Substantial empirical evidence of the contribution of social and behavioral factors to functional status and the onset and progression of disease has accumulated over the past few decades. Research on social and behavioral determinants of health was inspired to a substantial extent by three landmark papers. The analysis by McGinnis and Foege (1993) of the “actual causes of death” showed the large contribution of behaviors such as smoking, diet and activity, and alcohol as well as socioeconomic status to premature morality. Link and Phelan (1995) argued that social conditions related to socioeconomic resources such as money, social ties, and knowledge are “fundamental causes” of disease. Further, the Whitehall Study of British civil servants (Marmot et al., 1984) demonstrated significant decreases in rates of mortality at each step up in “occupational grade” despite the fact that all of those followed had access to health care. Taken together, these papers provided a compelling argument for examination of the role of social and behavioral factors in the determination of health.
Much of the subsequent research on social conditions and their associated behavioral risks have been aimed at the elimination of avoidable and unjust differences in morbidity and mortality among sociodemographic groups. However, efforts to address health disparities among groups in the United States are not the only reason to consider social and behavioral determinants of health. In the past few years, the relatively poor health status of the U.S. population as a whole relative to that of the populations
of other countries has fostered interest in understanding the reasons for this situation. The nation lags in life expectancy, maternal mortality, and infant mortality; and in the conditions that contribute to these outcomes, including injuries and homicides, sexually transmitted diseases, adolescent pregnancy, heart disease, obesity, diabetes, disability, chronic lung disease, HIV/AIDS, and drug-related mortality (NRC and IOM, 2013). Bradley and Taylor (2013) characterized the fact that the United States has higher rates of morbidity and mortality (CIA, 2011; OECD, 2011a; United Nations, 2009) than other countries of the Organisation for Economic Co-operation and Development (OECD)—even though it spends more on health care than those nations (OECD, 2011b) both in absolute terms and as a portion of the country’s gross domestic product—as the American health care paradox.
This unfavorable balance between health care costs and the health of the U.S. population suggests that the way in which the country allocates spending for health care is suboptimal. Some of this may reflect waste and inefficiency in the delivery of health care (Berwick and Hackbarth, 2012; IOM, 2010). However, it may also reflect insufficient attention by the current health care system to the major determinants of health and illness. Increasing evidence indicates that the life conditions outside the encounters with the health care system matter far more than the condition for which a patient seeks care in the exam room. The best available estimates suggest that the conditions for which patients seek medical care (accounting both for access to care and the quality of the care that is received) account for only about 10 percent of early deaths, whereas health behaviors and social conditions are estimated to account for more than half of such deaths (McGinnis et al., 2002). In contrast to the OECD countries that Bradley et al. (2011) studied, the United States allocates relatively more of its resources to health care and relatively less on social services. Across all countries, those that had the highest ratio of spending on social services to spending on health care had the best population health statistics.
Traditionally, research and interventions on the social and behavioral determinants of health have largely been the purview of public health, which has focused on prevention of disease and the maintenance of the public’s health. Public health researchers and practitioners have long believed that improving the health status of Americans requires addressing the social determinants of health, which are defined as “circumstances in which people are born, grow up, live, work, and age, as well as the health systems they utilize” (CDC, 2013). The goals set for the U.S. population in Healthy People 2020, which include improving health status and eliminating disparities, are explicit about the need to address social and physical environments of populations to promote good health and ensure healthy development and behaviors across the life course. Health care systems, in contrast, have pri-
marily focused on the treatment of disease in individual patients, and until recently, social determinants of health have not been linked in the United States to clinical practice or health care delivery systems. Conversely, several OECD countries incorporate social and behavioral information in their provision of health care and as part of their electronic health records (EHRs) (OECD, 2013). The United Kingdom, for example, collects information on depression, anxiety, alcohol and tobacco use, as well as physical activity levels (McIntosh et al., 2004; NICE, 2009, 2010, 2013a,b). Countries that are longtime users of EHRs, such as Denmark, New Zealand, and Sweden, have benefited from the interoperable use of patient data. General practitioners and hospitals are able to access patient information, such as physician notes, examinations, prescribed medications, across the health system; and health care facilities are able to plan across primary, secondary, and long-term care settings (Gray et al., 2011).
In recent years, changes have begun in the United States, prompted, in part, by concern about the unsustainability of the growth of health care costs and poor overall public health statistics (NCHS, 2006). This is best exemplified by the conceptualization of the “triple aim” by Berwick et al. (2008). They posit that improvements to health outcomes in the United States require the simultaneous pursuit of improvements to the experience of health care, improvements to the health of populations, and reductions in the per capita costs of health care. These are not independent goals but rely upon each other in the pursuit of achieving high-value health care. The nation’s response to the triple aim has resulted in the creation of the National Strategy for Quality Improvement in Health Care that aims to improve the quality of health and health care by aligning public and private interests, in turn, having all parts of the health system working together toward a common goal of improved health for all Americans (HHS, no date).
Changes in policy affecting incentives for new approaches to health care delivery included in the Patient Protection and Affordable Care Act,1 and other policy innovations are encouraging the formation of more coordinated systems that have a greater capacity to address the social and behavioral needs of individual patients and to pay more attention to public health (HHS, no date). Accountable care organizations (ACOs)—groups of doctors, hospitals, and other health care providers, who provide coordinated care to patients—and other group practices are incentivized to maintain the health of the populations that they serve and reduce health care utilization (PwC, 2010). To the extent that the provision of better services and interventions meet their patients’ social needs, and to eliminate behavioral risk
1 Public Law 111-148.
and reduce the use of health care services, these systems will want to assess the social and behavioral determinants of health.
