the standards for the classification of federal data on race and ethnicity (OMH, 2010). In 1997, the OMB announced revisions to the Standards for the Classification of Federal Data on Race and Ethnicity. The current standards have five categories for race (American Indian or Alaska Native, Asian, Black or African American, Native Hawaiian or Other Pacific Islander, and White) and two categories for ethnicity (Hispanic or Latino and Not Hispanic or Latino). When race and ethnicity are collected separately, the number of White and Black persons who are Hispanic must be identifiable, and capable of being reported in that category. If a combined format is used to collect racial and ethnic data, the minimum acceptable categories are American Indian or Alaskan Native; Asian or Pacific Islander; Black, not of Hispanic origin; Hispanic; and White (OMB, 2003). Currently, Meaningful Use Stage 2 is using OMB’s categories of race and ethnicity and is in the process of reviewing added categories similar to the U.S. Census.

The U.S. Census has been collecting information on race, in some form, since the late 1700s (U.S. Census Bureau, 2014b). Throughout the decades, many changes and adaptations resulted in the adding of race and ethnic categories. As of 2010, the U.S. Census uses one unique category for collecting information on ethnicity, which is whether the person is of Hispanic, Latino, or Spanish origin. If the answer is yes, then four options are given for describing this background, including an option to write in information (PRB, 2009). Multiple ethnic origins may be selected. For race, the U.S. Census lists 15 possible options. This question also includes the option to

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

write in a tribe for persons of American Indian or Alaska Native race, and allows written specification of Other Asian and Other Pacific Islander. An option to write in some other race is also provided, as is the option to select multiple races. The following are the two questions on race and ethnicity (PRB, 2009)1:

Question 5: Is the person of Hispanic, Latino, or Spanish origin?

images No, not Hispanic, Latino, or Spanish origin

images Yes, Mexican, Mexican American, Chicano

images Yes, Puerto Rican

images Yes, Cuban

images Yes, another Hispanic, Latino, or Spanish origin (with fill in option)

Question 6: What is the person’s race? Mark one or more races to indicate what this person considers himself/herself to be.

images White

images Black, African American, or Negro

images American Indian or Alaskan Native (with fill in option)

images Asian Indian

images Chinese

images Filipino

images Japanese

images Korean

images Vietnamese

images Native Hawaiian

images Guamanian or Chamorro

images Samoan

images Other Pacific Islander (with fill in option)

images Other Asian (with fill in option)

images Some other race (with fill in option)

Common Metric

The U.S. Census 2010 questions 5 and 6 are detailed metrics of race and ethnicity used at the federal level for assessing the demographic composition of the United States.

Ratings of Measures by Committee

The OMB racial and ethnic group measures provide a minimum set of categories, while the U.S. Census items provide a more comprehensive

_________________

1 The 2010 Census questionnaire was mailed to every household in the United States with directions indicating that the person who filled out the form would be identified as Person 1. Person 1 was also asked to answer questions about every household member, including identifying race and ethnicity (PRB, 2009). These two questions will need to be adapted accordingly for patient self-reporting for use in EHRs.

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

and specific description of race and ethnicity. The U.S. Census race and ethnicity questions will be useful for health care providers to determine which patients to screen for certain conditions (based on the epidemiology of those conditions across race and ethnic groups) and will help identify populations for which cultural competency training may be warranted for clinical staff. Additionally, the U.S. Census questions allow health systems to track patient outcomes across multiple racial and ethnic origins and are easily comparable to national data on the same groups. These measures are highly feasible to collect, and should be self-reported rather than determined by clinical staff. Because of these considerations the committee rated the measures of the race/ethnicity domain as follows in Table 4-2.

Limitations of Measure

Limitations of the U.S. Census questions include the time and financial costs to operationalize these detailed metrics (i.e., entering write-in categories into the EHR).

Specific Populations

Children and adolescents may find self-reporting their racial and ethnic identity to be challenging. The committee recommends that race/ethnicity be ascertained from parents of children and adolescents up to age 18 years. Older adolescents may be able to validly report their race/ethnic identity; however, further qualitative research is needed to identify the specific age at which youth can validly self-report this personal characteristic.

Country of Origin/U.S. Born or Non-U.S. Born

The United States is composed mostly of immigrants, some more recent and others from prior generations. As a health-related domain, immigration is a complex concept with several subdomains—including country of origin, years since immigration, immigration status, acculturation, primary language, preferred language for health care encounter, literacy, race, and ethnicity—all of which in varying degrees impact health. Being native or foreign-born and degree of acculturation have implications for health. First-generation immigrants tend to have better health outcomes than acculturated and U.S. born second or later generational individuals (Singh and Miller, 2004). For example, immigrant Latinos are less likely to be depressed and anxious, have lower cancer and cardiovascular rates, and have better infant birth weight than do U.S. born Latinos (Franzini et al., 2001; Lara et al., 2005).

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

TABLE 4-2 Ratings of the Measures on Race and Ethnicity

Domain Measure Standard Measure and Freely Available) (*** = standard, * = no standard) Usefulness (*** = most useful, * = least useful) Feasible (*** = most feasible, * = least feasible) Lack of Sensitive Information or Patient Discomfort (*** = least sensitive, * = most sensitive) Committee Judgment of the Measure (*** = highest rating, * = lowest rating)
U.S. Census: Race/Ethnicity *** *** *** *** ***
OMB Definition *** ** *** *** ***
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

Identification and Description of Measure

Country of origin and length of time in the United States are relatively straightforward variables to collect as exemplified by questions included in the U.S. Census. Acculturation is a more complex concept for which there is no universally accepted “gold standard.” However, researchers frequently use proxy measures for acculturation that involve single items or composite variables based on questions relating to the following: country of origin, time spent in the United States, generation in the United States (via country of origin of parents/caregivers), and language used in medical appointments and/or most frequently used in the home. Language is thought to be the strongest single predictor of acculturation (Alegria, 2009; Arcia et al., 2001; Marin and Gamba, 1996). Most of the literature base for acculturation measures is derived from studies among Latinos migrating to the United States.

In addition to country of origin, there are additional measures from the U.S. Census Bureau’s long-form questions that can be used. (Below are the questions if read orally):

11a. Does this person speak a language other than English at Home?

images   Yes

images   No → skip to question 12

11b. What is this language? (Fill in information)

11c. How well does this person speak English?

images   Very well

images   Well

images   Not well

images   Not at all

12. Where was this person born?

images   The United States. (Fill in state)

images   Outside of the United States. (Fill in name of country)

13. Is this person a citizen of the United States?

images   Yes, born in the United States → SKIP to question 15a

images   Yes, born in Puerto Rico, Guam, the U.S. Virgin Islands, or Northern Marianas

images   Yes, born abroad of U.S. citizen parent or parents

images   Yes, U.S. citizen by naturalization

14. When did this person come to live in the United States? (Fill in year)

The committee suggests using only question 12 (Where was this person born?) and question 14 (When did this person come to live in the United States?). For question 12, the standard response option is to list the state, including the District of Columbia. For those born in U.S. territories, such as Puerto Rico or Guam, individuals are instructed to respond “outside of the United States.” For question 14, the response option is the year an indi-

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

vidual came to live in the United States for the purpose of measuring period of entry, not the total years lived in the United States (Malone et al., 2003).

Common Metric

There is no commonly accepted metric for all of the concepts embedded here at this time.

Ratings of Measure by Committee

The two questions suggested by the committee are freely available and are highly feasible in a clinical setting. The brevity of the questions makes them easy to respond to and can be nonburdensome to the patient and on administrative personnel. Results from these questions can alert a health care provider to ask about the patient’s preferred language and can potentially result in effective culturally and linguistically appropriate treatment. There may be sensitive issues in asking a patient’s country of origin, which might inhibit accurate reporting and adversely affect patient–provider communication and trust. Because of these considerations, the committee rated the measure as follows in Table 4-3.

Limitations of Measure

As stated earlier, the clinical care team should be sensitive to individuals whose immigration status is questionable. Individuals whose immigration status is questionable may feel particularly vulnerable and may opt to not seek care or not follow up with their treatment if they feel threatened.

Specific Populations

A specific population for this domain measure is infants, children, and adolescent patients. For this population one can ask information on the parents’ country of origin. Because younger-age populations’ health outcomes are associated with their parents’ social and economic backgrounds. Knowing parental country of origin/nativity may assist in improving the clinical outcome of these populations. Concerning immigration status, an infant, child, or adolescent of immigrant or refugee parents should be given special attention because questions are not asked of parents.

Education

Education is a well-established determinant of health at all stages of the life span. Educational attainment assesses the human capital dimension

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

TABLE 4-3 Ratings of the Measure on Country of Origin/U.S. Born Versus Non-U.S. Born

Domain Measure Standard Measure and Freely Available (*** = most, * = least) Usefulness (*** = most useful, * = least useful) Feasible (*** = most feasible, * = least feasible) Lack of Sensitive Information or Patient Discomfort (*** = least sensitive, * = most sensitive) Committee Judgment of the Measure (*** = most, * = least)
U.S. Census: Country of Origin *** ** *** * **
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

of socioeconomic status (SES). It is strongly associated with other indices of SES, including occupational prestige and household income. Internationally, and to some extent among disadvantaged groups in the United States, increasing female education has been associated with considerable improvements in health and well-being.

Identification and Description of Measure

The committee reviewed two measures of education: one for highest degree earned, and another for highest year of schooling completed. Each has been shown to relate to a wide range of health outcomes. Although the two measures are highly related, some research has found a linear association of years of schooling and health while others have found discontinuities associated with earning degrees. The committee evaluated the education measures originally developed by the U.S. Census and expanded by the MacArthur Research Network on SES & Health (MacArthur Research Network on SES & Health, 2008).

For highest level of school that an individual completed, he or she is asked:

What is the highest level of school you have completed? Check one.

Elementary School High School College Graduate/Professional School
01___ 09___ 13___ 17___
02___ 10___ 14___ 18___
03___ 11___ 15___ 19___
04___ 12___ 16___ 20+___
05___      
06___      
07___      
08___      

For highest degree earned by the individual, he or she is asked:

What is the highest degree you earned? Check one.

images   High school diploma

images   GED

images   Vocational certificate (post high school or GED)

images   Association degree (junior college)

images   Bachelor’s degree

images   Master’s degree

images   Doctorate

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

Common Metric

There is not a commonly accepted metric for these measures at this time.

Ratings of Measure by Committee

There is no fee for use of these measures, and their clarity and brevity make them highly feasible to be asked in a clinical setting or self-reported before the clinical encounter. These two measures will be useful for population health management. Education levels correlate with many health indicators, particularly as a health determinant and in their links to SES and health literacy, making it useful to capture them on a clinical level (Commission to Build a Healthier America, 2009; Woolf and Braveman, 2011; Woolf et al., 2007). The committee does not believe either question is sensitive. Because of these considerations, the committee rated the measure as follows in Table 4-4.

Limitations of Measure

A potential limitation of the two measures arises from the fact that neither captures the quality of the education received.

Specific Populations

Educational attainment can be problematic to measure for young adults whose education is not fixed. For those pursuing education later in life, a change to the frequency of capturing the information (originally a one-time capture) would be warranted. For most individuals, educational attainment plateaus by age 25. For children and adolescents, the most appropriate measure of education is their parent’s educational attainment, rather than their grade level, because the intent of the concept is to characterize one aspect of an individual’s SES. The agreement between adolescents and parents regarding parental educational attainment is moderately strong, with older adolescents reporting higher levels of agreement (Ensminger et al., 2000). For this reason, the committee recommends obtaining parental education directly from parents. However, adolescents 14 years and older may be able to provide reasonably valid assessments of their parents’ educational attainment (Ensminger et al., 2000).

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

TABLE 4-4 Ratings of the Measure for Education

Domain Measure Standard Measure and Freely Available (*** = standard, * = no standard) Usefulness (*** = most useful, * = least useful) Feasible (*** = most feasible, * = least feasible) Lack of Sensitive Information or Patient Discomfort (*** = least sensitive, * = most sensitive) Committee Judgment of the Measure (*** = highest rating, * = lowest rating)
Educational Attainment *** *** *** *** ***
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

Other Measures Reviewed

The committee believes that the measure of education should be as standardized as possible and, to this end, has recommended two common measures: one for highest grade level and one for highest degree obtained. There are several alternative questions that can be used to provide assessments of these metrics. The committee suggests that the MacArthur network questions be used because of the ease of administration and simplicity.

Employment

A large body of work has linked employment status, the type of occupation a person is engaged in, and specific job characteristics (including physical and psychosocial characteristics) to a broad range of health outcomes. There are many ways in which an individual’s work exposures can be characterized. The simplest involves classifying persons based on their employment status. A second more complex option requires obtaining more information on the type of job so employed persons can be further classified in terms of the type of occupation or the occupational category they belong to, based on standard classifications (such as the U.S. Census). Yet a third option is to obtain measures of specific physical exposures (e.g., chemicals, noise, dust) or psychosocial exposures (e.g., demands, control, support) at work. All of these work dimensions have been shown to be strongly predictive of health and could have clinical and population utility.

Identification and Description of Measure

A person’s employment status reflects their level of engagement with the workforce and has relevance to the clinician. It is also useful for population monitoring of employment trends. Although apparently straightforward, the questions used to characterize employment status in national surveys are often complex and involve a relatively large set of questions.

Standard measures used in national surveys such as the National Health and Nutrition Examination Survey (NHANES) are useful but were judged by the committee to be too long and complex to be included in the EHR. The committee evaluated the Multi-Ethnic Study of Atherosclerosis (MESA), a simple measure used in many population studies. These categories allow for a simple classification of the patient’s current employment. The question and categories are as follows (MESA, 2005):

Choose one of the following which best describes your current occupation:

images   Homemaker, not working outside the home

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

images   Employed (or self-employed) full time

images   Employed (or self-employed) part time

images   Employed, but on leave for health reasons

images   Employed, but temporarily away from my job (other than health reasons)

images   Unemployed or laid off 6 months or less

images   Unemployed or laid off more than 6 months

images   Retired from my usual occupation and not working

images   Retired from my usual occupation but working for pay

images   Retired from my usual occupation but volunteering

Common Metric

There is not a commonly accepted metric for this measure at this time.

Ratings of Measure by Committee

The measure is freely available and the committee did not find the measures to be sensitive for patients. The committee found this question to be somewhat useful for systemic inclusion in EHRs. Due to its brevity, it is feasible to complete in the clinical setting. The question is straightforward and is not seen as being highly sensitive. Because of these considerations, the committee rated the measure as follows in Table 4-5.

Limitations of Measure

One of the limitations for the MESA measure is that it does not ask if the person is unemployed due to a disability. This is something that can be considered and addressed by possibly adding it as an option for selection. Additionally, the measure does not capture employment history or military service.

Specific Populations

For children and adolescents, parental employment status should be obtained.

Other Measures Reviewed

The committee gave serious consideration to the inclusion of other work and occupational measures, including characterization of type of occupation and other physical and psychosocial exposures at work for those employed. Although the committee recognized the value of identifying

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

TABLE 4-5 Ratings of the Measure for Employment

Domain Measure Standard Measure and Freely Available (*** = standard, * = no standard) Usefulness (*** = most useful, * = least useful) Feasible (*** = most feasible, * = least feasible) Lack of Sensitive Information or Patient Discomfort (*** = least sensitive, * = most sensitive) Committee Judgment of the Measure (*** = highest rating, * = lowest rating)
MESA Employment Question ** ** *** ** **
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

occupational categories, the process required to collect and then adequately code these data was judged too burdensome to be recommended for inclusion at this time. However, as discussed in Chapter 7, as the tools and instruments for characterizing occupation develop, this may become a high-priority measure for future inclusion in the EHR, given the high relevance of occupation and occupational exposure to the health of employed adults. For example, the National Institute for Occupational Safety and Health (NIOSH) is currently developing and standardizing specific measures that capture a patient’s industry and occupation, including measures on work schedule, employment status, and external causes related to injury and poisoning (i.e., ICD-10 codes) (NIOSH, 2014). Although this information can be useful in identifying work exposures and conditions that are linked to health outcomes, the committee concluded that coding this information in an EHR was time intensive.

Financial Resource Strain: Food and Housing Insecurity

Financial resource strain is a composite of both subjective evaluation of economic difficulties and specific sources of strain, such as food insecurity or housing insecurity (Kahn and Pearlin, 2006). Food insecurity occurs when the availability of food is limited or uncertain (Scott and Wehler, 1998). It has been of interest not only as a reflection of overall economic strain but also because of its potential role in eating patterns that contribute to being overweight or obese (Dinour et al., 2007; Seligman et al., 2007). Individuals who experience periods of insufficient food availability may overconsume calories when food becomes available (Alaimo et al., 2001; Polivy, 1996; Taren et al., 1990; Townsend et al., 2001). In addition to obesity, food insecurity has been associated with physical health, mental health, and nutrition (Siefert et al., 2001; Szanton et al., 2010). Financial resource strain and insecurity (e.g., food, housing) are interconnected with one another along with other variables, often making it challenging to measure forms of insecurity independently from one another (Kahn and Pearlin, 2006; Siefert et al., 2001; Szanton et al., 2010). The phrases “food insufficiency” and “food insecurity” appear in the literature and are sometimes used interchangeably. Housing insecurity can range from an individual situation to community settings and is hard to measure because it is confounded with other variables (Kushel et al., 2006) (e.g., underemployment/unemployment, low wages, housing costs, and lack of access to the Supplemental Nutrition Assistance Program [SNAP]).

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

Identification and Description of Measures

The committee considered collection of patient income as a domain; however, the measures are complex, and short measures do not take into consideration issues such as wealth and assets, and that family income needs to be adjusted for the number of people dependent on the income. Additionally, individuals may not be comfortable disclosing this information, especially on an annual basis. Patient income overlaps with financial resource strain and geocoded median neighborhood income, which were ranked higher by the committee.

Overall financial resource strain For overall financial resource strain, the committee considered the work of Kahn and Pearlin (2006), a study of aging, stress, and the health consequences of repeated financial resource strain. They offer two approaches for the assessment of financial resource strain: one addressing current strain, and one addressing financial strain throughout the life span. The research demonstrated that both current financial strain and the number of periods of financial strain across the life span affect health outcomes. An alternative approach was employed in the Study of Women’s Health Across the Nation (SWAN) and the Coronary Artery Risk Development in Young Adults (CARDIA) studies (see, for example, Hall et al. [2009] and Puterman et al. [2013]) that uses a single-item question. These studies indicate the single-item question to be a valid measure of general financial resource strain.

How hard is it for you to pay for the very basics like food, housing, medical care, and heating? Would you say it is…

Very hard

Somewhat hard

Not hard at all

Patients are asked to circle one of the options. The answer is then scored on a scale of 1 (very hard) to 3 (not at all), and unknown answers are scored as a negative number. Assessments were made at study entry and during the study at years 2, 5, 7, 10, 15, 20, and 25 (CARDIA, no date). Evidence from the CARDIA study demonstrates the value of measuring the difficulty of paying for basics over time (e.g., financial resource strain), because there appear to be cumulative effects (e.g., incident hypertension). The effects are independent of other SES measures (Matthews et al., 2002).