EHRs hold the potential to serve as essential tools for improving quality, increasing efficiency, and expanding access to the health system (Friedman, 2006; Friedman et al., 2010). They provide crucial information to providers treating individual patients, to health systems about population health, and to researchers about the determinants of health and the effectiveness of treatment. The inclusion of social and behavioral domains in EHRs is vital to quality, efficiency, and access.
There are inherent risks to collecting personal data in an electronic format. Safeguards have been enacted to counteract potential harms. Health information is protected by a federal law, known as the Health Insurance Portability and Accountability Act of 1996 (HIPAA),2 which restricts what health care professionals can reveal about their patients’ medical status. Given the seriousness of breaches of confidentiality and the extent to which these can undermine the value of EHRs, electronic information must be well-protected in a vigorous manner. Further, for EHRs to achieve their full potential, data will need to be collected consistently across the nation. This requires a commitment from all components in a health system—including the patient’s interest and willingness to provide data, some of which might be considered to be sensitive information to the individual.
The patient health record, which traces its origin to the Mayo Clinic (Melton, 1996), the Presbyterian Hospital (Lamb, 1955; Openchowski, 1925), and the Flexner Report a century ago (Flexner, 1910), serves “to recall observations, to inform others, to instruct students, to gain knowledge, to monitor performance, and to justify interventions” (Reiser, 1991, p. 902). Early adopters of electronic health data began writing programs to store and retrieve patient records in 1958 (Stead, 1989). By 1991, the Institute of Medicine (IOM) identified the computer-based record as an essential technology for health care (NRC, 1997). Growth of interest in the EHR has paralleled growth in other types of electronic technologies, including mobile communications, online social networks, and sensors.
At many institutions today, the legal health record—which is defined by federal and state regulations—is actually a combination of electronic systems and paper sources. The term EHR loosely refers to the electronic
2 Public Law 104-191, 110 Statute 1936.
version of the patient health record, but the term is ambiguous. The “EHR system” comprises both the database that holds the patient information and the software tools used to collect, store, and manage the information, along with the tools needed to support decision making and analyze data (McDonald et al., 2014). Therefore, in practice, the EHR refers to those portions of the patient health record that happen to be stored in a particular EHR system. For example, institutions with two EHR systems (e.g., one for inpatient care and one for outpatient care) may split their legal health record into two EHRs. Furthermore, an EHR system is often referred to simply as an “EHR.” The term EHR data is sometimes used to be clear that the concept refers specifically to the information rather than to the whole system (McDonald et al., 2014). Figure 1-1 illustrates the components of an EHR system.
An EHR system’s decision-making tools include data-driven alerts and reminders, order sets, displays to visualize information, calculators, list
FIGURE 1-1 The legal patient record may comprise electronic and paper information from several sources. In the simplest case, a health provider may be served by a single electronic health record system (EHRS), whose database constitutes the entire legal patient record. Some organizations have more than one EHRS. Ancillary systems such as the clinical laboratory and registration systems have their own databases, which may be considered separate from the legal patient record or may be considered part of it; in addition, they usually upload information to the EHRS’s database. There may also be links to outside sources of information, which may upload information or remain purely as a link.
managers, search tools, data validations, and links to knowledge resources (McDonald et al., in press). These tools provide the opportunity to improve decisions and to reduce errors. In the context of social determinants of health, they enable the clinician to efficiently capture the determinants, keep track of them, and apply them at the point of care, incorporating evidence-based practices drawn from recent literature. When it is set up, the EHR system can steer health care practice to use social and behavioral determinants extensively and appropriately to improve health care outcomes (HealthIt.gov, no date–a,b).
EHR systems have, unfortunately, not yet achieved their potential. As of 2009 only 4 to 16 percent of clinicians and hospitals were found to be using EHRs (Blumenthal, 2009; Blumenthal and Tavenner, 2010), with few using truly comprehensive systems. Recent scientific reports include examples of unintentional and adverse clinical consequences in health care settings using EHR systems (Han et al., 2005) because of the discrepancy between health care work and information system design or implementation (Rosenbloom et al., 2006). Still, numerous studies indicate positive results in using EHRs, as is shown in the following sections.
Efforts to recover from the 2008 financial crisis provided an opportunity for improvement. The American Recovery and Reinvestment Act of 2009 (ARRA)3 included the Health Information Technology for Economic and Clinical Health Act (HITECH) provision, which provided billions of dollars in incentives to use EHR systems to create “significant and measurable improvements” in population health outcomes through a transformed health care delivery system. HITECH required that a certified EHR system be used in a meaningful manner with the electronic exchange of health information and reporting of quality measures. Since 2009, recent published estimates (2012) indicate that 40 percent of office-based physicians have adopted an EHR and 44 percent of hospitals reported having a basic EHR system (RWJF et al., 2013).4
The inclusion of information on social and behavioral determinants of health in EHRs could direct clinical utility in cases in which knowledge of the condition is relevant to diagnosis, treatment, or prognosis. The data in EHRs are useful tools for health care providers, including hospitals and
3 Public Law 111-115.
4 Since publication of the Phase 1 report, new data on use of EHRs point to expanded use. For example, results from the National Ambulatory Medical Care Survey, released in May 2014, revealed that in 2012 nearly 72 percent of office-based physicians used some type of EHR system. This is a significant increase from approximately 35 percent in 2007 (Hsiao et al., 2014).
health care centers, so they may track patient health and illnesses, medical procedures and prognosis, family histories, and laboratory results. Further, EHRs enable computer-based decision support during order entry and prescribing medication. In a study whose findings were published in the New England Journal of Medicine, people with diabetes seen by doctors who used EHRs were 35 percent more likely to get all of the recommended screening measures, such as eye exams and blood sugar tests, than patients whose doctors relied on paper records. Moreover, they were 15 percent more likely to have favorable outcomes on those measures (Cebul et al., 2011). Health networks that use common data platforms are also able to share information across health care providers to coordinate patient services. This sharing of patient data allows the health system to efficiently and effectively provide patient care. (See, for example, Box 1-1.) Networks can also use those data to set reminders on when a patient is due for preventive screenings and alerts on contraindications on medications, among other more administrative functions. While numerous challenges exist, the resulting improvement in care coordination, case management, and health care quality this enables will benefit the primary stakeholder—the patient.
Patients, like their health care providers, can use the data in their EHRs to inform themselves and become more involved in their medical care. Patient empowerment plays an integral role in improving quality of care. An informed and actively involved patient can be more engaged in disease self-management and is better able to adhere to the recommendations of his or her health care provider recommendations. Patients who have access to personal health data can obtain their laboratory results; receive drug and appointment alerts; record their nonprescribed medicines and treatments; monitor and track their illness, treatment, and progress; and learn about the prognosis for their illness (Pagliari et al., 2007), potentially resulting in improved quality of care. (See, for example, Box 1-2.)
Electronic health data provide valuable information on “the distribution of disease, function, and well-being within a population” (Friedman et al., 2013, p. 1560). Perhaps the most common use of EHRs for managing population health is the development of registries that help manage chronic disease and promote prevention. EHRs may provide additional information needed to create a comprehensive public health surveillance system by complementing the data available from existing administrative sources such as the Centers for Medicare & Medicaid Services (CMS) and the Veterans Health Administration (Elliott et al., 2012).
Although many ACOs take a conventional medical approach in viewing their role in managing population health in relation to their panels of
The Case of Veronica: Including Community Health Workers,
Advocacy Groups, and Citizens to Promote Healthy Neighbors
“Veronica,” a patient of Dr. Rishi Manchanda from South Central Los Angeles, had previously sought care at an emergency department (ED) for recurrent and worsening headaches, accompanied by fatigue and malaise. She was given medication for pain and told to return if she did not get better. She returned twice, still in pain. Subsequent workups included a computed tomography (CT) scan, routine blood tests, and a lumbar puncture but revealed nothing clinically wrong. Each of these three ED visits cost more than Veronica’s monthly rent. Veronica’s headaches persisted; she took more sick days from work and she worried about losing her job and about adequately caring for her young children.
When Veronica came to his clinic, Dr. Manchanda and his colleagues probed further into Veronica’s symptoms. The clinic’s routine intake process includes the collection of social data on housing. When asked about her living conditions, Veronica revealed that her apartment was damp, infested by roaches, and full of mold. She could not afford to move and the landlord would not repair the leaky plumbing of her small, ground-floor apartment. The diagnosis, Dr. Manchanda thought, was migraine headache triggered by chronic allergies and complicated by sinus congestion. Allergens in the damp apartment also probably accounted for her son’s frightening asthma flares, another source of anxiety for Veronica.
The medical staff connected Veronica to a community health worker, who could visit her at home and help her obtain and take the medications she needed to relieve her symptoms. At the same time, she was linked to a tenants’ rights advocacy group that petitioned the landlord—this time with a doctor’s note in hand—to make the improvements that were in keeping with building codes that were part of his contractual agreements and were in keeping with local building codes. Veronica and her son got better. Veronica had no further ED visits and her needs were fully met in a nearby “patient-centered home” clinic (Manchanda, 2013).
patients, others are defining population health as the health of individuals in a geopolitical unit (Hacker and Walker, 2013). Even though both types of ACOs would benefit by incorporating and addressing social and behavioral determinants of health, those with the latter perspective are more likely to incorporate a broader view of the determinants of health including social services, public health, and environmental factors (Noble and Casalino, 2013). An ACO can perhaps best manage community health using data systems that merge clinical data obtained from medical encounters and stored in EHRs with community data obtained from a variety of sources and stored in community information systems. A community information system provides compositional and contextual information about the environments where individuals reside, work, and learn. (See, for example,
Box 1-3.) Knowledge of the distribution of community resources and environmental factors that can affect the risk of disease may well become just as important for managing patients’ health as knowledge of clinical indicators such as body mass index.
Primary care specialties in the United States have largely endorsed the patient-centered medical home model, which combines the transformation of primary care practice with payment reform to incentivize the core elements of the model. One of the key functions of a patient-centered medical home is the coordination of patient care by helping patients access community resources, facilitating referrals, linking patients to health care and social services, and ensuring the effective transfer of information (Arend et al., 2012; Stange et al., 2010; Wagner et al., 2012).
Integrating social and behavioral determinants of health into EHRs could allow providers and public health agencies to better describe and monitor patterns of heath and outcomes of care for the entire population (Friedman et al., 2013; HealthIT.gov, no date–b). Capturing social determinants of health in EHR data will allow health care providers to better characterize, understand the causes of, and identify appropriate interven-
The Case of Sonia: Kaiser Permanente in Northern
California’s Domestic Violence Program
“Sonia” is a 38-year-old Mexican-American woman who has been married for 20 years, and the mother of two grown children. She has been a long-term hospital employee who had recently been promoted to a supervisory position. At a routine checkup, when the physician asked how things were at home, Sonia shrugged and looked away. A gentle request, “Tell me more,” led her to reveal that although she had been separated from her husband for 10 years, he continually terrorized her. She was humiliated that the neighbors had called the police because of his angry shouting. Recently, he had threatened to firebomb her home. When the physician offered a referral to a domestic violence evaluator, Sonia accepted the referral and subsequently joined a support group that she credits for “helping me find a path out of the relationship.” She gained confidence to call the police for help, to contact a lawyer, who obtained a restraining order, and then to file for divorce.