As stated earlier, financial resource strain has various composites, food insecurity and housing insecurity being two of those components. Food insufficiency is defined as food intake that is inadequate because of lack of

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

resources (Briefel and Woteki, 1992). It is a valid measure for this domain, it has good evidence of validation (Lee and Frongillo, 2001), and it was seen to be actionable in a clinical setting.

Food insufficiency The food insufficiency measure has been used in NHANES III to measure an individual’s food intake based on the reported adequacy of the family’s food resources (Alaimo et al., 1998). The first question of the five-item set has been used to define restricted household food supplies or too little food intake among adults or children in the house hold. The question, when delivered orally, is (Siefert et al., 2001):

Which of the following describes the amount of food your household has to eat:

images   Enough to eat

images   Sometimes not enough to eat

images   Often not enough to eat

This single question has shown to hold external and face validity for the measurement of food insufficiency (Alaimo et al., 1998; Basiotis, 1992; Briefel and Woteki, 1992; Christofar and Basiotis, 1992).

Housing insecurity The financial resource strain subdomain that is the most difficult to measure is housing insecurity. From the review of the literature, there does not appear to be a measure that looks at housing insecurity by itself. Many of the studies look at housing instability, food insecurity, and economic instability combined (Kushel et al., 2006; Wallace et al., 2013). For example, in a study by Kushel et al. (2006), low-income individuals first self-reported if they had difficulty in paying rent, mortgage, or utility bills in the past year. Positive respondents were then asked “whether or not they had to move in with friends or family because they had no other choice.”

Common Metric

There is no common metric for financial resource strain.

Ratings of Measures by Committee

The single question for overall financial resource strain is accepted, freely available, and shows a strong association with current health status. The single question used to measure food insufficiency is a standard measure that is also freely available and can be useful in identifying individuals and their household members who are having difficulty in accessing the

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

appropriate amount of food on a regular basis. However, patients may feel uncomfortable answering that question. The committee noted that the single-question measure of financial resource strain includes food insecurity. There is no standard measure for housing insecurity. Because of these considerations, the committee rated the measures as follows in Table 4-6.

Limitations of Measures

As previously stated, financial resource strain and insecurity, such as food or housing, are interconnected and are therefore difficult to measure independent of one another.

Specific Populations

The food insufficiency questions should be asked of parents to characterize a child’s family’s food insecurity. Low-income individuals are a vulnerable population and should be asked these questions on a regular basis.

Other Measures Reviewed

In addition to the CARDIA and NHANES questions for evaluating financial resource strain, two other measures the committee reviewed were the BEST Index and the Elder Index developed by Wider Opportunities for Women (WOW, 2014). Both of these indices measure the income needed to achieve food, housing, and income security, as well as other expenses. However, these indices are population-based and are not made for use as an individual measure. Food insecurity can also be measured using a two-item instrument developed from affirmative responses given to the U.S. Department of Agriculture 18-item Household Food Security Survey (Hager et al., 2010; USDA, 2014). These two questions distinguish between hunger and food insecurity; however, this measure’s limitation is that it has only been validated for families with children.

PSYCHOLOGICAL DOMAIN MEASURES

The candidate psychological domains covered below are described in the Ansari et al. (2003) model as psychosocial risk factors, whereas in the Kaplan et al. (2000) model the domains fall within the category of individual risk factors.

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

TABLE 4-6 Ratings for the Measures on Financial Resource Strain

Domain Measures Standard Measure and Freely Available (*** = standard, * = no standard) Usefulness (*** = most useful, * = least useful) Feasible (*** = most feasible, * = least feasible) Lack of Sensitive Information or Patient Discomfort (*** = least sensitive, * = most sensitive) Committee Judgment of the Measure (*** = highest rating, * = lowest rating)
Overall Financial Resource Strain *** *** *** ** ***
Food Insufficiency *** ** ** ** ***
Housing Insecurity * ** * ** *
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

Health Literacy

Health literacy, as stated in Chapter 3, is “the degree to which individuals have the capacity to obtain, process, and understand basic health-related decisions” (IOM, 2004, p. 20). Limited health literacy is common in the United States: 35 percent have basic or below basic health literacy; 53 percent have intermediate health literacy; and only 12 percent of adults are classified as proficient (Almader-Douglas, 2013; Kutner et al., 2006 ). Further, limited health literacy is associated with poor health outcomes (IOM, 2009), including higher hospitalization rates, greater use of emergency rooms as a source of regular care, and more adverse disease outcomes, as well as poor adherence to medications and a limited knowledge of health conditions (Baker et al., 1998; Berkman et al., 2004; Schillinger et al., 2002). Limited health literacy may place individuals at risk for poor health for several reasons, including creating difficulties in navigation through a convoluted health care system, in patient–provider interactions, and in self-care (HHS, 2008; Paasche-Orlow and Wolf, 2007). Health literacy is a complex domain with various dimensions (e.g., education, preferred language, culture, vision/hearing/cognitive ability). It is exacerbated by the complexity of health information and use of scientific medical terminology by the clinical health care team that may be unfamiliar to patients.

Identification and Description of Measure

The most widely used measures in the literature relating limited health literacy to adverse health outcomes are the Test of Functional Health Literacy in Adults (TOFHLA) and the Rapid Estimate of Adult Literacy in Medicine (REALM) (IOM, 2009). However, these scales are lengthy. The TOHFLA requires approximately 22 minutes to complete (NC Program on Health Literacy, 2014) and the REALM is a 66-item test (AHRQ, 2009). They also have complicated scoring procedures, are time-consuming to administer, and may require additional training of staff in order to administer them effectively.

A three-question measure has been derived to assess health literacy. These questions are as follows (Chew et al., 2004):

1.   How often do you have someone (like a family member, friend, hospital/clinic worker or caregiver) help you read hospital materials?

Always Often Sometimes Occasionally Never

2.   How often do you have problems learning about your medical condition because of difficulty understanding written information?

All of the time Most of the time Some of the time

A little of the time None of the time

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

3.   How confident are you filling out forms by yourself?

All of the time Most of the time Some of the time

A little of the time None of the time

Responses are scored on a five-point Likert scale from zero to four. Chew et al. (2004) and Wallace et al. (2006) found that the three questions identified individuals with inadequate health literacy and compared favorably on receiver operator curve test characteristics with the Short TOFHLA area under the curve.

Common Metric

There is no common metric for this domain.

Rating of Measure by Committee

The three-question measure validated by Chew et al. (2004, 2008) may be useful for the clinical team in identifying those patients who have less than adequate or marginal health literacy. The brevity of the questions and the simplicity of the scoring make it feasible and nonburdensome for the patients to answer. Health literacy is a potentially sensitive area for patients, although framing this question as a query about the need for assistance might diminish potential stigmatization (compared with actual literacy tests).

However, although health literacy can be viewed as a characteristic of the individual, it operates in the context of the health care system—the clarity of communications from the health care system and individual providers’ communication skills—as well as the patient’s health literacy. As a result, adverse health effects of low health literacy can be reduced not just by identifying the needs and capacity of the individual patient but also by assuring the clarity of communication with all patients no matter what their literacy level.

All patients deserve clear communication, not just those deemed to have low health literacy levels. Thus, many have suggested that health care providers adopt a “universal precautions” approach to health literacy (Brown et al., 2004; Oates and Paasche-Orlow, 2009; Paasche-Orlow and Wolf, 2007; Rudd, 2010; Volandes and Paasche-Orlow, 2007). This approach offers strategies for clear communication using plain language for clear communication with all patients. An example is the use of teach-back techniques (IOM, 2004) that ask a patient to describe to a member of the clinical care team her or his understanding of their treatment plan. Such techniques can be useful in determining whether all patients comprehend

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

and retain the information being provided to them. The Agency for Healthcare Research and Quality has developed a Health Literacy Universal Precautions Toolkit that details these approaches (AHRQ, 2014).

The committee concluded that health literacy can be best addressed through a universal precautions approach to ensure clear and effective communication with all patients, rather than measurement of the level of health literacy in individual patients. EHRs may have a role in assessing and documenting patient comprehension, although this potential is beyond the scope of this committee’s review. Because of these considerations, the committee rated the measure as follows in Table 4-7.

Specific Populations

Although the REALM has been adapted for adolescents (Davis et al., 2006), brief measures of health literacy, such as the three questions that the committee evaluated, have not been adapted for pediatric populations. It may be more appropriate to assess parental health literacy for young children; parental levels of health literacy have been shown to relate to children’s receipt of health services and their health outcomes (Sanders et al., 2009). As with adult literacy, a systems-focused approach may be more appropriate than an individual approach.

Other Measures Reviewed

The committee considered other scales, specifically, REALM-66, REALM-Short Form (SF), Short Assessment of Health Literacy for Spanish-speaking Adults (SAHLASA)-50, TOFHLA, Spanish (S)TOFHLA, and Newest Vital Signs (NVS). The majority of these scales have internal reliability, and their results correlated with at least one other scale. The REALM (and its abbreviated versions) and the TOFHLA (and its abbreviated versions) have been most studied for their correlation with health outcomes. However, these scales are not feasible to ask in a clinical setting.

Stress

Stress is a subjective state arising when an individual believes that he or she does not possess the resources to cope with a threatening situation, resulting in tension, restlessness, nervousness, or anxiousness. Acute and chronic stresses are types of stress experienced by individuals that have been linked to health outcomes. Acute stress is episodic and manifests during times of increased demands or pressures in response to anticipated threats (APA, 2014). Acute stress is short term and can have transient health

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

TABLE 4-7 Ratings of the Measure on Health Literacy

Domain Measure Standard Measure and Freely Available (*** = standard, * = no standard) Usefulness (*** = most useful, * = least useful) Feasible (*** = most feasible, * = least feasible) Lack of Sensitive Information or Patient Discomfort (*** = least sensitive, * = most sensitive) Committee Judgment of the Measure (*** = highest rating, * = lowest rating)
Three-Question Measure from Chew et al. (2004) *** * *** ** *
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

effects, such as emotional distress and muscular problems. More serious health effects may emerge if such exposures are severe or persist over time. The chronic—sometimes termed toxic—stress of experiencing situations over an extended time that are perceived to be unmanageable and uncontrollable can create an “allostatic load” that increases the likelihood of serious health consequences (e.g., high blood pressure, cardiovascular disease, depression) (APA, 2014; Seeman et al., 2001). Adversities experienced in childhood may engender toxic stress that not only affects the child’s health and well-being but may increase disease risk in adulthood as well (Felitti et al., 1998; Shonkoff et al., 2012).

Identification and Description of Measures

The committee considered two different approaches to stress measurement. Given the evidence of the importance of adverse early life exposures and links to adversity and stress, the committee examined the Adverse Childhood Experiences (ACE) tool. The ACE assesses chronic stress associated with experiencing multiple adversities in childhood (up to age 18). For example, individuals who had six or more ACEs were more likely to have a premature death than were those without ACEs, dying 20 years earlier on average (60.6 years, 95 percent confidence intervals [CIs] = 56.2, 65.4, versus 79.1 years, 95 percent CI = 78.4, 79.9, respectively) (Brown et al., 2009). Additionally, in another study, results indicate an increased graded-dose response between ACE scores and comorbid outcomes of substance abuse, impaired memory, sexuality (early intercourse, promiscuity, or sexual dissatisfaction), aggression, and somatic disturbances (Anda et al., 2005). The original ACE index developed by Felitti et al. (1998) asked adults 17 questions regarding exposures such as abuse and neglect, parental marital status, mental illness, and incarceration. ACE researchers and the CDC developed a standardized ACE module for use in the Behavioral Risk Factor Surveillance System (BRFSS). The following are the adapted BRFSS ACE’s questions (Institute for Safe Families, 2013):

While you were growing up, that is during your first 18 years of life, how often, if ever, did a parent, step-parent, or another adult living in your home…

1.   How often did a parent or adult in your home ever swear at you, insult you, or put you down?

More than once Once Never

2.   How often did your parents or an adult in your home ever hit, beat, kick or physically hurt you in any way? Do not include spanking.

More than once Once Never

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

3.   How often did anyone at least 5 years older than you or an adult ever touch you sexually?

More than once Once Never

4.   How often did anyone at least 5 years older than you or an adult ever try to make you touch them sexually?

More than once Once Never

5.   How often did anyone at least 5 years older than you or an adult ever force you to have sex?

More than once Once Never

6.   How often did your parents or adults in your home ever slap, hit, kick, punch, or beat each other up?

More than once Once Never

7.   Did you live with anyone who was a problem drinker or alcoholic?

8.   Did you live with anyone who used illegal street drugs or who abused prescription medications?

9.   Did you live with anyone who was depressed, mentally ill, or suicidal?

10. Were your parents separated or divorced?

11. Did you live with anyone who served time or was sentenced to serve time in prison, jail, or other correctional facility?

Scores on the ACE obtained in adulthood are associated with various poor health outcomes (e.g., impaired memory, substance abuse, somatic disturbances), as stated in the above text.

To obtain a measure of current stress, the committee evaluated a single question developed by Elo et al. (2003). This question is associated with indicators of health and psychosocial work characteristics, and it can be used for monitoring stress in work-life contexts. The single-item question is:

Stress means a situation in which a person feels tense, restless, nervous, or anxious, or is unable to sleep at night because his/her mind is troubled all the time. Do you feel this kind of stress these days?

Not at all A little bit Somewhat Quite a bit Very much

The response is recorded on a five-point Likert scale ranging from 1—indicating not at all, 2—a little bit, 3—somewhat, 4—quite a bit, to 5—indicating very much. This single question shows content validity as well as concurrent criterion validity. The single question converged with items on psychological symptoms and sleep disturbances and with validated measures of well-being (Elo et al., 2003).

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

Common Metric

No common metric is available for either chronic or acute stress.

Ratings of Measures by Committee

The ACE measure is increasingly used in research, and evidence is accumulating that adverse childhood experiences set a trajectory for poor health into adulthood. It is of moderate length and asks very sensitive information about the patient’s background exposures. Its usefulness is primarily to understand the patient’s background and factors that may help the clinician understand the patient’s current health and health behaviors. It is deemed of modest priority for inclusion in the EHR. ACE questions can be asked only once in an adult’s life span.

The single-question measure by Elo et al. (2003) is freely available and has been assessed by the committee to be highly feasible for inclusion in the EHR and it is moderately sensitive in nature, potentially causing patient discomfort. The measure can also be linked to the PROMIS Emotional Distress (Depression and Anxiety) Short Form scales, which also are reliable and are valid ways to assess stress.

Because of these considerations, the committee rated the measures as follows in Table 4-8.

Limitations of Measures

ACE is a retrospective assessment of exposures to profound family stressors. Validation of the scale derives from studies of associated risk for chronic disease and early mortality. It is not clear what action can be taken should patients have high scores, independent of standard treatment for the potential health consequences of these family stressors that the patients experience.

The single-question (Elo et al., 2003) stress measure is scored on a five-point Likert scale that has been primarily tested in Scandinavian populations. There is no clinical cutoff for determining when interventions, such as referral to stress management, are warranted.

Specific Populations

The adverse childhood experiences measure can be used for adult populations; comparable versions for pediatric populations have yet to be validated for either the ACEs or the Elo et al. (2003) stress measure.

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

TABLE 4-8 Ratings of the Measures for Stress

Domain Measures Standard Measure and Freely Available (*** = standard, * = no standard) Usefulness (*** = most useful, * = least useful) Feasible (*** = most feasible, * = least feasible) Lack of Sensitive Information or Patient Discomfort (*** = least sensitive, * = most sensitive) Committee Judgment of the Measure (*** = highest rating, * = lowest rating)
Adverse Childhood Experiences (ACE) *** * ** * *
 
Single Question from Elo et al. (2003) *** *** *** ** ***
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

Other Measures Reviewed

The committee reviewed the British-developed “distress thermometer,” which is a visual analog scale to rate emotional distress ranging from 0 (no distress) to 10 (extreme distress) (Roth et al., 1998). However, this tool’s limitation is that it also measures depression and anxiety and does not specifically screen for stress (Mitchell, 2007), making a diagnosis more challenging. In addition, the committee considered the National Institutes of Health (NIH) Toolbox Perceived Stress Survey, but due to its length (10 questions), it was not seen as feasible for the clinical encounter (Slotkin et al., 2012).

Depression

There are many well-validated measures of depressive symptoms. For the EHR, the committee reviewed several screening and monitoring measures. The committee deliberated on one measure that screens the patient for depression symptoms and, if positive, would lead to further referrals for clinical evaluation, and on a second measure for monitoring symptom changes over time.

Identification and Description of Measures

For an initial patient screen for depression the committee considered the Patient Health Questionnaire (PHQ) for depression, which is a commonly used screener in diverse clinical settings. The PHQ-9 contains nine symptom items rated on a four-point scale that measures the frequency of experiencing depressive symptoms during the past 2 weeks, from “not at all” (0 points) to “nearly every day” (3 points). The PHQ-9 is a reliable instrument, with Cronbach’s alpha coefficients ranging from 0.86 to 0.89, and test-retest reliability across 48 hours with an alpha coefficient of 0.84 (Kroenke et al., 2001). This scale and the shortened version described below are associated with other measures of negative emotions, quality of life, number of office visits, and disability measures, all of which support the validity of the scale.

Two questions taken from the PHQ-9 have been validated for use as a screen for depression. The briefer PHQ-2, with a cutoff of greater than or equal to three, has a sensitivity of 83 percent and a specificity of 92 percent for major depression, relative to independent structured interviews by mental health professionals (Kroenke et al., 2003). The PHQ-2 question is the following (Kroenke et al., 2003) and is scored from 0 (not at all) to 3 (nearly every day):

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

Over the past 2 weeks, how often have you been bothered by any of the following problems:

  1. Little interest or pleasure in doing things
    Not at all Several days More than half the days Nearly every day
  2. Feeling down, depressed or hopeless
    Not at all Several days More than half the days Nearly every day

For monitoring patients with a positive initial depression screen, the committee considered the Patient Reported Outcomes Measurement Information System (PROMIS) Depression Scale, which measures mood in the last week. It is available in short forms (eight-item, six-item, and four-item) as well as in a computerized adaptive test (CAT) version. The PROMIS-8b short form has also been shown to assess depressive symptoms as well as other longer scales (e.g., PHQ-9, Center for Epidemiology Studies Depression Scale [CESD]-10) (Kim et al., 2012), having an item correlation with CESD-10 of 0.83 to 0.88 (Amtmann et al., 2014; Choi et al., 2014). PROMIS depression instruments are available in adult and pediatric versions and are related to other measures of negative emotion, especially anxiety. The scale is a valid instrument for monitoring patients who initially screen positive for depressive symptoms or are depressed. The PROMIS-8b for depression asks how often in the past 7 days—from 1 (never) to 5 (always)—a person had each of eight feelings:

In the past 7 days:

  1. I felt worthless
  2. I felt that I had nothing to look forward to
  3. I felt helpless
  4. I felt sad
  5. I felt like a failure
  6. I felt depressed
  7. I felt unhappy
  8. I felt hopeless

Common Metric

Both the PHQ-2 and the PROMIS-8b short form can be scored on the PROMIS Depression T-scale, which serves as the common metric for depressive symptoms. The PROMIS Depression T-score has a mean of 50 (centered on the U.S. population average) and a standard deviation of 10 (PROMIS, 2014b). Developers of PROMIS are conducting research that allows clinicians and investigators to translate the scores from other depres-

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

sion measures, specifically the Beck Depression Inventory, and the PHQ-9, into PROMIS T-scores.