Sonia’s abusive situation was detected during routine screening for interpersonal violence (IPV). EHR tools such as prompts to screen for IPV, care paths, charting and documentation, and an easily accessible referral protocol facilitate the provision of a caring, effecting, and efficient response to IPV by health care professionals. However, EHR prompts and tools are best paired with appropriate training in order to successfully identify cues, including nonverbal responses (McCaw et al., 2002).
The Case of Benjamin:
Sharing EHR Records to Address Health
“Benjamin,” a 9-month-old, was hospitalized for difficulty breathing at Cincinnati’s Children Hospital and Medical Center. He suffered from respiratory problems, as well as chronic asthma. A resident caring for Benjamin learned that the family had recently filed a complaint with the health department due to mold in their apartment. Rather than make the necessary repairs, the landlord filed to evict Benjamin’s family for their complaints. Once this health linkage was discovered, Benjamin was referred to the Cincinnati Child Health Law Partnership (Child HeLP).
The partnership between Cincinnati Children’s Hospital and Medical Center and the Legal Aid Society of Greater Cincinnati allows the sharing of information through the patient’s electronic health records (EHRs). Once a physician or social worker enters the referral, it is automatically transferred to Child HeLP. Information is seamlessly transferred between physicians and Legal Aid through EHRs, allowing the patient or the patient’s family to be well-informed throughout the process.
The Legal Aid Society was able to intervene and stop the family’s eviction, and also helped Benjamin’s family look for new, safer housing. The family was able to move into a new home where Benjamin is no longer exposed to asthma triggers such as mold (Cincinnati Children’s Hospital, 2012).
tions that health systems (and non–health care systems) can make to reduce health disparities (HealthIT.gov, no date–c; ONC, 2013), which will allow critical social problems and also costly problems for the health system and society as a whole to be addressed. The addition of these variables has great potential to improve the quality, safety, and efficiency of health services delivery and to support national goals of improving health and eliminating health disparities.
The capture of a core set of standard social and behavioral determinants of health as variables in the EHR advances data harmonization and has the potential to unleash unprecedented opportunities for health research. For example, EHRs can be used to evaluate practice variations and their associations with health outcomes, which in turn will result in improved patient care. Conventional clinical trials, pragmatic clinical trials, clinical epidemiology, and health services research will benefit from enhanced electronic datasets. EHRs can also enable the conduct of registry-based randomized clinical trials (RRCTs), a new form of clinical research
trial that takes advantage of computerized patient registries (Lauer and D’Agostino, 2013). These trials are more cost-effective than traditional randomized clinical trials because of their more efficient use of time and resources. For example, Fröbert et al. (2013), using the RRCT model, evaluated whether routine intracoronary thrombus aspiration (removal of a blood clot within the heart by the use of an aspirator) before primary percutaneous coronary intervention (unblocking of a coronary artery by inflating a balloon, causing a larger opening of the artery) reduced mortality. Michael Lauer, director of cardiovascular sciences for the National Heart, Lung, and Blood Institute, noted that the study was completed at a fraction of the cost ($300,000) compared with that required for a traditional clinical trial and was completed within a shorter period of time (Lauer and D’Agostino, 2013; National Heart, Lung, and Blood Institute, 2013).
A recent report on precision medicine envisions new taxonomies of diseases defined by their mechanisms and based on the availability of digital information in EHRs linked with genomic and other information (NRC, 2011). The potential for the prevention as well as the treatment of these diseases will be limited, however, if the underlying research fails to include the full range of determinants spanning all the clinical, genetic, epigenetic, and environmental variables that affect health. Social and behavioral data can describe potentially modifiable conditions that, along with clinical and biological data, can provide more preventive, diagnostic, and therapeutic options for improving individual and population health (Barrett et al., 2013).
The social and behavioral information in EHRs can advance both basic and applied research. For example, information on environmental attributes linked to a patient’s EHR can facilitate population research on the causal impact of changes in these environmental attributes on behavioral change, biomarkers of risk, and health outcomes. Longitudinal data on patients derived from EHRs will be valuable in establishing causality. This type of evidence is fundamental for establishing policies in a variety of health-related areas. In addition, and perhaps of relevance to practitioners, the availability of these data would enhance clinical research on the extent to which consideration of social and environmental factors are useful in improving the outcomes of care (such as for hypertension and diabetes control). Finally, clinical research on clinician knowledge of these factors may improve diagnosis, treatment, and follow up; allow better risk stratification; and enhance prediction of outcomes of care.
The “Meaningful Use” requirements of HITECH provisions were structured to maximize the effectiveness of EHRs once they are adopted. Profes-
sionals and hospitals that are eligible for incentives through HITECH are required to attest to or to measure performance on a series of objectives defined by CMS. The objectives specify EHR system functions and quality measures such as the use of computerized provider order entry, the collection of demographic data, and the use of clinical decision support. The objectives are organized into four categories: improve quality, safety, and efficiency and reduce health disparities; engage patients and families; improve care coordination and public health; and ensure adequate privacy and security protections for protected health information (HealthIT.gov, no date–c). The Meaningful Use program was divided into three stages. Stage 1 took effect in 2011, and Stages 2 and 3 (which have been given extensions) are expected to be in place in 2014 and 2017, respectively. As a general guideline, the focus of Stage 1 is data capture and sharing, the focus of Stage 2 is on advancing clinical processes, and the focus of Stage 3 is on improved outcomes (HealthIT.gov, no date–c).
Meaningful Use is defined through a public process. The Meaningful Use Workgroup of the Health Information Technology Policy Committee (HIT Policy Committee) defines a set of objectives and measures for each stage through a series of public meetings. The HIT Policy Committee, which is a federal advisory committee of the Office of the National Coordinator for Health Information Technology (ONC), hears the recommendations of the Meaningful Use Workgroup, other workgroups, and tiger teams (an assembled team of specialists) and drafts a letter to ONC with its recommended objectives and measures. ONC shares them with CMS, and ONC and CMS work jointly to define both the Meaningful Use requirements for eligible professionals and hospitals (released by CMS) and the requirements for EHR system certification (released by ONC). A proposed rule is first released, and then a final rule is released after public comment.
Deliberations within the HIT Policy Committee and its workgroups address the balance among moving as quickly as possible because of the urgency of achieving health care reform, the desire to improve patient outcomes, and the timing of incentives (which were front loaded); and moving more slowly because of limited capabilities in currently available EHR systems, the time needed to implement EHR systems, the realities of small clinical practices, and the desire to learn from previous experience with Meaningful Use before new stages are defined.5 As of October 2013, about one-half of eligible professionals and two-thirds of eligible hospitals had achieved Meaningful Use Stage 1, which represents a huge improvement over the 2009 baseline level of achievement (King and Adler-Milstein, 2013). Additionally, CMS released data at the end of April 2014 indicating that 88 percent of eligible professionals have registered for the Medicare or
5 Personal communication, G. Hripcsak, Colombia University, October 21, 2013.
Medicaid EHR incentive programs. Seventy percent of these professionals completed requirements and received incentive payments. Ninety-five percent of eligible hospitals had registered, with 91 percent completing Stage 1 requirements (ONC, 2014). Although progress continues, few providers and hospitals have completed adoption of Stage 2 (HealthIT.gov, 2014). Of particular relevance to our task, currently only 41 percent of hospitals are able to send and receive messages to organizations outside the hospital system regarding patient information, creating gaps in the potential for outside linkages to other public health resources (HealthIT.gov, 2014).
April 2014 also saw the release of the report by the JASON/MITRE Corporation, A Robust Health Data Infrastructure, which noted that “the current lack of interoperability among the data resources for EHRs is a major impediment to the unencumbered exchange of health information and the development of a robust health data infrastructure” (AHRQ, 2014, pp. 5–6). The report is referenced in Chapter 6 of this report.6
Meaningful Use represents a lever that can be used to steer health systems to better incorporate social and behavioral determinants of health. Some of these determinants have already been incorporated into Meaningful Use Stages 1 and 2 to some extent. Stage 1 includes the collection of information on a patient’s preferred language, gender, race, ethnicity, and smoking status (HHS and CMS, 2010). CMS opted to use the Office of Management and Budget’s (OMB’s) five categories for race and two categories for ethnicity. An optional Stage 1 menu objective for hospitals was included to collect advance directives for patients ages 65 years and older.
The CMS Final Rule for Meaningful Use Stage 2 maintained the social determinants of health from Stage 1, but gender was changed to sex so that it aligned with vital statistics reporting, and family health history was added as a menu objective (HHS, 2012). Furthermore, the summary of care record for patients who are transitioned or referred to another provider or care setting was required to include functional status, including activities of daily living and cognitive and disability status, if the provider knows it (i.e., if it is already recorded in the EHR). It was decided not to mandate the collection of disability status as a demographic variable because of the data collection burden and the lack of an agreed-upon definition. Gender identity and sexual orientation were considered but not included because of lack of consensus in public comments on whether doing so would be useful, the degree of sensitivity of the information, and how it would be recorded.
As of December 2013, the Meaningful Use Workgroup was developing recommendations for Stage 3. An August 2013 draft included items such as functional status with activities of daily living, relevant social and financial information, and relevant environmental factors affecting the patient’s
6 This text has been revised since the release of the Phase 1 report.
health; and the draft included the patient submission of information such as functional status (CPeH, 2013). At its August meeting, the HIT Policy Committee requested a change in emphasis so that all objectives included in the Meaningful Use Stage 3 definition were clearly linked to concrete health outcomes that were aligned with the national priorities. A new framework was created, and the workgroup was scheduled to present its recommendations to the HIT Policy Committee in March 2014 (Meaningful Use Workgroup, 2013).7
With the National Institutes of Health at the helm, a collaboration among the Association of State and Territorial Health Officials, the Blue Shield of California Foundation, the California HealthCare Foundation, the Centers for Disease Control and Prevention, CMS, The Lisa and John Pritzker Family Fund, the Robert Wood Johnson Foundation, and the Substance Abuse and Mental Health Services Administration was formed. Together, they requested that the IOM convene a committee of experts “to identify domains and measures that capture the social determinants of health to inform the development of recommendations for Stage 3 meaningful use of electronic health records (EHRs).” A 13-member committee was selected to address the charge. The committee comprised experts in the fields of social determinants of health, health information technology, behavioral and psychological issues, and measurement. (See Appendix D for the biographical sketches of the committee members.)
This study was conducted in two phases. Box 1-4 contains the complete statement of task for this study.
To meet its charge in Phase 1, the committee first established the rationale for adding social and behavioral domains into EHRs and considered how EHRs may assist providers in their decision making in a way that will result in improved health outcomes for their patients, regardless of Meaningful Use adoption and implementation. The committee held two information-gathering meetings during Phase 1 in order to clarify its statement of task; learn about meaningful use objectives; and hear from other
7 ONC’s Meaningful Use Workgroups were being restructured over the summer of 2014, and in July 2014, the HIT Policy Workgroup released its recommendations for Stage 3 Meaningful Use to ONC (Health IT Policy Committee, 2014). At the time of publication, it was unclear when ONC will be moving these recommendations forward to CMS and if and when CMS would accept them or request additional work for ONC on Stage 3 requirements.