The PHQ-9 is based on the DSM-IV criteria for depression and, as such, maps onto the diagnostic criteria for depression. Thus, the clinical cutoff of yes or no risk for depression is a common diagnostic metric that can be approximated by the PHQ-2.

Ratings of Measure by Committee

The PHQ-2 will be useful for health care providers to identify those patients at risk for depression. Depression is comorbid with many chronic illnesses and, when treated, may lead to improvement in health more generally. The brevity of the scale and its simplicity of scoring make it highly feasible. It is a somewhat sensitive measure because of the stigma associated with depression, and it should be followed by fuller evaluation and support services. The committee believes the PHQ-2 will be useful in EHR because of the impact of depression on many illnesses and disabilities and because effective treatments are available. It is likely to be especially useful to include during known periods in the life span of increased vulnerability to depression, such as postpartum women, perimenopausal women, and the elderly.

The PROMIS-8b short form is helpful in monitoring change in symptoms over time. Like the PHQ-2, the information is somewhat sensitive because of stigma associated with depression. An advantage of PROMIS is that studies have been conducted to allow cross-scale comparisons.

Because of these considerations, the committee rated the measures as follows in Table 4-9.

Limitations of Measure

The primary limitation of the PHQ-2 is a concern about the management challenges that the health system may face in following up on what may be a substantial number of patients who have a positive depression screen. However, given the impact of depression on many aspects of life, including its contribution to other disease states, this is a legitimate and an important demand on the health care system.

Specific Populations

There are life periods when special care is required for diagnosis and treatment for pediatric samples, for women, and for the elderly. One advan-

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

TABLE 4-9 Ratings of the Measures for Depression

Domain Measures Standard Measure and Freely Available (*** = standard, * = no standard) Usefulness (*** = most useful, * = least useful) Feasible (*** = most feasible, * = least feasible) Lack of Sensitive Information or Patient Discomfort (*** = least sensitive, * = most sensitive) Committee Judgment of the Measure (*** = highest rating, * = lowest rating)
PHQ-2 *** *** *** ** ***
PROMIS-8b *** * ** ** *
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

tage of the PROMIS Depression Scale is that there is a pediatric version that is conceptually harmonized with the adult version. For detection of postpartum depression and depression among the elderly, either the PHQ-2 or the PROMIS-8b depression measures would be a reasonable choice.

Depending on the level of disability, individuals with intellectual disability may present with atypical symptoms of mood disorders (e.g., depression, anxiety), and they may have limited speech capabilities (Hurley, 2006). Individuals with mild levels of intellectual disabilities can use self-reported measures (e.g., PHQ-9, PROMIS-8b) for diagnosing depressive symptoms or depression. However, more research is needed for developing adequate measures for individuals with severe intellectual disabilities (Hermans and Evenhuis, 2010).

Other Measures Reviewed

The committee considered other widely used and validated scales, specifically, the CES-D scale and the WHO-K6 or K10 scales. The latter scales cover serious mental illness more broadly, whereas the CES-D is designed for use in epidemiological studies. The CES-D has clinical cutoffs and can be used to monitor changes in depressive symptoms during treatment. However, it contains 20 items, which makes it less feasible within the constraints of the EHR.

Anxiety

Like depression scales, anxiety measures are plentiful. For the EHR, the committee deliberated on several measures, including the Generalized Anxiety Disorder Scale (GAD-7) and the PROMIS Anxiety Scales.

Identification and Description of Measures

The GAD-7 contains seven items based on clinical diagnostic criteria for generalized anxiety disorder. The questions concern anxious feelings, worrying, and trouble relaxing in the past 2 weeks. It has excellent reliability, with Cronbach’s alpha coefficient equal to 0.92, a test-retest reliability of 0.83, and a sensitivity and specificity of 89 percent and 82 percent, respectively (Spitzer et al., 2006). It is correlated with other anxiety scales and diagnosis of generalized anxiety disorder. It is related to measures of quality of life, disability symptoms, and illness visits. The seven questions are as follows and each of the items is scored from 0 (not at all) to 3 (nearly every day) (Spitzer et al., 2006):

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

Over the past 2 weeks, how often have you been bothered by the following problems?

  1. Feeling nervous, anxious, or on edge
  2. Not being able to stop or control worrying
  3. Worrying too much about different things
  4. Having trouble relaxing
  5. Being so restless that it is hard to sit still
  6. Becoming easily annoyed or irritable
  7. Feeling afraid as if something awful might happen

Like the PROMIS depression instruments, the PROMIS anxiety instruments are available in short versions (eight-item, seven-item, six-item, and four-item). All the short PROMIS anxiety instruments are similar in reliability and precision for screening anxiety symptoms. The PROMIS anxiety instruments are available in adult, pediatric, and parent proxy versions (PROMIS, 2014a). This measure asks about anxiety levels in the past week for symptoms, including fear, anxiousness, misery, and hyper-arousal symptoms. The PROMIS-7a has an internal consistency reliability score of 0.90 (PROMIS, 2014a) and is highly correlated with the Mood and Anxiety Symptoms Questionnaire (MASQ) (Pilkonis et al., 2011). This instrument asks the following questions about how often in the past week the person experienced each of the following feelings, from never (0) to always (5):

In the past 7 days…

  1. I felt fearful
  2. I found it hard to focus on anything other than my anxiety
  3. My worries overwhelmed me
  4. I felt uneasy
  5. I felt nervous
  6. I felt like I needed help for my anxiety
  7. I felt anxious

Common Metric

Like other PROMIS scales, PROMIS anxiety scores are T-scores with a standard deviation of 10 developed based on large community and clinical samples. The PROMIS Anxiety T-score has a mean of 50 (centered on the U.S. population average) and a standard deviation of 10 (PROMIS, 2014a). Ongoing investigations are providing ways to link measures provided by other anxiety scales (e.g., MASQ, GAD-7, and the Positive and Negative Affect Scale [PANAS]) to the PROMIS norms (Schalet et al., 2013).

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

Ratings of Measures by Committee

Because anxiety often accompanies depressive symptoms, the committee downgraded the usefulness of a separate measure of anxiety within the EHR. Like measures of depression, it is feasible to collect, especially if the PROMIS CAT measure is used. Because of these considerations, the committee rated the measures as follows in Table 4-10.

Limitations of Measure

The primary limitation concerns whether the health system can manage adequate follow-up of patients likely to report high levels of anxiety symptoms.

Specific Populations

As stated earlier, the PROMIS anxiety instruments include both adult and pediatric versions. The pediatric version is self-reported by children as young as 8 years of age, and using parent-proxy, it is reported for children of 5 to 7 years of age. There is no PROMIS measure for children younger than 5 years of age.

Other Measures Reviewed

The GAD-2 short form instrument, derived from GAD-7, had a sensitivity of 86 percent for generalized anxiety disorder, 76 percent for panic disorder, 70 percent for social anxiety disorder, along with reliable specificity (83 percent to 81 percent) for these disorders (Kroenke et al., 2007). The GAD-2 questions can be used in combination with the PHQ-2 to screen for anxiety and depression.

Conscientiousness

Conscientiousness is a complex trait composed of propensity to be self-controlled, to be task and goal directed, to delay gratification, and to follow norms and rules. It is challenging to measure the full array of factors that constitute conscientiousness.

Identification and Description of Measure

Conscientiousness is measured typically as part of personality scales that assess five major personality characteristics, sometimes called the “Big

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

TABLE 4-10 Ratings of the Measures for Anxiety

Domain Measure Standard Measure and Freely Available (*** = standard, * = no standard) Usefulness (*** = most useful, * = least useful) Feasible (*** = most feasible, * = least feasible) Lack of Sensitive Information or Patient Discomfort (*** = least sensitive, * = most sensitive) Committee Judgment of the Measure (*** = highest rating, * = lowest rating)
GAD-7 *** ** *** ** **
PROMIS-7a *** ** *** ** **
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

Five.” The scales measuring the Big Five have several variants, with up to 105 items. There have been efforts to develop briefer versions of the Big Five personality characteristics (see Bogg and Roberts [2004] for an overview). A commonly used one for research purposes is the Big Five Inventory (BFI), which has nine items measuring conscientiousness; these items are rated on a five-point scale, ranging from 1 (disagree strongly) to 5 (agree strongly) and summed after appropriate reverse scoring of specific items (Rammstedt and John, 2007). Even shorter versions are available and validated. However, in general, they are not as reliable and valid as the longer scales. One of the shorter scales is the BFI-10 items, which contains two items that comprise the conscientious subscale (Rammstedt and John, 2007):

I see myself as someone who:

images   Tends to be lazy

images   Does a thorough job

Common Metric

There is no current common metric for conscientiousness.

Ratings of Measure by Committee

Use of the measure of conscientiousness, based on the Neuroticism-Extraversion-Openness (NEO) Personality Inventory Revised, was determined to have some limitations because it is copyrighted and lengthy. The BFI (44 questions, with 9 for conscientiousness) is available for researchers, and the owner of the copyright may grant permission to use these measures to EHR vendors in the interest of patient care and research.2

Although the evidence is strong that conscientiousness is consistently related to longevity and positive health behaviors, including adherence, the committee deemed it low priority for inclusion in EHRs because of a lack of evidence on how to intervene on patient conscientiousness. It is theoretically possible to develop programs to aid patients in developing organizational skills, which is one part of conscientiousness, but that is only one aspect, and no evidence exists on the effectiveness of such training. Because of these considerations, the committee rated the measure as follows in Table 4-11.

_________________

2 Personal communication, Oliver John, University of California, Berkeley, August 5, 2014.

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

TABLE 4-11 Ratings of the Measure for Conscientiousness

Domain Measure Standard Measure and Freely Available (*** = standard, * = no standard) Usefulness (*** = most useful, * = least useful) Feasible (*** = most feasible, * = least feasible) Lack of Sensitive Information or Patient Discomfort (*** = least sensitive, * = most sensitive) Committee Judgment of the Measure (*** = highest rating, * = lowest rating)
Big Five Inventory-2 of 10 Questions ** ** ** *** *
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

Limitations of Measure

A potential limitation of using the BFI or its subscales is that the items are not unambiguously freely available to all users (John, 2007–2009). Permission is routinely granted for free use for research and noncommercial use. The copyright holder for the BFI has indicated his willingness to give permission for use in EHRs because these will be used to advance treatment and research.

Specific Populations

A pediatric version of the BFI, long form, that is completed by parents has been developed (John, 2007–2009).

Other Measures Reviewed

There is a two-item conscientiousness scale from the 10-item Personality Inventory. Each item is rated on a 7-point scale ranging from 1 (disagree strongly) to 7 (agree strongly) with the stem, “I see myself as” followed by the two items: (a) dependable, self-disciplined, and (b) disorganized, careless. The latter is reverse scored. This scale is freely available, feasible, low burden, and not sensitive. Note that this is a separate measure than that noted above. However, there is little research using the shorter measures in relation to health and there is only a moderate correlation between the two-item measures and longer measures (John, 2007–2009).

Patient Engagement/Activation

Patient engagement/activation encompasses an individual’s attitudes, skills, and knowledge that enable him or her to engage in health care in an active, full, and meaningful manner.

Identification and Description of Measure

Currently, only one measure has been validated for the assessment of this psychological asset—the Patient Activation Measure (PAM). A PAM score has the ability to predict health care outcomes (e.g., medication adherence, hospitalizations) (Inigma, 2014). The PAM instrument began as a 22-item questionnaire that measured unidimensional, self-management variables using a Guttman-like measure (Hibbard et al., 2004), and it was then shortened to a 13-item measure. The shorter version has a small loss in precision within some subgroups. Because the PAM instrument is not freely available, the items are not listed here.

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

Common Metric

There is no common metric for patient engagement/activation at this time.

Ratings of Measure by Committee

The PAM is not freely available to health care providers or health systems because of copyright protections. Recent research reports an association between patient engagement/activation and health outcomes (Brenk-Franz et al., 2013; Hibbard and Greene, 2013), but perhaps because of limitations on its use, there is limited research on which to base estimates of usefulness. Because of these considerations, the committee rated the measure as follows in Table 4-12.

Limitations of Measure

The PAM is not freely available to health care providers due to copyright protections.

Other Measures Reviewed

Measures related to the PAM have been developed for patients with specific diseases, but no validated measures of patient activation that would be appropriate for the diverse types of patients served by EHRs have been published.

Optimism

Identification and Description of Measure

The prevailing measure of optimism is the Life Orientation Test-Revised (LOT-R), which contains six questions regarding expectations for positive and negative outcomes. Each question is rated on a four-point scale. Test-retest reliability is 0.79 across 28 months, and Cronbach’s alpha coefficient is 0.78 for the six questions, demonstrating acceptable internal validity (Scheier et al., 1994). It correlates with other positive assets, such as self-esteem and self-mastery, and negatively with negative attributes, such as anxiety and neuroticism. The LOT-R items are the following (Scheier et al., 1994)3:

_________________

3 The original LOT-R has four filler questions; they are omitted here.

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

TABLE 4-12 Ratings of the Measure for Patient Engagement/Activation

Domain Measure Standard Measure and Freely Available (*** = standard, * = no standard) Usefulness (*** = most useful, * = least useful) Feasible (*** = most feasible, * = least feasible) Lack of Sensitive Information or Patient Discomfort (*** = least sensitive, * = most sensitive) Committee Judgment of the Measure (*** = most, * = least)
PAM * * ** *** *
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

Be as honest as you can throughout, and try not to let your responses to one question influence your responses to other questions. There are no right or wrong answers.

  1. In uncertain times, I usually expect the best.
  2. If something can go wrong for me, it will.
  3. I’m always optimistic about my future.
  4. I hardly ever expect things to go my way.
  5. I rarely count on good things happening to me.
  6. Overall, I expect more good things to happen to me than bad.

The questions are scored on a five-point Likert scale, from 0 (strongly disagree) to 4 (strongly agree). Given that most samples score on average to be somewhat optimistic, it may be useful to examine the three positive expectation and three negative expectation subscales separately.

Common Metric

There is not a commonly accepted metric for optimism at this time.

Ratings of Measure by Committee

This scale is rated as low in patient burden, high in ease of administration, and unlikely to cause stigmatization. The measure was judged to be moderately useful in the context of delivery of clinical care and identifying individuals who might require additional supports. This domain was the most highly rated of the positive assets because of the strength and consistency of association of the evidence linking optimism with adherence to behavior change, positive health behaviors, and mortality. Reservations about the measure’s actionability led to the lower overall ratings on the measure’s usefulness. Because of these considerations, the committee rated the measure as follows in Table 4-13.

Limitations of Measure

The LOT-R is a research instrument and is not intended for clinical applications. There is no clinical cutoff for optimism (University of Miami Department of Psychology, 2007).

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

TABLE 4-13 Ratings of the Measure for Optimism

Domain Measure Standard Measure and Freely Available (*** = standard, * = no standard) Usefulness (*** = most useful, * = least useful) Feasible (*** = most feasible, * = least feasible) Lack of Sensitive Information or Patient Discomfort (*** = least sensitive, * = most sensitive) Committee Judgment of the Measure (*** = highest rating, * = lowest rating)
Life Orientation Test-Revised *** ** *** *** **
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

Specific Populations

The LOT has been adapted for youth self-report (YLOT, between third and sixth grade) (Ey et al., 2005) and for parent-proxy report for younger children (PLOT) (Lemola et al., 2010).

Self-Efficacy

Self-efficacy is defined as the confidence to carry out or produce a specific behavior or make a change in a specific behavior (e.g., exercise three times per week for 60 minutes). It is defined within the context of social learning theory, which has proved useful to guide interventions requiring behavior change. Another perspective is generalized self-efficacy, which is the confidence that one can deal with demands and stressful circumstances more broadly.

Identification and Description of Measures

One way of measuring self-efficacy for specific behaviors is by using Bandura’s Guide for Constructing Self-Efficacy Scales (Bandura, 2006). Construction of these scales requires analysis of the specific domain, including knowledge of the behavior involved and assessment of aspects of behavior control, and identifications of challenges that may derail a person’s success. According to Bandura (2006, p. 310), “Behavior is better predicted by people’s beliefs in their capabilities to do whatever is needed to succeed.” Efficacy scales are unipolar, ranging from minimum strength (0) to maximum strength (100), with intervals of 10 to form the ratio scales. The Alcohol Abstinence Self-Efficacy Scale (DiClemente et al., 1994), for example, is a 20-item scale developed to assess a person’s self-efficacy regarding alcohol abstinence. Specific scales have been developed and validated for other behaviors (e.g., smoking cessation [Etter et al., 2000], nutrition-related or dietary patterns [Anderson et al., 2000], physical exercise [Schwarzer and Renner, 2000]).

Generalized self-efficacy can be assessed by using NIH’s self-efficacy measure, the NIH Toolbox Self-Efficacy Survey. This survey measures an individual’s generalized confidence to handle stressful circumstances. It contains 10 questions rated according to how true the statement is of the person on a five-point scale. It has versions for adult and pediatric samples. The 10-item questions are as follows (NIH, 2006–2012):

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

Please read the sentence and describe how true it is of you in general.

  1. I can manage to solve difficult problems if I try hard enough.
  2. If someone opposes me, I can find the means and ways to get what I want.
  3. It is easy for me to stick to my aims and accomplish my goals.
  4. I am confident that I could deal efficiently with unexpected events.
  5. Thanks to my talents and skills, I know how to handle unexpected situations.
  6. I can solve most problems if I try hard enough.
  7. I stay calm when facing difficulties because I can handle them.
  8. When I have a problem, I can find several ways to solve it.
  9. If I am in trouble, I can think of a solution.
  10. I can handle whatever comes my way.

The questions are scored on a five-point scale ranging from 1 (never) to 5 (very often).

Common Metric

There is no common metric for self-efficacy at this time. The NIH Toolbox measure of generalized self-efficacy has normative data that could be used to develop a common metric for self-efficacy among adults.