Statement of Task
The Institute of Medicine will convene a committee to identify domains and measures that capture the social determinants of health to inform the development of recommendations for Stage 3 meaningful use of electronic health records (EHRs). The committee’s work will be conducted in two phases and will produce two products. As part of its work, the committee will:
Phase 1 (accomplished in this report)
- Identify specific domains to be considered by the Office of the National Coordinator,
- Specify criteria that should be used in deciding which domains should be included,
- Identify core social and behavioral domains to be included in all EHRs, and
- Identify any domains that should be included for specific populations or settings defined by age, socioeconomic status, race/ethnicity, disease, or other characteristics.
A brief Phase 1 report will be produced and submitted to the sponsors by the end of March 2014.
Phase 2 (to be addressed in a forthcoming report)
The committee will consider the following questions:
- What specific measures under each domain specified in Phase 1 should be included in EHRs? The committee will examine both data elements and mechanisms for data collection.
- What are the obstacles to adding these measures to the EHR and how can these obstacles be overcome?
- What are the possibilities for linking EHRs to public health departments, social service agencies, or other relevant non–health care organizations? Identify case studies, if possible, of where this has been done and how issues of privacy have been addressed.
A final report that includes the Phase 1 report and addresses the Phase 2 questions will be the final product.
The committee will make recommendations where appropriate.
experts in the field, stakeholders, and the public on domains that the committee should consider. (See the meeting agendas in Appendix C.) After each information-gathering meeting, the committee met in closed session for discussion and deliberation.
Before the first meeting and throughout the study process, the committee reviewed relevant literature. Its formal review of the literature focused on identifying peer-reviewed, published literature, reports from governmental agencies, and other IOM reports that were germane to the statement of task. The committee used the Ovid Embase, Ovid Medline, and Web of Science search engines, setting limits and using in its search specific medical subject headings terms pertinent to components of social and behavioral determinants of health. Given the vast literature on the range of social and behavioral determinants of health, systematic reviews were used when possible. The committee prioritized U.S. Preventive Services Task Force guidelines, as well as the Cochrane Database of Systematic Reviews.
For this study, the committee uses the term candidate to refer to the “core” domains (the third item of the Statement of Task) because the specific task for the Phase 1 report was to identify domains that should be considered by ONC for Stage 3 Meaningful Use. In this context the core domains are those that are “candidates” for being selected for Meaningful Use. The committee erred on the side of inclusion for its Phase 1 report while also trying to limit the number of candidate domains. Consequently, the committee further winnowed the list of candidate domains to a smaller number of recommended “core” domains during Phase 2. Throughout the study, the term domain refers to determinants of health that could include health conditions that, in turn, influence other health outcomes. The committee also established the following working definitions for “domains,” “measures,” “data sources,” and “EHRs”: (1) the “domain” is the definition of the conceptual variable, (2) the “measure” is the specific instrument through which the domain is assessed or operationalized, (3) the “data source” is where the measure can be obtained, and (4) “EHRs” are collections of electronic data stored and used by health care providers to manage patients’ health. For the purposes of this study, the committee employed a definition on social and behavioral determinants of health used in the National Research Council’s report Proposed Revisions to the Common Rule for the Protection of Human Subjects in the Behavioral and Social Sciences (NRC, 2014), noted in Box 1-5.
The study was limited by the need to keep a very tight timeline for preparation and publication of the Phase 1 report to provide ONC and CMS the opportunity to consider the committee’s candidate domains as part of Meaningful Use Stage 3. The committee first met in September 2013 and wrote this first report after its two initial meetings. Guided by a review of existing conceptual frameworks, the committee first identified an outline of the full set of domains for committee review and then narrowed these to a smaller number of domains best suited for consideration for inclusion in EHRs using evidence-based criteria and consensus methods.
Finally, the identification of thresholds for each measure was deter-
Social and Behavioral Determinants of Health Definition
“The term ‘behavioral’ refers to overt actions; to underlying psychological processes such as cognition, emotion, temperament, and motivation; and to bio-behavioral interactions. The term ‘social’ encompasses sociocultural, socioeconomic, and socio-demographic status; biosocial interactions; and the various levels of social context from small groups to complex cultural systems and societal influences” (Office of Behavioral and Social Science Research, 2010).
mined to be outside the scope of work of the committee described in the statement of task that the sponsor agencies presented to the committee. CMS uses thresholds to set the bar for the reporting of measures to achieve certification. For example, to measure smoking status, the Meaningful Use Stage 1 threshold is “more than 50 percent of all unique patients 13 years or older seen by the [eligible physician] have smoking status recorded as structured data” (CMS, 2010, p. 1).
Prior to the release of its Phase 1 report on April 8, 2014, the committee began to address its task for Phase 2. In fact, while its Phase 1 report was being reviewed by independent experts, the committee held its third public meeting. Its purpose was to learn from invited experts about measurement of social and behavioral determinants of health and successful implementation strategies for including measures of the domains in EHRs. A fourth public meeting was held that April to present the Phase 1 report to interested participants and receive feedback on the report. The meeting also allowed the committee to hear from speakers about the best ways to collect information, successes and challenges in linking EHR data between public health departments and other relevant organizations, and how systems can be developed in which data flow freely among all relevant users. A key component was learning about patient privacy protection issues in adding potentially sensitive social and behavioral data elements into EHRs. Finally, a panel of speakers addressed obstacles in adding measures to EHRs and suggested ways to overcome these barriers for the patient, provider, system, and society. Following each information-gathering meeting, the committee met in closed session for discussion and deliberation.
This report is a synthesis of Phase 1 and Phase 2 of the study. The Phase 1 report, woven into this report largely unchanged as Chapters 1–3, describes the committee’s process of selection of candidate domains for consideration for inclusion in all EHRs, including the conceptual frameworks used, the discussion of possible domains, and the criteria considered in the selection of domains (Chapter 2) and how specific populations are addressed (Chapter 3). Chapter 3 also identifies the evidence used to establish a candidate set of domains that the committee agrees should be considered for inclusion in all EHRs.
During the course of its Phase 2 work, the committee did make a few edits for clarification. For example, the domain name Tobacco Use and Exposure was more descriptive of the evidence reviewed for that domain than was the name Nicotine Use and Exposure. Accordingly, the name of the domain was changed throughout the report.
The material added during Phase 2 starts with Chapter 4, which details the measures for each domain that the committee reviewed. Chapter 5 considers the measures relative to one another on the basis of usefulness, readiness, and the committee’s overall judgment, and the committee recommends a parsimonious panel of measures for inclusion in all EHRs. Chapter 6 details challenges and opportunities in adding new data to EHRs, including addressing patient privacy issues, and examples are provided of how data can be shared with local public health departments and community agencies. Chapter 7 identifies the opportunities and challenges engendered by the adoption of the recommended panel of measures in all EHRs, including implications for future research. It also identifies the need for ongoing assessment and processes to consider adding additional measures as they become ready for inclusion in EHRs. A preface is included in this report, written by the committee co-chairs. Appendix A includes descriptions of all of the domains reviewed and not selected by the committee, and Appendix B contains a commissioned paper authored by an independent consultant to the committee. Appendix C includes the meeting agendas, and the committee member biographies are available in Appendix D.
AHRQ (Agency for Healthcare Research and Quality). 2014. A robust health data infrastructure. AHRQ Publication No. 14-0041-ef. Washington, DC: Agency for Healthcare Research and Quality.
Arend, J., J. Tsang-Quinn, C. Levine, and D. Thomas. 2012. The patient-centered medical home: History, components, and review of the evidence. Mount Sinai Journal of Medicine: A Journal of Translational and Personalized Medicine 79(4):433–450.
Barrett, M. A., O. Humblet, R. A. Hiatt, and N. E. Adler. 2013. Big data and disease prevention: From quantified self to quantified communities. Big Data 1(3):168–175.
Berwick, D. M., and A. D. Hackbarth. 2012. Eliminating waste in U.S. health care. JAMA 307(14):1513–1516.
Berwick, D. M., T. W. Nolan, and J. Whittington. 2008. The triple aim: Care, health, and cost. Health Affairs 27(3):759–769.
Blumenthal, D. 2009. Stimulating the adoption of health information technology. New England Journal of Medicine 360(15):1477–1479.
Blumenthal, D., and M. Tavenner. 2010. The “meaningful use” regulation for electronic health records. New England Journal of Medicine 363(6):501–504.
Bradley, E. H., and L. A. Taylor. 2013. The healthcare paradox. Why spending more is getting us less. New York: Public Affairs.
Bradley, E. H., B. R. Elkins, J. Herrin, and B. Elbel. 2011. Health and social service expenditures: Associations with health outcomes. British Medical Journal Quality & Safety 20(10):826–831.
CDC (Centers for Disease Control and Prevention). 2013. Social determinants of health. http://www.cdc.gov/socialdeterminants (accessed December 12, 2013).
Cebul, R. D., T. E. Love, A. K. Jain, and C. J. Hebert. 2011. Electronic health records and quality of diabetes care. New England Journal of Medicine 365(9):825–833.
CIA (Central Intelligence Agency). 2011. World factbook. Washington, DC: Central Intelligence Agency.
Cincinnati Children’s Hospital. 2012. Cincinnati Child Health Law Partnership (Child HeLP) Video File. http://www.cincinnatichildrens.org/service/g/gen-pediatrics/services/child-help (accessed December 12, 2013).
CMS (Centers for Medicare & Medicaid Services). 2010. Eligible professional meaningful use core measures. Measure 9 of 15. Stage 1. Washington, DC: CMS. http://www.cms.gov/Regulations-and-Guidance/Legislation/EHRIncentivePrograms/downloads/9RecordSmokingStatus.pdf (accessed November 6, 2013).
CPeH (Consumer Partnership for eHealth). 2013. Leveraging meaningful use to reduce health disparities: An action plan. Washington, DC: National Partnerships for Women and Families.
Elliott, A. F., A. Davidson, F. Lum, M. F. Chiang, J. B. Saaddine, X. Zhang, J. E. Crews, and C.-F. Chou. 2012. Use of electronic health records and administrative data for public health surveillance of eye health and vision-related conditions in the United States. American Journal of Ophthalmology 154(6 Suppl.):s63–s70.
Flexner, A. 1910. Medical education in the United States and Canada. A report to the Carnegie Foundation for the Advancement of Teaching. Boston, MA: The Merrymount Press.
Friedman, C. P., A. K. Wong, and D. Blumenthal. 2010. Achieving a nationwide learning health system. Science Translational Medicine 2(57):57cm29.
Friedman, D. J. 2006. Assessing the potential of national strategies for electronic health records for population health monitoring and research. Vital Health Statistics 2(143):1–83.
Friedman, D. J., R. G. Parrish, and J. A. Ross. 2013. Electronic health records and US public health: Current realities and future promise. American Journal of Public Health 103(9):1560–1567.