Ratings of Measures by Committee

There is no standard measure of self-efficacy for specific behaviors. Although any measure that may be constructed is typically short, easy to ascertain, and unlikely to be sensitive, a dictionary of behaviors would need to be developed to be useful for patient care. The NIH Toolbox is widely used in research and clinical settings and is standardized. It can be easily collected although it is somewhat long. Its usefulness is limited, however, because it is not specific to the situations involved in health care. Because of these considerations, the committee rated the measures as follows in Table 4-14.

Limitations of Measure

The NIH Toolbox measure was recently developed. Further research is needed to better understand its performance and utility in clinical settings.

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

TABLE 4-14 Ratings of the Measures for Self Efficacy

Domain Measures Standard Measure and Freely Available (*** = standard, * = no standard) Usefulness (*** = most useful, * = least useful) Feasible (*** = most feasible, * = least feasible) Lack of Sensitive Information or Patient Discomfort (*** = least sensitive, * = most sensitive) Committee Judgment of the Measure (*** = highest rating, * = lowest rating)
Self-Efficacy Scales for Specific Behaviors * * * *** *
NIH Toolbox of Generalized Self-Efficacy *** * *** *** *
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

Specific Populations

There are no pediatric measures of self-efficacy at this time.

BEHAVIORAL DOMAIN MEASURES

The candidate behavioral domains include dietary patterns, physical activity, tobacco use and exposure, and alcohol use.

Dietary Patterns

Dietary patterns (summary measures of food consumption) are often examined in the research setting to capture their associations with health, and they are increasingly used in the clinical setting to guide lifestyle counseling. Nutrition is important to health but complex to measure because it encompasses caloric intake, the macronutrients found in food (e.g., fats, proteins, carbohydrates), the micronutrients (e.g., vitamins, minerals), and non-nutritional ingredients (e.g., sugar).

After examining possible measures of nutrition, the committee focused on the subdomain of fruit and vegetable intake and frequency because of the availability of appropriate measures for this aspect of nutrition. Although caloric intake, sugar intake, and energy expenditure are also important components of nutrition, short validated measures assessing them do not exist at this time.

Identification and Description of Measure

Consumption of fruits and vegetables is highly correlated with health outcomes. For the EHR, the committee evaluated the two-question measure developed and used in a British study by Wardle et al. (2000) and validated by Cappuccio et al. (2003) for fruit and vegetable intake with biomarkers. These same measures have been used and validated in various food assessment studies across various British populations. (See, for example, Baker and Wardle, 2003; Little et al., 1999; and Wardle et al., 2005.) These two questions are:

  1. How many pieces of fruit, of any sort, do you eat on a typical day?
  2. How many portions of vegetables, excluding potatoes, do you eat on a typical day?

These questions have high specificity, identifying more than 80 percent of individuals with biomarker profiles indicative of low fruit and vegetable

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

intake (i.e., consuming less than five portions of fruits and vegetables per day), which is the case for the majority of the U.S. population. These studies used the British Dietetic Association’s guidelines (BDA, 2014) to define what it meant by portion, which is comparable to the U.S. Department of Agriculture’s Recommended Guidelines for Americans (USDA, 2013). It is important to note, however, that portion size corresponds to the amount of a single food item an individual eats in one sitting, which is different from serving size, a standardized unit of a measured single food (CDC, 2006a).

Common Metric

There is no common metric for dietary patterns at this time.

Ratings of Measure by Committee

The two-question measure used by Cappuccio et al. (2003) was assessed by the committee to be highly feasible and with few concerns about the measure containing sensitive information and thus causing patient discomfort.

Additionally, it was thought to represent a useful screening tool as a marker of a healthy or unhealthy diet. However, while the committee considers collection of data on nutrition to be a priority, this measure to capture data on dietary patterns is not as robust as other measures the committee reviewed; it was also viewed to be only moderately useful due to limitations of the measure (see below) and because the clinical intervention is unclear.

Because of these considerations, the committee rated the measure as follows in Table 4-15.

Limitations of Measure

This measure captures only one aspect of nutrition. Other limitations of the measure include a lack of clarity regarding the term portion and insufficient information on the measure’s stability and the needed frequency of screening.

Weight is routinely collected by the clinical care team, and the committee considered that addressing issues related to a patient being overweight or obese would trigger behavioral counseling interventions related to healthy diet and weight even without a measure of dietary patterns. Typical health risk assessments most likely include fruit and vegetable consumption as well as other risks such as those associated with processed meats.

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

TABLE 4-15 Ratings of the Measure for Dietary Patterns

Domain Measure Standard Measure and Freely Available (*** = standard, * = no standard) Usefulness (*** = most useful, * = least useful) Feasible (*** = most feasible, * = least feasible) Lack of Sensitive Information or Patient Discomfort (*** = least sensitive, * = most sensitive) Committee Judgment of the Measure (*** = highest rating, * = lowest rating)
Fruit and Vegetable Consumption *** ** *** *** **
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

Specific Populations

Typical and recommended dietary patterns of children vary from adults, as do rewards for certain dietary patterns. For children and adolescents, consideration should be given to adding questions on intake of sweetened beverages, such as soda and fruit drinks. However, it is noted that adults also suffer from obesity linked with consumption of sweetened beverages. Further, because sugar is included in so many foods, it was unclear to the committee if measuring only sugar-sweetened beverages would be sufficient. This is an emerging area of science.

Other Measures Reviewed

Measures of specific dietary patterns (e.g., the eight-item Starting the Conversation screen, the Dietary Approaches to Stop Hypertension diet, and the Mediterranean diet) were reviewed, but they are potentially more time-consuming for patients to respond to; some measures are not clinically freely available; and if high-risk patients are identified, behavioral interventions need to be easily referable from the clinical setting. The Block Food Questionnaire available from NutritionQuest is a strong measure of sugar consumption, but it has more than 125 questions and is a proprietary tool.

Physical Activity

Physical activity refers to skeletal muscle movement resulting in energy expenditure that exceeds resting levels (Caspersen et al., 1985). Physical activity can be done purposefully, or as part of daily life, work, school, or fun; it can be a modifiable determinant of health.

Identification and Description of Measure

Physical activity can be assessed using direct ascertainment provided by devices that detect movement or via self-reported questionnaires. Methods of direct ascertainment include devices such as accelerometers, pedometers, and heart rate monitors. These devices must be worn by the patient as a wristwatch or a waistband or attached to clothing. They assess the amount an individual moves (i.e., motion) or the impact of movement on physiology, such as the change in heart rate associated with motion. Physical activity questionnaires ask respondents to report recent participation in movement behaviors, typically over a short interval or as a diary of types and duration of specific types of physical activities. Self-report assessment

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

methods may require retrospective recall or involve prospective data gathering of data for 1 to 7 days.

In both cases, a measure of metabolic equivalent of task minutes (MET-minutes) can be obtained. Based on the stipulation that health benefits are achieved with 500 to 1,000 MET-minutes per week, the Physical Activity Guidelines Advisory Committee, formed by the Office of Disease Prevention and Health Promotion within the U.S. Department of Health and Human Services (HHS), recommended 150–300 minutes per week of moderate-intensity or 75 minutes per week of vigorous-intensity activity to provide substantial health benefit (Physical Activity Guidelines Advisory Committee, 2008).

The committee evaluated the measurement properties of two specific measures: the Exercise Vital Sign and accelerometry. Both can be used to produce MET-minutes per week. The Exercise Vital Sign is a modified version of the physical activity questions in the BRFFS. It is a two-question measure that does not have licensing fees. Feasibility studies have shown that it can be readily integrated into the EHR (Coleman, 2012). The measure assesses minutes per week spent in moderate to vigorous exercise. The questions are:

  1. On average, how many days per week do you engage in moderate to strenuous exercise (like walking fast, running, jogging, dancing, swimming, biking, or other activities that cause a light or heavy sweat)?
  2. On average, how many minutes do you engage in exercise at this level?

The first question has a categorical response option set (0–7 days), and the second question is recorded in blocks of 10 minutes, from 0–150 or greater. The two numbers are multiplied to display minutes per week of moderate-vigorous activity, which can also be converted into the three-category clinically useful variable: inactive, insufficiently active, or sufficiently active.

Compared with accelerometers, the Exercise Vital Sign is markedly easier to administer (more feasible) and is more practical for clinical settings. The two-question Exercise Vital Sign has adequate reliability for screening the physical activity level of a population (Coleman, 2012). In addition, a study shows that the Exercise Vital Sign measure has both good face and discriminant validity when used in EHRs (Coleman, 2012).

Accelerometers are sensors that detect motion and provide an objective measure of MET-minutes. They can be worn as a wristwatch and used to assess physical activity behavior over several days, and they can provide real-time feedback to users or clinicians. Although the use of these sensors

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

in research studies is commonplace, there is a lack of standards for placement of the sensors, sampling frequency, and defining a valid “day.” There are numerous devices available, including accelerometers embedded within smartphones. The validity of accelerometer assessment using smartphones requires more investigation; however, given their widespread availability, this research is likely to be done soon. Patient compliance can be a problem, and the cost of accelerometers presents another barrier to routine use in clinical settings.

Common Metric

Clinically relevant groupings of activity behaviors have been developed using METs as a measure of physical activity intensity. These include: light intensity defined as 1.1–2.9 METs; moderate intensity, defined as 3.0–5.9 METs, such as brisk walking or gardening; and vigorous intensity, defined as 6.0 METs or more, such as running or fast cycling (Physical Activity Guidelines Advisory Committee, 2008). The Centers for Disease Control and Prevention divides activity behaviors into low intensity (e.g., walking), moderate intensity (e.g., brisk walking), and high intensity (e.g., jogging) based on these MET classifications.

One approach for converting the Exercise Vital Sign measure into MET-minutes per week is to multiply the number of minutes spent in moderate-to-vigorous activity by 4.5 METs, which is the midpoint MET level for moderate activity. This computation provides a crude approximation of MET-minutes per week; however, it will underestimate values for individuals who spend more time in vigorous than in moderate activity.

Much scientific evidence linking physical activity with health benefits evaluates the number of MET-minutes per week in association with an outcome, such as rates of disease, biomarkers, or fitness levels. MET-minutes are the product of the MET level, which is based on the type of activity being performed and the duration of the behavior. Use of MET-minutes as a common metric allows different types of aerobic activities with different intensity levels to be related on a single scale.

Ratings of Measure by Committee

The Exercise Vital Sign is standard and freely available. Compared with accelerometers, the Exercise Vital Sign is markedly easier to administer (more feasible) and is more practical for clinical settings. Because of these considerations, the committee rated the measures as follows in Table 4-16.

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

TABLE 4-16 Ratings of the Measure for Physical Activity

Domain Measures Standard Measure and Freely Available (*** = standard, * = no standard) Usefulness (*** = most useful, * = least useful) Feasible (*** = most feasible, * = least feasible) Lack of Sensitive Information or Patient Discomfort (*** = least sensitive, * = most sensitive) Committee Judgment of the Measure (*** = highest rating, * = lowest rating)
Exercise Vital Sign *** *** *** *** ***
Accelerometer *** ** * *** *
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

Limitations of Measures

Sedentary behavior was not reviewed as the measures and tools are not yet well developed.

Specific Populations

Accelerometry can be used among children and adolescents, and in research contexts, it has been used with infants. The Exercise Vital Sign has not been validated for pediatric populations, which is a topic for future research. Valid and reliable measures for geriatric populations exist to measure physical activity in older adults, but they were not prioritized by the committee owing to their lengthy forms. The health care team will need training on how to use these measures with people with disabilities or high-need patients (NIH, 2011).

Other Measures Reviewed

The International Physical Activity Questionnaire has a short form with nine questions. The committee reviewed this and found the measure to be acceptable, but it is more time consuming than the two-question measure.

Tobacco Use and Exposure

Tobacco use and tobacco-related illnesses are the leading cause of morbidity and mortality in the United States. Evidence strongly suggests that a health care provider’s explicit interest in a patient’s tobacco use can assist the patient taking steps toward stopping tobacco use. Tobacco use is a major cause of excess mortality among cancer-related deaths and is also a cause of heart disease, stroke, and chronic obstructive pulmonary disease (CDC, 2014c). The U.S. Preventive Services Task Force (USPSTF) recommends (A grade) that clinical care providers ask about tobacco use and provide tobacco cessation interventions for those who use tobacco products. Nicotine addiction has been studied intensely for more than 50 years. Nicotine contained in tobacco leads to dependence in many people. Based on the most recent 2012 National Health Interview Survey (NHIS) (CDC, 2014a), 18 percent of U.S. adults (18 years and older) are current cigarette smokers, while 21 percent were former smokers (Blackwell et al., 2014). The large majority of current smokers meet criteria for nicotine dependence.

Identifying tobacco-using persons is the first step to treatment. Practice guidelines from HHS and the American Psychiatric Association suggest asking patients at each visit whether they use tobacco. The tobacco use

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

measure for Stage 2 Meaningful Use requires that more than 80 percent of all unique patients 13 years old or older seen by the eligible provider have smoking status recorded as structured data (CMS, 2012) with one of the following Systematized Nomenclature of Medicine (SNOMED) codes:

  • Current every day smoker
  • Current some day smoker
  • Former smoker
  • Never smoker
  • Smoker, current status unknown
  • Unknown if ever smoked
  • Heavy tobacco smoker
  • Light tobacco smoker

The ways in which tobacco counseling and treatment are handled by EHRs remains to be settled but was considered out of scope for this committee.

Identification and Description of Measure

Smoking status questions, lifetime and current, have been asked on the NHIS for almost half a century and are used to assess progress toward achieving the Healthy People 2020 objectives. Although the questions have slightly changed throughout the years, the basis of measuring the prevalence of lifetime and current smoking status remains. The NHIS includes the following categories to measure if the patient is a current or every day smoker, former some day smoker, former smoker, or never smoker (Adsit and Fiore, 2013; ASPE, no date–a):

  1. Have you smoked at least 100 cigarettes in your entire life?
    Yes No Refused Do not know, and if yes:
  2. Do you NOW smoke cigarettes every day, some days or not at all?
    Every day Some days Not at all Refused Do not know

A “current every day smoker” or “current some day smoker” has smoked at least 100 cigarettes and still regularly smokes every day or periodically, yet consistently. A “former smoker” has smoked at least 100 cigarettes but does not currently smoke. A “never smoker” has not smoked 100 cigarettes. “Smoker, current status unknown” is known to have smoked at least 100 cigarettes, but whether they currently still smoke is unknown (Adsit and Fiore, 2013).

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

Common Metric

There is no common metric for tobacco use or exposure at this time.

Ratings of Measure by Committee

The smoking status measure is standard and freely available. Collecting this information in a systematic way in EHR is useful because smoking has been linked to many preventable diseases and is a major factor influencing life expectancy. The committee identified an additional measurement instrument that provides options to measure degree of dependence on nicotine and did not find the questions to be as sensitive in nature as other measures of dependency and thus considers them to be feasible to ask. Because of these considerations, the committee rated the measure as follows in Table 4-17.

Limitations of Measure

Occasional or intermittent smokers may be missed with these screening questions. The measure also is limited in only asking questions about cigarette use and does not ask about tobacco exposure (e.g., if patient lives with someone who smokes indoors). Also, the NHIS is used for ages 18 and above.

Specific Populations

Adolescents can be asked just one question because the first NIHS question may not identify those who have recently taken up smoking. A more appropriate question is taken from the Youth Risk Behavior Survey (ASPE, no date–b; Kann et al., 2014):

On how many of the past 30 days did you smoke a cigarette?
None 1–30 days Refused Do not know

For adolescents, cigarette smoking is defined as smoking cigarettes on at least 1 day during the 30 days before the survey. Pregnant women are a vulnerable population, but the committee concluded that the age-appropriate screening question(s) should be used as listed above.

Other Measures Reviewed

Other measures assess degree of dependence on nicotine. All have acceptably high levels of validity and reliability and have been shown to be

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

TABLE 4-17 Ratings of the Measure for Tobacco Use and Exposure

Domain Measure Standard Measure and Freely Available (*** = standard, * = no standard) Usefulness (*** = most useful, * = least useful) Feasible (*** = most feasible, * = least feasible) Lack of Sensitive Information or Patient Discomfort (*** = least sensitive, * = most sensitive) Committee Judgment of the Measure (*** = highest rating, * = lowest rating)
NHIS Smoking Status Questions *** *** *** *** ***
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

useful in clinical assessment of the dimensions of a patient’s dependence on nicotine. These measures are discussed below.

The National Survey on Drug Use and Health has both standard and modified versions; both measures are similar to the NHIS (Ryan et al., 2012). Longer measures are also available, such as the Tobacco Use Supplement to the Current Population Survey (TUS-CPS), which many consider a standard for complete information on national-, state-, and substate-level data from U.S. households regarding smoking, use of tobacco products, and tobacco-related norms, attitudes, and policies (NCI, 2013). The Nicotine Dependence Syndrome Scale (NDSS) (Shiffman et al., 2004), the Wisconsin Inventory of Smoking Dependences Motives (WISDM) (Piper et al., 2004), and the Fagerstrom Test of Nicotine Dependence (FTND) (Etter et al., 1999; Storr et al., 2005) create numeric scores that place the smoker on a range from low dependence to high dependence by focusing on endpoint definitions of dependence, such as heavy smoking, time to first cigarette in the morning, and smoking despite consequences. The Hooked on Nicotine Checklist (HONC) is well suited for use with smokers whose cigarette consumption is low, and it is a reliable and valid measure of diminished autonomy over tobacco (DiFranza et al., 2002).

The Tobacco Dependence Screener (TDS) is a self-report measure that assesses 10 DSM-IV tobacco dependence criteria (Piper et al., 2008). The FTND tends to yield better predictions of cessation outcomes than the NDSS, WISDM, or TDS. Research also shows that the NDSS and WISDM are more sensitive in detecting particular smoking motives (Piper et al., 2006). A single question can screen for smokers in population-based research: “Have you smoked one or more cigarettes in the past month?” (GEM, 2011). However, the committee concluded that the two-question measure aligned better with Meaningful Use SNOMED codes.

The Nicotine Dependence Scale for Adolescents (NDSA) is a six-question measure developed by the FTND and NDSS (NCI, 2012). The scale was designed primarily for survey research, thereby having limited clinical utility.

Alcohol Use

Alcohol is one of the most widely used substances in the world. Because it has both positive and negative health effects depending on type of use, measuring alcohol use involves measuring a continuum of risk. The USPSTF defines alcohol misuse as a variety of behaviors. Multiple validated and reliable measures exist for screening purposes. These measures address frequency of use and associated dependence.

Chapter 3 documents the association between alcohol misuse and adverse health effects. In making recommendations for screening for alco-

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

hol misuse and abuse, the committee believes that benefits gained in guiding patients toward safer alcohol use and/or into treatment for alcohol dependence outweigh the increased time required during a typical primary care encounter. Increased screening enables early intervention.

Identification and Description of Measure

Among multi-question scales measuring alcohol misuse and dependence that have acceptably high reliability and validity, the Alcohol Use Disorders Identification Test Consumption (AUDIT-C) is a three-question instrument, modified from the AUDIT 10-question instrument. It helps identify hazardous drinkers or those who have active alcohol use disorders (Bradley et al., 2007). This instrument is scored across a continuum from “no use” to “serious use,” and it has a different established norm for men and women. The higher the AUDIT-C score the more likely the patient’s drinking affects his or her health and safety.

The AUDIT-C has a sensitivity of 86 percent among patients with heavy drinking or dependence and a specificity of 72 percent (Bush et al., 1998). Cutoff scores greater than or equal to three on the AUDIT-C identify 90 percent of patients with active alcohol abuse or dependence and 98 percent of patients with heavy drinking (Bradley et al., 2007; Bush et al., 1998). This screening instrument appears to be practical for identifying active alcohol abuse or dependence (Bush et al., 1998). The three AUDIT-C questions are:

 

1.   How often do you have a drink containing alcohol?

a.   Never

b.   Monthly or less

c.   2–4 times a month

d.   2–3 times a week

e.   4 or more times a week

2.   How many standard drinks containing alcohol do you have on a typical day?

a.   1 or 2

b.   3 or 4

c.   5 or 6

d.   7 to 9

e.   10 or more

3.   How often do you have six or more drinks on one occasion?

a.   Never

b.   Less than monthly

c.   Monthly

d.   Weekly

e.   Daily or almost daily

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

The questions are scored on a scale of 0 to 12: a = 0 points, b = 1 point, c = 2 points, d = 3 points, and e = 4 points. A score greater than 4 for men or 3 for women is considered to be heavy or hazardous drinking (Babor et al., 2001; Bradley et al., 2007).

Common Metric

While there is no common metric, the USPSTF has guidelines that could be followed regarding alcohol consumption.

Ratings of Measure by Committee

The AUDIT-C questions are freely available to administer in clinical settings, and it is a useful health tool in monitoring patients who are alcohol dependent or who have problem drinking behaviors. The AUDIT-C question takes approximately 1 minute to complete, making it feasible and practical for a patient to complete before a clinical visit. The committee does not believe this is a sensitive question for patients to answer. Because of these considerations, the committee rated the measure as follows in Table 4-18.

Limitations of Measure

If the response to this screen is rated as a positive indication of alcohol use or dependence, the care provider will need to use a more complete test to indicate the extent of the dependence and to refer the patient for appropriate care. Research using the CAGE and AUDIT measures in emergency department settings has shown their use to be feasible, yet they face several barriers owing to time demands and lack of resources to offer screening and brief interventions. Documenting problem drinking may equate to asking the patient to place in the record information contrary to their legal interest. The American College of Emergency Physicians offers a resource kit titled Alcohol Screening and Brief Intervention Resource Kit to their members, which provides information on the benefits of screening, facts about problem drinking, and templates on how to locate community resources (American College of Emergency Physicians, no date; Degutis, 1998; D’Onofrio and Degutis, 2004/2005).

Specific Populations

Screening for alcohol use and intervening with pregnant women is important and challenging (Chang, 2004). Women often alter their drinking patterns after they learn they are pregnant. Quantity and frequency questions on the screening instruments may not show the true risk of women

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

TABLE 4-18 Ratings of the Measure for Alcohol Use

Domain Measure Standard Measure and Freely Available (*** = standard, * = no standard) Usefulness (*** = most useful, * = least useful) Feasible (*** = most feasible, * = least feasible) Lack of Sensitive Information or Patient Discomfort (*** = least sensitive, * = most sensitive) Committee Judgment of the Measure (*** = highest rating, * = lowest rating)
AUDIT-C *** *** *** ** ***
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

with high alcohol levels, a condition early in gestation likely to harm the fetus (Smith et al., 1987). Asking screening questions before pregnancy elicits more accurate measures of a woman’s drinking behavior.

Federal and state prison populations typically show high rates of alcohol misuse, abuse, or dependency. Financial constraints of prison systems may play a role in the extent and quality of the intervention programs offered to those with positive screening results. Most states mandate screening and assessment of driving while intoxicated (DWI) offenses. Sentencing guidelines also recommend that all DWI offenders be screened for alcohol misuse/abuse. The screening instruments discussed above were not developed using prisoner populations nor were they tested in the context of the criminal justice setting. For example, offenders who may feel coerced into screening and treatment or fear being penalized (e.g., unfavorable terms of parole, loss of child custody) if they admit to their actual drinking behavior. Consequently, the validity of the screening results will be in question.

The AUDIT-C has been shown to provide reliable and valid assessment among adolescents ages 14 to 18 years old (Knight et al., 2003). For adolescents and young adults, attention should be given to any level of drinking as well as binge drinking.

Other Measures Reviewed

The committee reviewed several measures for alcohol use, including the single screening question, “On any single occasion during the past 3 months, have you had more than five drinks containing alcohol?,” which can detect at-risk drinking and current alcohol use disorders (Fleming, 2004/2005; Taj et al., 1998). Administration of the single question in a primary care setting demonstrated a positive predictive value of 74 percent and a negative predictive value of 88 percent for problem drinking, with a sensitivity of 62 percent and specificity of 93 percent (Taj et al., 1998). However, the three-question AUDIT-C screen is more reliable in identifying problem drinking. The full 10-question AUDIT has established norms for indicating abuse and alcohol dependence. Scores of 8 or more are considered an indicator of harmful drinking, with a 92 percent sensitivity and 94 percent specificity (Babor et al., 2001). Although this instrument is viewed by many as being very accurate, it is also considered too time consuming in a health care setting because of its lengthy questions. The MAST 25-question instrument was not included because of its length. It was deemed impractical as an initial screening tool in a care setting. It remains an option for use in those screened positive using the single question or AUDIT-C. The CAGE four-question measure was also reviewed by the committee. The questionnaire identifies lifetime abuse or dependence, and most patients in whom alcohol abuse is detected are either actively addressing their substance abuse

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

or are in recovery (American College of Emergency Physicians, no date; Degutis, 1998; D’Onofrio and Degutis, 2004/2005).

INDIVIDUAL-LEVEL SOCIAL RELATIONSHIPS AND LIVING CONDITIONS DOMAIN MEASURES

Social Connections and Social Isolation

Social relationships have been identified as a major psychosocial risk factor for health, and they have been identified as potential resources or buffers mitigating the impact of other risk factors for health, such as stress, and facilitating recovery from acute and chronic diseases (Cassel, 1976; Cobb, 1976). A voluminous body of research documents health effects of a range of social relationships (see review by Holt-Lunstad et al. [2010]), most notably social integration versus isolation (House, 2002), social support (Cohen and Syme, 1985; Uchino, 2009), and loneliness (Cacioppo et al., 2002). The committee considered the importance and relevance of the broad domain of social relationships for EHRs, focusing on the three subdomains just noted.

Identification and Description of Measure

Based on the recent meta-analytic review by Holt-Lunstad et al. (2010), the committee concluded that the updated and adapted Berkman-Syme Social Network Index could be adopted into EHRs. The index derived by Berkman and Syme (1979)—a four-question measure of social integration versus isolation (marital status, frequency of contact with other people, participation in religious activities, and participation in other club or organization activities)—showed an increase risk of all-cause mortality for individuals who were socially isolated. This result has been replicated (with few exceptions) in dozens of studies of broad community populations over more than three decades (e.g., reviews from House et al. [1988] and Holt-Lunstad et al. [2010], and most recently in the NHANES III [Pantell et al., 2013]). The Pantell et al. (2013) study found the mortality relative risks ranged from 1.5 up to 3.0 or higher for the most isolated (lowest quintile) compared to the rest of the population. These risks (and prevalence level for the high-risk category) generally equal or exceed those from cigarette smoking and a wide range of other major behavioral and biomedical risk factors for mortality. Each of the component questions in the measure has also been significantly predictive of all-cause mortality, though expectedly with lower relative risks. This measure has been similarly predictive of a wide range of specific causes of mortality and incidence and course of major diseases, especially cardiovascular disease.

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

The Berkman-Syme Social Network Index, and its scoring, have been appropriately constituted and used in the NHANES III survey. The questions are as follows4 (Pantell et al., 2013):

  1. In a typical week, how many times do you talk on the telephone with family, friends, or neighbors?
  2. How often do you get together with friends or relatives?
  3. How often do you attend church or religious services?
  4. How often do you attend meetings of the clubs or organizations you belong to?

These categories form an ordinal scale that assesses the number of types of social relationships on which a person is connected and not isolated and has standard scoring. Individuals receive one point for each of the following: being married or living together with someone in a partnership at the time of questioning; averaging three or more social interactions per week (assessed with questions one and two, above); reporting attending church or other religious services more than four times per year (assessed with question three, above); and reporting that they belong to a club or organization (assess with question four, above). A score of 0 represents the highest level of social isolation and a score of 4 represents the lowest level of social isolation (Pantell et al., 2013). NHANES data can provide national norms for the data, and for their relation to a range of other psychosocial, behavioral, and biomedical risk factors.

These questions also have high relevance for clinical practice as they provide a picture of the social context in which patients live. Marital status, active participation in religious or other organizations, and regular informal contact with friends and family are resources that can help patients adhere to medical regimes, promote health behaviors and deter risky ones, provide avenues for health screening, and generally provide both structure and support in individuals’ lives.

Common Metric

There is no common metric for social connection and isolation at this time.

_________________

4 Marital status information is collected in the demographic section of NHANES that includes the following options: married, widowed, divorced, separated, never married, and living with partner. Marital status is part of the scoring and not included in this list of questions. It will need to be adapted accordingly for patient self-reporting for use in EHRs.

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

Ratings of Measure by Committee

The four-question set is without copyright protections. On the basis of their strongly established relationship to health, potential value in clinical practice, ease of administration, and the lack of sensitive information that might cause patient discomfort, the committee has given this measure the following ratings, see Table 4-19.

Limitations of Measure

The committee did not identify any limitations in using the four-question measure.

Specific Populations

The suggested social integration versus isolation measure has demonstrated to be easily administered and strongly relates to health across the full range of the adult population. Such evidence is lacking for children, especially at younger ages. The social integration versus isolation measure for a child’s parent or guardian may prove useful, as might measures of attachment and quality of relationship with parents or guardians. These are considered in other domains examined by the committee. Tools exist for geriatric populations to measure social isolation and disconnectedness, as older adults and those in worse health tend to experience greater levels of social isolation. However, the committee felt the four-question NHANES III measure was appropriate for use in all adults.

Other Measures Reviewed

Although similar evidence exists for the relation of measures of social support and of loneliness to health, it is not as strong and consistent as that for social integration versus isolation, nor is there as clearly a consensual measure that is easily administered (Holt-Lunstad et al., 2010; Steptoe et al., 2013). Both constructs, and measures of them, merit further consideration in future efforts to expand the inclusion of psychosocial factors in EHRs. Social contact via electronic media may emerge in the future as an additional aspect of social relationships.

Exposure to Violence

Interpersonal or domestic partner violence involves actual or threatened physical, sexual, psychological, or emotional abuse by a family member, caregiver, current or former spouse or partner, or dating partner. Inter-

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

TABLE 4-19 Ratings of the Measure for Social Connections and Social Isolation

Domain Measure Standard Measure and Freely Available (*** = standard, * = no standard) Usefulness (*** = most useful, * = least useful) Feasible (*** = most feasible, * = least feasible) Lack of Sensitive Information or Patient Discomfort (*** = least sensitive, * = most sensitive) Committee Judgment of the Measure (*** = highest rating, * = lowest rating)
NHANES III Social Connections and Isolation Questions *** *** *** *** ***
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

personal violence can take many forms, including child and elder abuse and neglect and intimate partner violence. Assessment and management of child abuse and recording such information in the clinical record are a matter of law and will not be addressed specifically in this section. Intimate partner violence refers to the experience of being hit, slapped, pushed, or otherwise harmed by someone identified as a romantic partner. Elder abuse may be perpetrated by intimate partners, other family members—including someone’s own children—as well as strangers. The USPSTF reviewed evidence related to violence against elders (USPSTF, 2012, 2013). Although mindful of the serious nature of elder abuse, the USPSTF found insufficient evidence linking improved health outcomes to routine assessment of elders for exposure to violence. The USPSTF did, however, recommend screening for intimate partner violence, but only for women of reproductive age.

Identification and Description of Measure

Despite strong evidence of the links between interpersonal violence and health, existing recommendations for screening are limited to intimate partner violence screening of women in childbearing years, following the USPSTF recommendation. Thus the committee limited its discussion of metrics and measures to intimate partner violence.

Intimate partner violence, a subdomain of interpersonal violence, refers specifically to violence within romantic relationships. Intimate partners can be of the same or the opposite sex (CDC, 2006b). The National Intimate Partner and Sexual Violence Survey documented that one in three women have experienced physical violence (Black et al., 2011). Intimate partner violence is associated with life-threatening injuries as well as long-term physical and mental health problems, and it may account for 20 percent of all homicides (de Boinville, 2013). Intimate partner violence has been implicated as a chronic stressor leading to substance abuse, depression, and other mental health problem. Health care costs are generally higher for those experiencing intimate partner violence, and undetected and untreated intimate partner violence can lead to poor health outcomes.

Screening for intimate partner violence in health care settings is one of the eight preventive health services now covered in new health plans without requiring a copayment, coinsurance, or deductible5 (HRSA, 2012). However, screening recommendations remain inconsistent across various health organizations and agencies. As noted above, the USPSTF calls for clinicians to screen women of childbearing age and offers no recommendations regarding assessment of women outside of this age range or for men. The American College of Gynecologists recommends screening of all

_________________

5 The Patient Protection and Affordable Care Act, Public Law 111-148 § 2713.

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

women (ACOG, 2012), while the World Health Organization (WHO) recommends against universal screening for any women (WHO, 2013). Screening for intimate partner violence is believed to have positive outcomes if the health provider has the ability to provide for or refer to interventional services.

At this time there is not a screen that has well-established psychometric properties (Rabin et al., 2009); nonetheless, it is important to screen women (IOM, 2011a) of reproductive age. Successful implementation of broad-based intimate partner violence screening rests in part on the ability to provide for or refer to interventional services (McCaw, 2013). The committee reviewed the HARK (Humiliation, Afraid, Rape, Kick), which is a four-question self-reported instrument that represents different components of interpersonal violence, including emotional, sexual, and physical abuse (Sohal et al., 2007). The questions are:

1.   Within the last year, have you been humiliated or emotionally abused in other ways by your partner or ex-partner?

Yes No

2.   Within the last year, have you been afraid of your partner or ex-partner?

Yes No

3.   Within the last year, have you been raped or forced to have any kind of sexual activity by your partner or ex-partner?

Yes No

4.   Within the last year, have you been kicked, hit, slapped, or otherwise physically hurt by your partner or ex-partner?

Yes No

Each question answered with a yes is a score of 1. Scores then can range from 0 to 4. The sensitivity of the optimal clinical cutoff score of one or more was 81 percent and the specificity is 95 percent (Nelson et al., 2012).

Common Metric

There is not a common metric for intimate partner violence at this time.

Ratings of Measure by Committee

The clinical questions are freely available. This four-question screen is clearly useful for clinical interventions, especially because this domain is linked to long-term health outcomes. The committee deemed this domain highly sensitive, noting the strong evidence of shame and desire to avoid reporting violence due to perceptions of increased risk (IOM, 2011a). Other concerns include unanticipated disclosure that could occur from the inad-

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

vertent distribution of an after-visit summary, and potential provider-based prejudice emerging from knowledge of a past history of abuse. Because of these considerations, the committee rated the measure as follows in Table 4-20.

Limitations of Measure

Limitations for this measure include that it is highly sensitive information potentially causing patient discomfort, and the narrow focus of intimate partner violence, rather than a larger lens of interpersonal violence. Another limitation may be for health care providers who lack experience in screening and follow-up interventions that are necessary for improved outcomes (IOM, 2011a).

Specific Populations

The USPSTF neglected to extend the recommendations for screening to women over the age of 46 or to men. Some writers (Connelly et al., 2000) posit the need for additional intensive screening for women with high-risk pregnancies.

Immigrant women may be hesitant to report intimate partner violence because of differences in cultural perceptions or for fear of deportation (Committee on Health Care for Underserved Women, 2012).

Adolescent females are a population that has reported experiencing physical dating violence (Silverman et al., 2004); however, assessments of intimate partner violence for adolescents or children are not currently available.

Other Measures Reviewed

The committee reviewed other assessments of risk of violence; however, most lacked standards for validity. The committee also reviewed instruments specific to perpetrators (e.g., The Violence Risk Scale [Olver et al., 2014]—sexual offender version—and the Historical, Clinical and Risk management-20 [HCR 20]); however, their validity and reliability to predict violence remains to be determined (Dolan and Doyle, 2000). According to the USPSTF, in addition to HARK there are five intimate partner violence screening tools with good sensitivity and specificity (USPSTF, 2012): Hurt, Insult, Threaten, Scream (HITS) (English and Spanish versions); Ongoing Abuse Screen/Ongoing Violence Assessment Tool (OAS/OVAT); Slapped, Threatened, and Throw (STaT); Modified Childhood Trauma Questionnaire–Short Form (CTQ-SF); and Woman Abuse Screen Tool (WAST). The HITS instrument includes four questions, can be used

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

TABLE 4-20 Ratings of the Measure for Exposure to Violence

Domain Measure Standard Measure and Freely Available (*** = standard, * = no standard) Usefulness (*** = most useful, * = least useful) Feasible (*** = most feasible, * = least feasible) Lack of Sensitive Information or Patient Discomfort (*** = least sensitive, * = most sensitive) Committee Judgment of the Measure (*** = highest rating, * = lowest rating)
HARK *** *** ** * ***
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

in a primary care setting, and is available in both English and Spanish. It can be self- or clinician-administered. STaT is a three-question self-report instrument that was tested in emergency department settings. All assess current or past exposure rather than risk of future exposure. The USPSTF found no risk inherent in the screening, and modest positive gains in safety, health, and injury mitigation arising from early intervention (USPSTF, 2012). Another measure reviewed by the committee was a two-question screener recommended by the Committee on Health Care for Underserved Women (2012), but it lacked a clinical cutoff.

NEIGHBORHOODS AND COMMUNITIES DOMAIN MEASURES

Neighborhood and Community Compositional Characteristics

A large body of work has used measures of neighborhood socioeconomic characteristics to investigate the impact of neighborhood contexts on health (Diez-Roux and Mair, 2010). Neighborhood measures have also been used as proxies for individual-level socioeconomic information when it is not available. Area socioeconomic characteristics are derived from various summary measures of the compositional characteristics of neighborhoods and communities that can be created from routinely collected census data. The “areas” for which these measures can be calculated are many, but census tracts are one of the ones most commonly used as proxies for residential neighborhoods. Examples of measures include median household income; percent below poverty; the percent of persons who have graduated college; the percent of persons in managerial, professional, or executive occupations; and the unemployment rate. Various summary measures of “area socioeconomic position,” derived theoretically or using techniques such as factor analysis, have been created (Diez-Roux and Mair, 2010). Neighborhood and community compositional characteristics, such as area socioeconomic measures, have been shown to be related to various health outcomes. Even though individual-level characteristics have a stronger association with health outcomes than do neighborhood characteristics, there is an independent contribution of neighborhood and community compositional characteristics above and beyond individual factors (Diez-Roux and Mair, 2010). Neighborhood and community compositional characteristics have also been shown to be useful in predicting health risk (Fiscella and Franks, 2001; Fiscella et al., 2009; Kim et al., 2010; Pollack et al., 2012; Vortuba and Kling, 2008) and outcomes of care for individual patients (Gerber et al., 2010). Recent work has also employed these measures in health services research (Nagasako et al., 2014).

In Chapter 3 (part of Phase 1), the committee considered race/ethnicity composition of an area as a domain under neighborhood and community

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

compositional characteristics. For this chapter (part of Phase 2), which focuses on the measures, the committee opted to focus on area socioeconomic geocodable (defined in the following section) measures and not on race/ethnic composite of an area because race/ethnicity of an individual patient is routinely collected in EHRs.

Identification and Description of Measure

To characterize neighborhood and community compositional characteristics (such as area SES) for a patient’s residential address, information needs to be obtained. This information must be collected in a standardized manner so residential addresses can then be “geocoded.” Geocoded residential addresses have geographic identifiers—latitude and longitudinal coordinates—attached to census codes. Once geocodes are available the location can be linked to geographically referenced data from the American Community Survey (ACS) to characterize area socioeconomic characteristics. The geocodes can also be used to link locations to various sources of neighborhood contextual characteristics, including measures of spatial access to resources, air pollution data, crime rates, or measures of the built environment—for example, walkability score (Philadelphia Department of Public Health, 2013; TRB and IOM, 2005; WHO, 2011). The committee prioritizes the physical environment, and as more geocodable data become available and can be linked to census data (i.e., economic or occupation indicators or measures of racial/ethnic composition) as well as measures created through other linkages to other data (i.e., population health surveys), this type of data linking can occur. This information may also be linked to geocoded patient data providing a demographic neighborhood and community profile of the patient’s living conditions.

The ACS provides data on the median household income for various census tracts (U.S. Census Bureau, 2014a). Median household income is a continuous measure that can be used to capture variability across areas. Because census tracts were at least initially defined to be approximately homogeneous in socioeconomic characteristics, and because they are used in many analyses as proxies for neighborhoods, the committee suggests using census tract measures. The median household income measured in current U.S. dollars at the level of the census tract can serve as the standard measure for this domain. In addition, because the ACS is carried out only on a sample of households each year, the committee suggests pooling the data across multiple years to obtain the estimates for income, with the number of years serving as a function of the sample size per year in any given tract.

Geocoding a patient’s residential address takes the attributes of a street address, compares them to a database of addresses in a geographic information system (GIS), and assigns coordinates based on the best match

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

(CA.gov, 2014). The four-digit extension is added on via program coding. An example of how to collect patient address in a standard way is listed below:

  • House Number + Directional (such as North, South, etc.) + Street Name
  • City
  • State
  • Zip Code + 4-digit extension

Common Metric

There is not a common metric for neighborhood and community compositional characteristics at this time.

Ratings of Measure by Committee

Geocodable patient address and census tract-median household income are neighborhood indicators that can be useful when systematically included in the EHR. The measures are standard and easy to obtain in a systematic way from the ACS; they are useful at a population and a clinical level (especially in a context where individual-level income data are unlikely to be available); they are feasible; and providing an address to enable geocoding and the resulting census-tract information is not sensitive. Because of these considerations, the committee rated the measures as follows in Table 4-21.

Limitations of Measure

Collecting a patient’s residential address in a standardized way is necessary for the geocodable data to be linked to the patient’s record. For patients who move often or are without stable and permanent housing, this could be difficult. Another limitation is that in order for geocoding to work, it must be completely standardized. Geocoded addresses that are not properly formatted or collected will lack accuracy if addresses are coded using only zip codes (Rushton et al., 2007). Zip codes are not geographic areas and do not have exact spatial bounds. As such, there is no real correlation between zip codes and census geography; thus, statistical analyses are conducted to estimate zip code populations associated with health outcomes (CA.gov, 2014).

Specific Populations

Median household income by census tract is relevant for all age groups. Other neighborhood characteristics such as proximity to schools and play-

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

TABLE 4-21 Ratings of the Measure for Neighborhood and Community Compositional Characteristics

Domain Measures Standard Measure and Freely Available (*** = standard, * = no standard) Usefulness (*** = most useful, * = least useful) Feasible (*** = most feasible, * = least feasible) Lack of Sensitive Information or Patient Discomfort (*** = least sensitive, * = most sensitive) Committee Judgment of the Measure (*** = highest rating, * = lowest rating)
Geocoded Residential Address *** *** *** *** ***
Census Tract-Median Household Income *** *** ** *** ***
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

grounds may be particularly relevant for children; environmental exposures may be particularly relevant to individuals with asthma or other respiratory ailments; and the age structure and proximity to pharmacies and health care may be particularly relevant to older individuals.

Other Measures Reviewed

A wealth of other physical and social environment data can be linked to geocoded address information. Measures of land use, urban design, and walkability include measures such as the extent to which an area includes residential and other (commercial or retail) uses, proximity to various types of uses (e.g., shops, social destinations), whether street networks are interconnected in ways that allow easy transportation by walking, and other features of design such as the presence of sidewalks that may encourage or detract from walking for transportation or leisure. These measures can be calculated using a GIS in conjunction with routine and specially collected geographically referenced data. Summary measures (such as the walkability score) that combine information from several of these domains have also been created. Access to resources such as healthy foods and recreational facilities can also be characterized using a GIS, as can environmental exposure data, such as levels of air pollutants or proximity to highways or hazardous sites. Geocoded address data can also be linked to crime data, when available, at a disaggregated data level or to other survey data that can be used to characterize constructs such as social capital and social cohesion or levels of safety for neighborhoods.

Although all these measures have potential clinical and population usefulness, the data required to create them is not routinely available in a standardized format. In addition, the processes used to create the measures can be complex, and a number of different measures exist. The validity and usefulness of different types of measures remains a topic of active research. For this reason the committee does not suggest the inclusion of any specific measures of these neighborhood contextual domains in the EHR at this point.

However, it is expected that the availability of geocode information in the EHR will stimulate further research on the value of these measures and may justify including additional measures in the EHR in a systematic way in the future. The committee hopes that in the future, variables related to compositional characteristics of the neighborhood and measures reflecting contextual characteristics, such as hazards and resources in the physical and social environment, will be standardized and routinely collected and thus able to be linked to all patient records.

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

SUMMARY OF CANDIDATE DOMAIN MEASURES REVIEWED BY THE COMMITTEE

A summary of the candidate domains and the related measures that were reviewed are shown in Table 4-22.

TABLE 4-22 Summary of All Domain Measures with Committee Ratings

Domains and Corresponding Measures Standard Measure and Freely Available (*** = standard, * = no standard Usefulness (*** = most, * = least)
Sexual Orientation

Self-Identity (1 Q)

*** *

Sexual Behavior (1 Q)

*** **
Race and Ethnicity

U.S. Census (2 Q)

*** ***

OMB Definition (2 Q)

*** **
Country of Origin/U.S. Born Versus Non-U.S. Born

U.S. Census (2 Q)

*** **
Education

Educational Attainment (2 Q)

*** ***
Employment

MESA Employment Question (1 Q)

** **
Financial Resource Strain

Overall Financial Resource Strain (1 Q)

*** ***

Food Insufficiency (1 Q)

*** **

Housing Insecurity (1 Q)

* *
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Feasible (*** = most feasible, * = least feasible) Lack of Sensitive Information or Patient Discomfort (*** = least sensitive, * = most sensitive) Committee Judgment of the Measure (*** = highest rating, * = lowest rating)
*** ** **
*** * **
*** *** ***
*** *** ***
*** * **
*** *** ***
*** ** **
*** ** ***
** ** ***
* * *
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Domains and Corresponding Measures Standard Measure and Freely Available (*** = standard, * = no standard Usefulness (*** = most, * = least)
Health Literacy

Chew et al. (2004) (3 Q)

*** *
Stress

Adverse Childhood Experiences (ACE) (11 Q)

*** *

Elo et al. (2003) (1 Q)

*** ***
Depression

Patient Health Questionnaire (PHQ)-2 (2 Q)

*** ***

PROMIS-8b (8 Q)

*** *
Anxiety

GAD-7 (7 Q)

*** **

PROMIS-7a (7 Q)

*** **
Conscientiousness

Big Five Inventory-10 Item (1 Q)

* **
Patient Engagement/Activation

PAM

* *
Optimism

Life Orientation Test-Revised (6 Q)

*** **
Self-Efficacy

Self-Efficacy Scales for Specific Behaviors

* *

NIH Toolbox (10 Q)

*** *
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Feasible (*** = most feasible, * = least feasible) Lack of Sensitive Information or Patient Discomfort (*** = least sensitive, * = most sensitive) Committee Judgment of the Measure (*** = highest rating, * = lowest rating)
*** ** *
** * *
*** ** ***
*** ** ***
** ** *
*** ** **
*** ** **
* *** *
** *** *
*** *** **
* *** *
*** *** *
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Domains and Corresponding Measures Standard Measure and Freely Available (*** = standard, * = no standard Usefulness (*** = most, * = least)
Dietary Patterns

Fruit and Vegetable Consumption (2 Q)

*** **
Physical Activity

Exercise Vital Sign (2 Q)

*** ***

Accelometer

*** **
Tobacco Use and Exposure

NHIS Smoking Status Questions (2 Q)

*** ***
Alcohol Use

AUDIT-C (3 Q)

*** ***
Social Connections and Social Isolation

NHANES III (4 Q)

*** ***
Exposure to Violence

Intimate Partner Violence: HARK (4 Q)

*** ***
Neighborhoods and Communities Compositional Characteristics

Geocoded Residential Address

*** ***

Census Tract-Median Household Income

*** ***

REFERENCES

ACOG (American College of Obstetricians and Gynecologists). 2012. Committee on Health Care for Underserved Women: Committee Opinion. http://www.acog.org/ResourcesAnd-Publications/Committee-Opinions/Committee-on-Health-Care-for-UnderservedWomen/Intimate-Partner-Violence (accessed October 23, 2014).

Adsit, R., and M. C. Fiore. 2013. Assessing tobacco use. The national landscape August 2013. Madison: University of Wisconsin School of Medicine and Public Health Center for Tobacco Research and Intervention.

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Feasible (*** = most feasible, * = least feasible) Lack of Sensitive Information or Patient Discomfort (*** = least sensitive, * = most sensitive) Committee Judgment of the Measure (*** = highest rating, * = lowest rating)
*** *** **
*** *** ***
* *** *
*** *** ***
*** ** ***
*** *** ***
** * ***
*** *** ***
** *** ***

 

AHRQ (Agency for Healthcare Research and Quality). 2009. Health literacy measurement tools: Fact sheet. http://www.ahrq.gov/professionals/quality-patient-safety/qualityresources/tools/literacy/index.html (accessed August 26, 2014).

AHRQ. 2014. Health literacy universal precautions toolkit. Rockville, MD: Agency for Healthcare Research and Quality. http://www.ahrq.gov/professionals/quality-patientsafety/quality-resources/tools/literacy-toolkit/index.html (accessed August 28, 2014).

Alaimo, K., R. R. Briefel, E. A. Frongillo, and C. M. Olson. 1998. Food insufficiency exists in the United States: Results from the third National Health and Nutrition Examination Survey (NHANES III). American Journal of Public Health 88(3):419–426.

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

Alaimo, K., C. M. Olson, and E. A. Frongillo, Jr. 2001. Low family income and food insufficiency in relation to overweight in US children: Is there a paradox? Archives of Pediatrics & Adolescent Medicine 155(10):1161–1167.

Alegria, M. 2009. The challenge of acculturation measures: What are we missing? A commentary on Thomson & Hoffman-Goetz. Social Science & Medicine 69(7):996–998.

Almader-Douglas, D. 2013. Culture in the context of health literacy: Update. M. Eberle. Bethesda, MD: National Library of Medicine. http://nnlm.gov/outreach/consumer/hlthlit.html (accessed December 9, 2013).

American College of Emergency Physicians. no date. Alcohol screening and brief intervention resource kit. ACEP Product No. 409036. http://www.acep.org (follow link to Practice Resources) (accessed May 16, 2014).

Amtmann, D., J. Kim, H. Chung, A. M. Bamer, R. L. Askew, S. Wu, K. F. Cook, K. L. Johnson. 2014. Comparing CESD-10, PHQ-9, and PROMIS depression instruments in individuals with multiple sclerosis. Rehabilitation Psychology 59(2):220–229.

Anda, R. F., V. J. Fellitti, J. D. Bremner, J. D. Walker, C. Whitfield, B. D. Perry, S. R. Dube, and W. H. Giles. 2005. The enduring effects of abuse and related adverse experiences in childhood. A convergence of evidence from neurobiology to epidemiology. European Archive of Psychiatry and Clinical Neuroscience 256:174–186.

Anderson, E. S., R. A. Winett, and J. R. Wojcik. 2000. Social-cognitive determinants of nutrition behavior among supermarket food shoppers: A structural equation analysis. Health Psychology 19(5):479–486.

Ansari, Z., N. J. Carson, M. J. Ackland, L. Vaughan, and A. Serraglio. 2003. A public health model of the social determinants of health. Soz Präventivmed 48(4):242–251.

APA (American Psychological Assocition). 2014. Stress: The different kinds of stress. http://www.apa.org/helpcenter/stress-kinds.aspx (accessed July 10, 2014).

Arcia, E., M. Skinner, D. Bailey, and V. Correa. 2001. Models of acculturation and health behaviors among Latino immigrants to the US. Social Science & Medicine 53(1):41–53.

ASPE (Assistant Secretary for Planning and Evaluation). no date–a. Percentage of U.S. Adults who smoke cigarettes. https://healthmeasures.aspe.hhs.gov/measure/37a (accessed August 5, 2014).

ASPE. no date–b. Percentage of U.S. high school students who smoke cigarettes. https://healthmeasures.aspe.hhs.gov/measure/37b (accessed August 7, 2014).

Babor, T. F., J. C. Higgins-Biddle, J. B. Saunders, and M. G. Monteiro. 2001. AUDIT: The alcohol use disorders identification test. Guidelines for use in primary care. Geneva, Switzerland: World Health Organization, Department of Mental Health and Substance Dependence.

Baker, A. H., and J. Wardle. 2003. Sex differences in fruit and vegetable intake in older adults. Appetite 40(3):269–275.

Baker, D. W., R. M. Parker, M. V. Williams, and W. S. Clark. 1998. Health literacy and the risk of hospital admission. Journal of General Internal Medicine 13(12):791–798.

Bandura, A. 2006. Guide for constructing self-efficacy scales. In Self-efficacy beliefs of adolescents, edited by F. Pajares and T. Urdan. Charlotte, NC: Information Age Publishing. Pp. 307–337.

Basiotis, P. P. 1992. Validity of the self-reported fodd sufficiency status item in the U.S. Department of Agriculture’s food consumption surveys. In American Council of Consumer Interests 38th annual conference: The proceedings, V. A. Haldmen, ed. Columbia, MO: American Council of Consumer Interests.

BDA (British Dietetic Association). 2014. Fruit and vegetables—how to get five-a-day. United Kingdom: British Dietetic Association.

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

Berkman, L. F., and S. L. Syme. 1979. Social networks, host resistance, and mortality: A nine-year follow-up study of Alameda County residents. American Journal of Epidemiology 109(2):186–204.

Berkman, N. D., D. A. DeWalt, M. P. Pignone, S. L. Sheridan, K. N. Lohr, L. Lux, S. F. Sutton, T. Swinson, and A. J. Bonito. 2004. Literacy and health outcomes. Rockville, MD: Agency for Healthcare Research and Quality.

Black, M. C., K. C. Basile, M. J. Breiding, S. G. Smith, M. L. Walters, M. T. Merrick, J. Chen, and M. R. Stevens. 2011. The National Intimate Partner and Sexual Violence Survey (NISVS): 2010 summary report. Atlanta, GA: National Center for Injury Prevention and Control, Centers for Disease Control and Prevention.

Blackwell, D. L., J. W. Lucas, and T. C. Clarke. 2014. Summary health statistics for U.S. adults: National Health Interview Survey, 2012. Washington, DC: National Center for Health Statistics.

Bogg, T., and B. W. Roberts. 2004. Conscientiousness and health-related behaviors: A meta-analysis of the leading behavioral contributors to mortality. Psychology Bulletin 130(6): 887–919.

Bradley, K. A., A. F. DeBenedetti, R. J. Volk, E. C. Williams, D. Frank, and D. R. Kivlahan. 2007. AUDIT-C as a brief screen for alcohol misuse in primary care. Alcoholism: Clinical and Experimental Research 31(7):1208–1217.

Brenk-Franz, K., J. H. Hibbard, W. J. Herrmann, T. Freund, J. Szecsenyi, S. Djalali, C. Steurer-Stey, A. Sonnichsen, F. Tiesler, M. Storch, N. Schneider, and J. Gensichen. 2013. Validation of the German version of the Patient Activation Measure 13 (PAM13-D) in an international multicentre study of primary care patients. PLoS ONE 8(9). doi: 10.137/journal. pone.0074786. http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0074786 (accessed April 3, 2014).

Briefel, R. R., and C. E. Woteki. 1992. Development of the food sufficiency questions for the third National Health and Nutrition Examination Survey. Journal of Nutrition Education 24(Suppl. 1):24s–28s.

Brown, D., R., R. Ludwig, G. A. Buck, M. D. Durham, T. Shumard, and S. S. Graham. 2004. Health literacy: Universal precautions needed. Journal of Allied Health 33(2):150–155.

Brown, D. W., R. F. Anda, H. Tiemeier, V. J. Felitti, V. J. Edwards, J. B. Croft, and W. H. Giles. 2009. Adverse childhood experiences and the risk of premature mortality. American Journal of Preventive Medicine 37(5):389–396.

Bush, K., D. R. Kivlahan, M. B. McDonell, S. D. Fihn, and K. A. Bradley. 1998. The AUDIT alcohol consumption questions (AUDIT-C): An effective brief screening test for problem drinking. Ambulatory Care Quality Improvement Project (ACQUIP). Alcohol use disorders identification test. Archives of Internal Medicine 158(16):1789–1795.

CA.gov. 2014. Geocoding service: Frequently asked questions. http://www.ehib.org/page.jsp?page_key=360#geocoding_FAQ_coordinatesystem (accessed August 13, 2014).

Cacioppo, J. T., L. C. Hawkley, L. E. Crawford, J. M. Ernst, M. H. Burleson, R. B. Kowalewski, W. B. Malarkey, E. Van Cauter, and G. G. Berntson. 2002. Loneliness and health: Potential mechanisms. Psychosomatic Medicine 64(3):407–417.

Cappuccio, F. P., E. Rink, L. Perkins-Porras, C. McKay, S. Hilton, and A. Steptoe. 2003. Estimation of fruit and vegetable intake using a two-item dietary questionnaire: A potential tool for primary health care workers. Nutrition, Metabolism, and Cardiovascular Diseases 13:12–19.

CARDIA (Coronary Artery Risk Development in Young Adults). no date. CARDIA. Exam Materials. http://www.cardia.dopm.uab.edu/exam-materials2 (accessed August 26, 2014).

Caspersen, C. J., K. E. Powell, and G. M. Christenson. 1985. Physical activity, exercise, and physical fitness: Definitions and distinctions for health-related research. Public Health Reports 100(2):126–151.

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

Cassel, J. 1976. The contribution of the social environment to host resistance. American Journal of Epidemiology 104(2):107–122.

CDC (Centers for Disease Control and Prevention). 2006a. Do increased portion sizes affect how much we eat? Research to Practice Series, no. 2. Atlanta, GA: National Center for Chronic Disease Prevention and Health Promotion, Division of Nutrition and Physical Activity.

CDC. 2006b. Preventing intimate partner violence, sexual violence, and child maltreatment. http://www.cdc.gov/ncipc/pub-res/research_agenda/07_violence.htm (accessed May 5, 2014).

CDC. 2014a. Current cigarette smoking among adults—United States, 2005–2012. Morbidity and Mortality Weekly Report 63(2):29–46.

CDC. 2014b. HIV among African Americans fact sheet. http://www.cdc.gov/hiv/risk/racialethnic/aa/facts/index.html (accessed July 21, 2014).

CDC. 2014c. Tobacco-related mortality. http://www.cdc.gov/tobacco/data_statistics/fact_sheets/health_effects/tobacco_related_mortality/index.htm#cigs (accessed July 30, 2014).

Chang, G. 2004. Screening and brief intervention in prenatal care settings. Alcohol Research and Health 28(2):80.

Chew, L. D., C. J. Bradley, and E. J. Boyko. 2004. Brief questions to identify patients with inadequate health literacy. Family Medicine 36(8):588–594.

Chew, L., J. Griffin, M. Partin, S. Noorbaloochi, J. Grill, A. Snyder, K. Bradley, S. Nugent, A. Baines, and M. VanRyn. 2008. Validation of screening questions for limited health literacy in a large VA outpatient population. Journal of General Internal Medicine 23(5):561–566.

CHIS (California Health Interview Survey). 2014. Report 3: Data processing procedures. Los Angeles, CA: UCLA Center for Health Policy Research.

CHIS. no date. LGBTdata.Com. California Health Interview Survey. http://www.lgbtdata.com/california-health-interview-survey-chis.html (accessed August 8, 2014).

Choi, S. W., B. Schalet, K. F. Cook, and D. Cella. 2014. Establishing a common metric for depressive symptoms: Linking the BDI-II, CES-D, and PHQ-9 to PROMIS Depression. Psychological Assessment 26(2):513–527.

Christofar, S. P., and P. P. Basiotis. 1992. Dietary intakes and selected characteristics of women ages 19-50 years and their children ages 1–5 years by reported perception of food sufficiency. Journal of Nutrition Education 24(2):53–58.

CMS (Centers for Medicare & Medicaid Services). 2012. Stage 2: Eligible professional: Meaningful Use core measures: Measure 5 of 17. Washington, DC: Centers for Medicare & Medicaid Services. http://www.cms.gov/Regulations-and-Guidance/Legislation/EHRIncentivePrograms/downloads/Stage2_EPCore_5_RecordSmokingStatus.pdf (accessed July 30, 2014).

Cobb, S. 1976. Social support as a moderator of life stress. Psychosomatic Medicine 38:300–314.

Cohen, S., and L. Syme. 1985. Social support and health. New York: Academic Press.

Coleman. 2012. Initial validation of an exercise “vital sign” in electronic medical records. Medicine Science in Sports Exercise 44(11):2071–2076.

Commission to Build a Healthier America. 2009. Beyond health care: New directions to a healthier America. Princeton, NJ: Robert Wood Johnson Foundation.

Committee on Health Care for Underserved Women. 2012. Intimate partner violence. Washington, DC: The American Congress of Obstetricians and Gynecologists. https://www.acog.org/Resources-And-Publications/Committee-Opinions/Committee-on-Health-Care-for-Underserved-Women/Intimate-Partner-Violence#21 (accessed July 7, 2014).

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

Connelly, C. D., R. R. Newton, J. Landsverk, and G. A. Aarons. 2000. Assessment of intimate partner violence among high-risk postpartum mothers: Concordance of clinical measures. Women & Health 31(1):21–37.

Davis, T. C., M. S. Wolf, C. L. Arnold, R. S. Byrd, S. W. Long, T. Springer, E. Kennen, and J. A. Bocchini. 2006. Development and validation of the Rapid Estimate of Adolescent Literacy in Medicine (REALM-Teen): A tool to screen adolescents for below-grade reading in health care settings. Pediatrics 118(6):e1707–e1714.

de Boinville, M. 2013. Screening for domestic violence in health care settings. ASPE Policy Brief (August):1–14.

Degutis, L. C. 1998. Screening for alcohol problems in emergency department patients with minor injury: Results and recommendations for practice and policy. Contemporary Drug Problems 25:463–475.

DiClemente, C. C., J. P. Carbonari, R. P. Montgomery, and S. O. Hughes. 1994. The alcohol abstinence self-efficacy scale. Journal of Studies on Alcohol and Drugs 55(2):141–148.

Diez-Roux, A. V., and C. Mair. 2010. Neighborhoods and health. Annals of the New York Academy of Sciences 1186(1):125–145.

DiFranza, J. R., J. A. Savageau, K. Fletcher, J. K. Ockene, N. A. Rigotti, A. D. McNeill, M. Coleman, and C. Wood. 2002. Measuring the loss of autonomy over nicotine use in adolescents: The DANDY (Development and Assessment of Nicotine Dependence in Youths) Study. Archives of Pediatric Adolescent Medicine 156(4):397–403.

Dinour, L. M., D. Bergen, and M.-C. Yeh. 2007. The food insecurity–obesity paradox: A review of the literature and the role food stamps may play. Journal of the American Dietetic Association 107(11):1952–1961.

Dolan, M., and M. Doyle. 2000. Violence risk prediction: Clinical and actuarial measures and the role of the psychopathy checklist. British Journal of Psychiatry 177(4):303–311.

D’Onofrio, G., and L. C. Degutis. 2004/2005. Screening and brief intervention in the emergency department. Alcohol Research & Health 28(2):63–72.

Elo, A.-L., A. Leppänen, and A. Jahkola. 2003. Validity of a single-item measure of stress symptoms. Scandinavian Journal of Work, Environment & Health 29(6):444–451.

Ensminger, M. E., C. B. Forrest, A. W. Riley, M. Kang, B. F. Green, B. Starfield, and S. A. Ryan. 2000. The validity of measures of socioeconomic status of adolescents. Journal of Adolescent Research 15(3):392–419.

Etter, J. F., T. V. Duc, and T. V. Perneger. 1999. Validity of the Fagerström test for nicotine dependence and of the heaviness of smoking index among relatively light smokers. Addiction 94(2):269–281.

Etter, J. F., M. M. Bergman, J. P. Humair, and T. V. Perneger. 2000. Development and validation of a scale measuring self-efficacy of current and former smokers. Addiction 95(6):901–913.

Ey, S., W. Hadley, D. N. Allen, S. Palmer, J. Klosky, D. Deptula, J. Thomas, and R. Cohen. 2005. A new measure of children’s optimism and pessimism: The youth life orientation test. Journal of Child Psychology and Psychiatry 46(5):548–558.

Felitti, V. J., R. F. Anda, D. Nordenberg, W. D. F., A. M. Spitz, V. Edwards, M. Koss, and J. S. Marks. 1998. Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults. The Adverse Childhood Experiences (ACE) Study. American Journal of Preventive Medicine 14(4):245–258.

Fenway Institute and Center for American Progress. 2013. Asking patients questions about sexual orientation and gender identity in clinical settings. Boston, MA: Fenway Institute and Center for American Progress.

Fiscella, K., and P. Franks. 2001. Impact of patient socioeconomic status on physician profiles: A comparison of census-derived and individual measures. Medical Care 39(1):8–14.

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

Fiscella, K., D. Tancredi, and P. Franks. 2009. Adding socioeconomic status to Framingham scoring to reduce disparities in coronary risk assessment. American Heart Journal 157(6):988–994.

Fleming, M. F. 2004/2005. Screening and brief intervention in primary care settings. Alcohol Research & Health 28(2):57–62.

Franzini, L., J. C. Ribble, and A. M. Keddie. 2001. Understanding the Hispanic paradox. Ethnicity & Disease 11(3):496–518.

GEM (Grid-Enabled Measures Database). 2011. Tobacco use screener. https://www.gem-beta.org/public/MeasureDetail.aspx?mid=1110&cat=2 (accessed August 7, 2014).

Gerber, Y., Y. Benyamini, U. Goldbourt, Y. Drory, and Israel Study Group on First Acute Myocardial Infarction. 2010. Neighborhood socioeconomic context and long-term survival after myocardial infarction. Circulation 121(3):375–383.

Hager, E. R., A. M. Quigg, M. M. Black, S. M. Coleman, T. Heeren, R. Rose-Jacobs, J. T. Cook, S. A. E. de Cuba, P. H. Casey, M. Chilton, D. B. Cutts, A. F. Meyers, and D. A. Frank. 2010. Development and validity of a 2-item screen to identify families at risk for food insecurity. Pediatrics 126(1):e26–e32.

Hall, M. H., K. A. Matthews, H. M. Kravitz, E. B. Gold, D. J. Buysse, J. T. Bromberg, J. F. Owens, and M.-F. Sowens. 2009. Race and financial strain are independent correlates of sleep in midlife women: The SWAN Sleep Study. Sleep 32(1):73–82.

Hermans, H., and H. M. Evenhuis. 2010. Characteristics of instruments screening for depression in adults with intellectual disabilities: Systematic review. Research in Developmental Disabilities 31(6):1109–1120.

HHS (U.S. Department of Health and Human Services). 2008. Plain language: A promising strategy for clearly communciating health information and improving health literacy. Washington, DC: Office of Disease Prevention and Health Promotion, HHS. http://www.health.gov/communication/literacy/plainlanguage/IssueBrief.pdf (accessed May 22, 2014).

Hibbard, J. H., and J. Greene. 2013. What the evidence shows about patient activation: Better health outcomes and care experiences; fewer data on costs. Health Affairs 32(2):207–214.

Hibbard, J. H., J. Stockard, E. R. Mahoney, and M. Tusler. 2004. Development of the Patient Activation Measure (PAM): Conceptualizing and measuring activation in patients and consumers. Health Services Research 39(4p1):1005–1026.

Holt-Lunstad, J., T. B. Smith, and J. B. Layton. 2010. Social relationships and mortality risk: A meta-analytic review. PLoS Medicine 7(7):e1000316. http://www.ncbi.nlm.nih.gov/pubmed/20668659 (accessed April 3, 2014).

House, J. S. 2002. Understanding social factors and inequalities in health: 20th century progress and 21st century prospects. Journal of Health and Social Behavior 43(2):125–142.

House, J. S., K. R. Landis, and D. Umberson. 1988. Social relationships and health. Science 241(4866):540–545.

HRSA (Health Resources and Services Administration). 2012. Women’s preventive services: Required health plan coverage guidelines. Rockville, MD: HRSA.

Hurley, A. D. 2006. Mood disorders in intellectual disability. Current Opinion in Psychiatry 19(5):465–469.

Inigma. 2014. Four levels of health activation. http://www.insigniahealth.com/solutions/patient-activation-measure (accessed July 10, 2014).

Institute for Safe Families. 2013. Findings from the Philadelphia Urban ACE Study. Philadelphia, PA: Institute for Safe Families.

IOM (Institute of Medicine). 2004. Health literacy: A prescription to end confusion. Washington, DC: The National Academies Press.

IOM. 2009. Measures of health literacy: Workshop summary. Washington, DC: The National Academies Press.

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

IOM. 2011a. Clinical preventive services for women: Closing the gaps. Washington, DC: The National Academies Press.

IOM. 2011b. For the public’s health: The role of measurement in action and accountability. Washington, DC: The National Academies Press.

John, O. P. 2007-2009. The Big Five Inventory: Frequently asked questions. http://www.ocf.berkeley.edu/~johnlab/bfi.htm (accessed July 4, 2014).

Kahn, J. R., and L. I. Pearlin. 2006. Financial strain over the life course and health among older adults. Journal of Health and Social Behavior 47(1):17–31.

Kann, L., S. Kinchen, S. L. Shanklin, K. H. Flint, J. Hawkins, W. A. Harris, R. Lowry, E. O’Malley Olsen, T. McManus, D. Chyen, L. Whittle, and E. Taylor. 2014. Youth Risk Behavior Suveillence-United States, 2013. Morbidity and Mortality Weekly Report 63(4):1–47.

Kaplan, G. A., S. A. Everson, and J. K. Lynch. 2000. The contribution of social and behavioral research to an understanding of the distribution of disease: A multilevel approach. Paper commissioned by the Committee on Capitalizing on Social Science and Behavioral Research to Improve the Public’s Health. Washington, DC: National Academy Press.

Kim, D., A. V. Diez-Roux, C. I. Kiefe, I. Kawachi, and K. Liu. 2010. Do neighborhood socioeconomic deprivation and low social cohesion predict coronary calcification?: The CARDIA study. American Journal of Epidemiology 172(3):288–298.

Kim, J., D. Amtmann, K. F. Cook, K. L. Johnson, A. M. Bammer, H. Chung, and R. L. Askew. 2012. Psychometric properties of three depression scales in people with multiple sclerosis. International Journal of MS Care 14(Suppl. 2):85–86.

Knight, J. R., L. Sherritt, S. K. Harris, E. C. Gates, and G. Chang. 2003. Validity of brief alcohol screening tests among adolescents: A comparison of the AUDIT, POSIT, CAGE, and CRAFFT. Alcoholism: Clinical and Experimental Research 27(1):67–73.

Kroenke, K., R. L. Spitzer, and J. B. W. Williams. 2001. The PHQ-9: Validity of a brief depression severity measure. Journal of General Internal Medicine 16(9):606–613.

Kroenke, K., R. L. Spitzer, and J. B. W. Williams. 2003. The Patient Health Questionnaire-2: Validity of a two-item depression screener. Medical Care 41(11):1284–1292.

Kroenke, K., R. L. Spitzer, J. B. Williams, P. O. Monahan, and B. Lowe. 2007. Anxiety disorders in primary care: Prevalence, impairment, comorbidity, and detection. Annals of Internal Medicine 146(5):317–325.

Kushel, M. B., R. Gupta, L. Gee, and J. S. Hass. 2006. Housing instability and food insecurity as barriers to health care among low-income Americans. Journal of General Internal Medicine 21(1):71–77.

Kutner, M., E. Greenberg, Y. Jen, and C. Paulsen. 2006. The health literacy of America’s adults: Results from the 2003 National Assessment of Adult Literacy (NCES 2006-483). U.S. Department of Education. Washington, DC: National Center for Education Statistics.

Lara, M., C. Gamboa, M. I. Kahramanian, L. S. Morales, and D. E. H. Bautista. 2005. Acculturation and Latino health in the United States: A review of the literature and its sociopolitical context. Annual Review of Public Health 26:367–397.

Lee, J. S., and E. A. Frongillo. 2001. Nutritional and health consequences are associated with food insecurity among U.S. elderly persons. Journal of Nutrition 131(5):1503–1509.

Lemola, S., K. Räikkönen, K. A. Matthews, M. F. Scheier, K. Heinonen, A.-K. Pesonen, N. Komsi, and J. Lahti. 2010. A new measure for dispositional optimism and pessimism in young children. European Journal of Personality 24(1):71–84.

Little, P., J. Barnett, B. Margetts, A. L. Kinmonth, J. Gabbay, R. Thompson, D. Warm, H. Warwick, and S. Wooton. 1999. The validity of dietary assessment in general practice. Journal of Epidemiology and Community Health 53(3):165–172.

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

MacArthur Research Network on SES & Health. 2008. Sociodemographic questionnaire. http://www.macses.ucsf.edu/research/socialenviron/sociodemographic.php (accessed July 31, 2014).

Malone, N., K. F. Baluja, J. M. Costanzo, and C. J. Davis. 2003. The foreign-born population: 2000. U.S. Census Bureau. http://www.census.gov/prod/2003pubs/c2kbr-34.pdf (accessed August 8, 2014).

Marin, G., and R. J. Gamba. 1996. A new measurement of acculturation for hispanics: The Bidimensional Acculturation Scale for Hispanics (BAS). Hispanic Journal of Behavioral Sciences 18(3):297–316.

Matthews, K. A., C. I. Kiefe, C. E. Lewis, K. Liu, S. Sidney, and C. Yunis. 2002. Socioeconomic trajectories and incident hypertension in a biracial cohort of young adults. Hypertension 39(3):772–776.

McCaw, B. 2013. Incorporating behavioral health in the EHR to improve care. Presentation to the Committee on the Recommended Social and Behavioral Domains and Measures for Electronic Health Records: Washington, DC: IOM.

MESA (Multi-Ethnic Study of Atherosclerosis). 2005. Multi-Ethnic Study of Atherosclerosis. Exam 4: Personal history self-administered. http://www.mesa-nhlbi.org/PublicDocs/MESAExam4Forms/V4_Personal_History_6-9-2005.pdf (accessed July 22, 2014).

Mitchell, A. J. 2007. Pooled results from 38 analyses of the accuracy of distress thermometer and other ultra-short methods of detecting cancer-related mood disorders. Journal of Clinical Oncology 25(29):4670–4681.

Nagasako, E. M., M. Reidhead, B. Waterman, and W. C. Dunagan. 2014. Adding socioeconomic data to hospital readmissions calculations may produce more useful results. Health Affairs 33(5):786–791.

NC (North Carolina) Program on Health Literacy. 2014. Literacy assessment instruments. http://www.nchealthliteracy.org/instruments.html (accessed August 28, 2014).

NCI (National Cancer Institute). 2012. NCI measures guide for youth tobacco research: Nicotine Dependence Scale for Adolescents (NDSA). http://cancercontrol.cancer.gov/brp/tcrb/nicotine_depend_scale_adol.html (accessed July 30, 2014).

NCI. 2013. Risk Factor Monitoring and Methods Branch Applied Research Program. Tobacco use supplement to the current population survey. Bethesda, MD: National Institutes of Health. http://appliedresearch.cancer.gov/tus-cps/TUS-CPS_fact_sheet.pdf (accessed July 30, 2014).

Nelson, H. D., C. Bougatsos, and I. Blazian. 2012. Screening women for intimate partner violence: A systematic review to update the U.S. Preventive Services Task Force recommendation. Annals of Internal Medicine 156(11):796–808.

NIH (National Institutes of Health). 2006–2012. NIH Toolbox Emotion: NIH TB Self-Efficacy CAT Age 18+. http://www.nihtoolbox.org/WhatAndWhy/Emotion/Emotional%20Health%20PDF/NIH%20TB%20SSE%20Self-Efficacy%20CAT%2018+.pdf (accessed July 17, 2014).

NIH. 2011. Identifying core behavioral and psychosocial data elements for the electronic health record Bethesda, MD: National Institutes of Health, Office of Behavioral and Social Sciences Research.

NIOSH (National Institute for Occupational Safety and Health). 2014. Comments from the National Institute for Occupational Safety and Health for the IOM Committee on the Recommended Social and Behavioral Domains and Measures for Electronic Health Records. Washington, DC: IOM.

Oates, D. J., and M. K. Paasche-Orlow. 2009. Health literacy: Communication strategies to improve patient comprehension of cardiovascular health. Circulation 119(7):1049–1051.

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

Olver, M. E., T. P. Nicholaichuk, D. A. Kingston, and S. C. Wong. 2014. A multisite examination of sexual violence risk and therapeutic change. Journal of Consulting and Clinical Psychology 82(2):312–324.

OMB (Office of Management and Budget). 2003. Revisions to the standards for the classification of federal data on race and ethnicity. http://www.whitehouse.gov/omb/fedreg_1997standards (accessed July 16, 2014).

OMH (Office of Minority Health). 2010. OMB standards for data on race and ethnicity. http://minorityhealth.hhs.gov/templates/browse.aspx?lvl=2&lvlID=172 (accessed July 15, 2015).

Paasche-Orlow, M., and M. S. Wolf. 2007. The causal pathways linking health literacy to health outcomes. American Journal of Health Behavior 31(Suppl. 1):s19–s26.

Pantell, M., D. Rehkopf, D. Jutte, S. L. Syme, J. Balmes, and N. Adler. 2013. Social isolation: A predictor of mortality comparable to traditional clinical risk factors. American Journal of Public Health 103(11):2056–2062.

Philadelphia Department of Public Health. 2013. Walkable access to healthy food in Philadelphia, 2010–2012. Philadelphia, PA: Philadelphia Department of Public Health.

Physical Activity Guidelines Advisory Committee. 2008. Physical Activity Guidelines Advisory Committee Report, 2008. Washington, DC: HHS.

Pilkonis, P. A., S. W. Choi, S. P. Reise, A. M. Stover, W. T. Riley, D. Cella. 2011. Item bansk for measuring emotional distress from the Patient Reported Outcomes Meseaurment Information System (PROMIS): Depression, anxiety, and anger. Assessment 18(3):263–283.

Piper, M. E., T. M. Piasecki, E. B. Federman, D. M. Bolt, S. S. Smith, M. C. Fiore, and T. B. Baker. 2004. The Wisconsin Inventory of Smoking Dependence Motives (WISDM-68). Journal of Consulting and Clinical Psychology 72(2):139–154.

Piper, M. E., D. E. McCarthy, and T. B. Baker. 2006. Assessing tobacco dependence: A guide to measure evaluation and selection. Nicotine & Tobacco Research 8(3):339–351.

Piper, M. E., D. E. McCarthy, D. M. Bolt, S. S. Smith, C. Lerman, N. Benowitz, M. C. Fiore, and T. B. Baker. 2008. Assessing dimensions of nicotine dependence: An evaluation of the Nicotine Dependence Syndrome Scale (NDSS) and the Wisconsin Inventory of Smoking Dependence Motives (WISDM). Nicotine & Tobacco Research 10(6):1009–1020.

Polivy, J. 1996. Psychological consequences of food restriction. Journal of the American Dietetic Association 96(6):589–592.

Pollack, C. E., M. E. Slaughter, B. A. Griffin, T. Dubowitz, and C. E. Bird. 2012. Neighborhood socioeconomic status and coronary heart disease risk prediction in a nationally representative sample. Public Health 126(10):827–835.

PRB (Population Reference Bureau). 2009. The 2010 Census questionnaire: Seven questions for everyone (accessed July 15, 2014).

PROMIS (Patient Recorded Outcomes Measurement Information System). 2014a. Anxiety: A brief guide to the PROMIS anxiety instruments. Silver Spring, MD: National Institutes of Health, PROMIS. https://www.assessmentcenter.net/documents/PROMIS%20Anxiety%20Scoring%20Manual.pdf (accessed June 7, 2014).

PROMIS. 2014b. Depression: A brief guide to the PROMIS depression instruments. Silver Spring, MD: National Institutes of Health, PROMIS. https://www.assessmentcenter.net/documents/PROMIS%20Depression%20Scoring%20Manual.pdf (accessed July 28, 2014).

Puterman, E., J. Haritatos, N. E. Adler, S. Sidney, J. E. Schwartz, and E. S. Epel. 2013. Indirect effect of financial strain on daily cortisol output through daily negative to positive affect index in the coronary artery risk development in young adults study. Psychoneuroendocrinology 38(12):2883–2889.

Rabin, R. F., J. M. Jennings, J. C. Campbell, and M. H. Bair-Merritt. 2009. Intimate partner violence screening tools. American Journal of Preventive Medicine 36(5):439–445.

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

Rammstedt, B., and O. P. John. 2007. Measuring personality in one minute or less: A 10-item short version of the Big Five Inventroy in English and German. Journal of Research in Personality 41:203–212.

Roth, A. J., A. B. Kornblith, L. Batel-Copel, E. Peabody, H. I. Scher, and J. C. Holland. 1998. Rapid screeening for psychologic distress in men with prostate carcinoma. American Cancer Society 82(10):1904–1908.

Rudd, R. E. 2010. Improving American’s health literacy. New England Journal of Medicine 363(24):2283–2285.

Rushton, G., M. P. Armstrong, J. Gittler, B. R. Greene, C. E. Pavlik, M. M. West, and D. L. Zimmerman, eds. (2007). Geocoding health data: The use of geographic codes in cancer prevention and control, research, and practice. Boca Raton, FL: CRC Press, Taylor & Francis Group.

Ryan, H., A. Trosclair, and J. Gfroerer. 2012. Adult current smoking: Differences in definitions and prevalence estimates—NHIS and NSDUH, 2008. Journal of Environmental and Public Health. doi:10.1155/2012/918368. http://www.hindawi.com/journals/jeph/2012/918368/cta (accessed July 30, 2014).

Sanders, L. M., S. Federico, P. Class, M. A. Abrams, and B. Dreyer. 2009. Literacy and child health: A systematic review. JAMA Pediatrics 163(2):131–140.

Schalet, B. D., K. F. Cook, S. W. Choi, and D. Cella. 2013. Establishing a common metric for self-reported anxiety: Linking the MASQ, PANAS, and GAD-7 to PROMIS Anxiety. Journal of Anxiety Disorders 28:88–96.

Scheier, M. F., C. S. Carver, and M. W. Bridges. 1994. Distinguishing optimism from neuroticism (and trait anxiety, self mastery, and self esteem): A re-evaluation of the life orientation test. Journal of Personality and Social Psychology 67:1063–1078.

Schillinger, D., K. Grumbach, J. Piette, F. Wang, D. Osmond, C. Daher, J. Palacios, G. D. Sullivan, and A. B. Bindman. 2002. Association of health literacy with diabetes outcomes. JAMA 288(4):475–482.

Schwarzer, R., and B. Renner. 2000. Social-cognitive predictors of health behavior: Action self-efficacy and coping self-efficacy. Health Psychology 19(5):487–495.

Scott, R. I., and C. A. Wehler. 1998. Food insecurity/food insufficiency: An impirical examination of alternative measures of food problems in impoverished U.S. households, Discussion paper no. 1176-98. Madison, WI: Institute for Research on Poverty.

Seeman, T. E., B. S. McEwen, J. W. Rowe, and B. H. Singer. 2001. Allostatic load as a marker of cumulative biological risk: MacArthur studies of successful aging. Proceedings of the National Academy of Sciences of the United States of America 98(8):4770–4775.

Seligman, H. K., A. B. Bindman, E. Vittinghoff, and A. M. Kanaya. 2007. Food insecurity is associated with diabetes mellitus: Results from the National Health Examination and Nutrition Examination Survey (NHANES) 1999-2002. Journal of General Internal Medicine 22:1018–1023.

Shiffman, S., A. J. Waters, and M. Hickcox. 2004. The nicotine dependence syndrome scale: A multidimensional measure of nicotine dependence. Nicotine & Tobacco Research 6(2):327–348.

Shonkoff, J. P., A. S. Garner, the Committee on Psychosocial Aspects of Child and Family, Health. Committee on Early Childhood, Adoption, and Dependent Care, and Section Developmental and Behavioral Pediatrics. 2012. The lifelong effects of early childhood adversity and toxic stress. Pediatrics 129(1):e232–e246.

Siefert, K., C. M. Heflin, M. E. Corcoran, and D. R. Williams. 2001. Food insufficiency and the physcial and mental health of low-income women. Women & Health 32(1–2):159–177.

Silverman, J. G., A. Raj, and K. Clements. 2004. Dating violence and associated sexual risk and pregnancy among adolescent girls in the United States. Pediatrics 114(2):e220–225.

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

Singh, G. K., and B. A. Miller. 2004. Health, life expectancy, and mortality patterns among immigrant populations in the United States. Canadian Journal of Public Health 95(3):14–21.

Slotkin, J., M. Kallen, J. Griffith, S. Magasi, J. Salsman, C. Nowinski, and R. Gershon. 2012. NIH Toolbox technical manual. Domain: Emotion. Subdomain: Stress and self-efficacy. Bethesda, MD: National Institutes of Health. http://www.nihtoolbox.org/HowDoI/TechnicalManual/Technical%20Manual%20sections/Toolbox%20Perceived%20Stress%20Survey%20Technical%20Manual.pdf (accessed August 6, 2014).

SMART (Sexual Minority Assessment Research Team). 2009. Best practices for asking questions about sexual orientation on surveys. Los Angeles, CA: UCLA, School of Law, The Williams Institute.

Smith, I. E., J. S. Lancaster, S. Moss-Wells, C. D. Coles, and A. Falek. 1987. Identifying high-risk pregnant drinkers: Biological and behavior correlates of continuous heavy drinking during pregnancy. Journal of Studies on Alcohol and Drugs 48(4):304–309.

Sohal, H., S. Eldridge, and G. Feder. 2007. The sensitivity and specificity of four questions (HARK) to identify intimate partner violence: A diagnostic accuracy study in general practice. BMC Family Practice 8(1):49–58.

Spitzer, R. L., K. Kreonke, J. B. W. Williams, and B. Lowe. 2006. A brief measure for assessing generalized anxiety disorder. The GAD-7. JAMA 166(10):1092–1097.

Steptoe, A., A. Shanker, P. Demakakos, and J. Wardle. 2013. Social isolation, loneliness, and all-cause mortality in older men and women. Proceedings of the National Academy of Sciences of the United States of America 110(15):5797–5801.

Stevens, S. S. 1946. On the theory of scales of measurement. Science 103(2684):677-680.

Storr, C. L., B. A. Reboussin, and J. C. Anthony. 2005. The Fagerström test for nicotine dependence: A comparison of standard scoring and latent class analysis approaches. Drug and Alcohol Dependence 80(2):241–250.

Szanton, S., R. J. Thorpe, and K. E. Whitfield. 2010. Life-course financial strain and health in african-Americans. Social Science & Medicine 71(2):259–265.

Taj, N., A. Devera-Sales, and D. C. Vinson. 1998. Screening for problem drinking: Does a single question work? Journal of Family Practice 46(4):328–335.

Taren, D. L., W. Clark, M. Chernesky, and E. Quirk. 1990. Weekly food servings and participation in social programs among low income families. American Journal of Public Health 80(11):1376–1378.

Townsend, M. S., J. Peerson, B. Love, C. Achterberg, and S. P. Murphy. 2001. Food insecurity is positively related to overweight in women. Journal of Nutrition 131(6):1738–1745.

TRB (Transportation Research Board) and IOM (Institute of Medicine). 2005. Does the built environment influence physical activity?: Examining the evidence. Washington, DC: The National Academies Press.

Uchino, B. N. 2009. Understanding the links between social support and physical health: A life-span perspective with emphasis on the separability of perceived and received support. Perspectives on Psychological Science 4(3):236-255.

University of Miami Department of Psychology. 2007. LOT-R (Life Orientation Test-Revised). http://www.psy.miami.edu/faculty/ccarver/sclLOT-R.html (accessed August 13, 2014).

U.S. Census Bureau. 2014a. American Community Survey. http://www.census.gov/acs/www/about_the_survey/american_community_survey (accessed July 31, 2014).

U.S. Census Bureau. 2014b. Index of questions. http://www.census.gov/history/www/through_the_decades/index_of_questions (accessed July 15, 2014).

USDA (U.S. Department of Agriculture). 2013. U.S. Fruit and vegetable consumption: Who, what, where, and how much. Washington, DC: U.S. Department of Agriculture.

USDA. 2014. Food security in the U.S.: History & background. http://www.ers.usda.gov/topics/food-nutrition-assistance/food-security-in-the-us/history-background.aspx (accessed July 22, 2014).

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×

USPSTF (U.S. Preventive Services Task Force). 2012. Screening women for intimate partner violence and elderly and vulnerable adults for abuse: Systematic review to update the 2004 U.S. Preventive Services Task Force Recommendation. Rockville, MD: Agency for Healthcare Research and Quality.

USPSTF. 2013. Screening for intimate partner violence and abuse of elderly and vulnerable adults. U.S. Preventive Services Task Force recommendation statement. http://www.uspreventiveservicestaskforce.org/uspstf12/ipvelder/ipvelderfinalrs.htm (accessed July 30, 2014).

Volandes, A. E., and M. K. Paasche-Orlow. 2007. Health literacy, health inequality and a just healthcare system. American Journal of Bioethics 7(11):5–10.

Vortuba, M. E., and J. R. Kling. 2008. Effects of neighborhood characteristics on the mortality of Black male youth: Evidence from Gautreaux. Ann Arbor, MI: National Poverty Center.

Wallace, L. S., E. S. Rogers, S. E. Rosko, D. B. Holiday, and B. D. Weiss. 2006. Screening items to identify patients with limited health literacy skills. Journal of General Internal Medicine 21(8):874–877.

Wallace, S., D. I. Padilla-Frausto, and S. Smith. 2013. Economic need among older Latinos: Applying the elder Economic Security Standard™ Index. Journal of Cross-Cultural Gerontology 28(3):239–250.

Wardle, J., K. Parmenter, and J. Waller. 2000. Nutrition knowledge and food intake. Appetite 34(3):269–275.

Wardle, J., S. Carnell, and L. Cooke. 2005. Parental control over feeding and children’s fruit and vegetable intake: How are they related? Journal of the American Dietetic Association 105(2):227–232.

WHO (World Health Organization). 2011. Air quality and health. http://www.who.int/mediacentre/factsheets/fs313/en/index.html (accessed January 15, 2014).

WHO. 2013. Responding to intimate partner violence and sexual violence against women: WHO clinical and policy guidelines. Geneva, Switzerland: WHO.

Woolf, S. H., and P. Braveman. 2011. Where health disparities begin: The role of social and economic determinants—and why current policies may make matters worse. Health Affairs 30(10):1852–1859.

Woolf, S. H., R. E. Johnson, R. L. Phillips, and M. Philipsen. 2007. Giving everyone the health of the educated: An examination of whether social change would save more lives than medical advances. American Journal of Public Health 97(4):679–683.

WOW (Wider Opportunities for Women). 2014. The economic security database. http://www.basiceconomicsecurity.org/more-info.aspx (accessed July 22, 2014).

Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 127
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 128
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 129
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 130
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 131
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 132
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 133
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 134
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 135
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 136
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 137
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 138
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 139
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 140
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 141
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 142
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 143
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 144
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 145
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 146
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 147
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 148
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 149
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 150
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 151
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 152
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 153
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 154
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 155
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 156
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 157
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 158
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 159
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 160
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 161
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 162
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 163
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 164
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 165
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 166
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 167
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 168
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 169
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 170
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 171
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 172
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 173
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 174
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 175
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 176
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 177
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 178
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 179
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 180
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 181
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 182
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 183
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 184
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 185
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 186
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 187
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 188
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 189
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 190
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 191
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 192
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 193
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 194
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 195
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 196
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 197
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 198
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 199
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 200
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 201
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 202
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 203
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 204
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 205
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 206
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 207
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 208
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 209
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 210
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 211
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 212
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 213
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 214
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 215
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 216
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 217
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 218
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 219
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 220
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 221
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 222
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 223
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 224
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 225
Suggested Citation:"4 Measures Reviewed for Each Candidate Domain." Institute of Medicine. 2014. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. Washington, DC: The National Academies Press. doi: 10.17226/18951.
×
Page 226
Next: 5 Recommended Core Domains and Measures »
Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2 Get This Book
×
Buy Paperback | $63.00 Buy Ebook | $49.99
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

Determinants of health - like physical activity levels and living conditions - have traditionally been the concern of public health and have not been linked closely to clinical practice. However, if standardized social and behavioral data can be incorporated into patient electronic health records (EHRs), those data can provide crucial information about factors that influence health and the effectiveness of treatment. Such information is useful for diagnosis, treatment choices, policy, health care system design, and innovations to improve health outcomes and reduce health care costs.

Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2 identifies domains and measures that capture the social determinants of health to inform the development of recommendations for the meaningful use of EHRs. This report is the second part of a two-part study. The Phase 1 report identified 17 domains for inclusion in EHRs. This report pinpoints 12 measures related to 11 of the initial domains and considers the implications of incorporating them into all EHRs. This book includes three chapters from the Phase 1 report in addition to the new Phase 2 material.

Standardized use of EHRs that include social and behavioral domains could provide better patient care, improve population health, and enable more informative research. The recommendations of Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2 will provide valuable information on which to base problem identification, clinical diagnoses, patient treatment, outcomes assessment, and population health measurement.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    Switch between the Original Pages, where you can read the report as it appeared in print, and Text Pages for the web version, where you can highlight and search the text.

    « Back Next »
  6. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  7. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  8. ×

    View our suggested citation for this chapter.

    « Back Next »
  9. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!