Fröbert, O., B. Lagerqvist, G. K. Olivecrona, E. Omerovic, T. Gudnason, M. Maeng, M. Aasa, O. Angerås, F. Calais, M. Danielewicz, D. Erlinge, L. Hellsten, U. Jensen, A. C. Johansson, A. Kåregren, J. Nilsson, L. Robertson, L. Sandhall, I. Sjögren, O. Östlund, J. Harnek, and S. K. James. 2013. Thrombus aspiration during ST-segment elevation myocardial infarction. New England Journal of Medicine 369(17):1587–1597.
Gray, B. H., T. Gowden, I. Johanse, and S. Koch. 2011. Electronic health records: An international perspective on “Meaningful Use.” Commonwealth Fund 28. http://www.commonwealthfund.org/Publications/Issue-Briefs/2011/Nov/Electronic-Health-RecordsInternational-Use.aspx (accessed February 24, 2014).
Hacker, K., and D. K. Walker. 2013. Achieving population health in accountable care organizations. American Journal of Public Health 103(7):1163–1167.
Han, Y. Y., J. A. Carcillo, S. T. Venkataraman, R. S. B. Clark, R. S. Watson, T. C. Nguyen, H. Bayir, and R. A. Orr. 2005. Unexpected increased mortality after implementation of a commercially sold computerized physician order entry system. Pediatrics 116(6):1506–1512.
Health IT Policy Committee. 2014. HITPC transmittal letter. July 15, 2014. Washington, DC: U.S. Department of Health and Human Services.
HealthIT.gov. no date–a. Benefits of EHRs. Improved diagnostics & patient outcomes. http://www.healthit.gov/providers-professionals/improved-diagnostics-patient-outcomes (accessed February 1, 2014).
HealthIT.gov. no date–b. How can electronic health records improve public and population health outcomes? http://www.healthit.gov/providers-professionals/faqs/how-can-electronic-health-records-improve-public-and-population-health-outcomes (accessed February 1, 2014).
HealthIT.gov. no date–c. Meaningful use definition & objectives. http://www.healthit.gov/providers-professionals/meaningful-use-definition-objectives (accessed November 11, 2013).
HHS (U.S. Department of Health and Human Services). 2012. Medicare and Medicaid programs; electronic health record incentive program—Stage 2. Federal Register 77(171): 53968–54162.
HHS. no date. National Quality Strategy (NQS). http://www.ahrq.gov/workingforquality (accessed February 24, 2014).
HHS and CMS. 2010. Medicare and Medicaid programs; electronic health record incentive program; final rule. Federal Register 75(144):44314–44588.
Hsiao, C.-J., E. Hing, and J. Ashman. 2014. Trends in electronic health record system use among office-based physicians: United States, 2007–2012. Atlanta, GA: Centers for Disease Control and Prevention, National Center for Health Statistics.
IOM (Institute of Medicine). 2010. The healthcare imperative: Lowering costs and improving outcomes: Workshop series summary. Washington, DC: The National Academies Press.
King, J., and J. Adler-Milstein. 2013. Hospital progress to meaningful use: State updates. Health Affairs Blog. http://healthaffairs.org/blog/category/health-it (accessed January 15, 2014).
Lamb, A. 1955. The Presbyterian Hospital and the Columbia-Presbyterian Medical Center, 1868–1943: A history of a great medical adventure. New York: Columbia University Press.
Lauer, M. S., and R. B. D’Agostino. 2013. The randomized registry trial—the next disruptive technology in clinical research? New England Journal of Medicine 369(17):1579–1581.
Link, B. G., and J. Phelan. 1995. Social conditions as fundamental causes of disease. Journal of Health and Social Behavior Special Issue 80–94.
Manchanda, R. 2013. The upstream doctors: Medical innovators track sickness to its source. New York: TED Books.
Marmot, M. G., M. J. Shipley, and G. Rose. 1984. Inequalities in death—specific explanations of a general pattern. Lancet 1(8384):1003–1006.
McCaw, B., H. M. Bauer, W. H. Berman, L. Mooney, M. Holmberg, and E. Hunkeler. 2002. Women referred for on-site domestic violence services in a managed care organization. Women & Health 35(2-3):23–40.
McDonald, C. J., P. C. Tang, and G. Hripcsak. 2014. Electronic health record systems. In Biomedical informatics: Computer application in health care and biomedicine, E. H. Shortliffe and J. J. Cimino, eds. London: Springer-Verlag. Pp. 391–421.
McGinnis, J., and W. H. Foege. 1993. Actual causes of death in the United States. JAMA 270(18):2207–2212.
McGinnis, J. M., P. Williams-Russo, and J. R. Knickman. 2002. The case for more active policy attention to health promotion. Health Affairs 21(2):78–93.
McIntosh, A., A. Cohen, N. Turnbull, L. Esmondel, P. Dennis, J. Eatock, C. Feetam, J. Hague, I. Hughes, J. Kelly, N. Kosky, G. Lear, L. Owens, J. Ratcliffe, and P. Salkovskis. 2004. Clinical guidelines and evidence review for panic disorder and generalised anxiety disorder. Sheffield, London: University of Sheffield, National Collaborating Centre for Primary Care. http://www.nice.org.uk/nicemedia/pdf/cg022fullguideline.pdf (accessed February 24, 2014).
Meaningful Use Workgroup. 2013. Stage 2 update. Presentation to the Health IT Policy Committee. http://www.healthit.gov/facas/sites/faca/files/MUWG_Stage3_13_Sep_4_FINAL_0.pdf (accessed February 1, 2014).
Melton, L. J. 1996. History of the Rochester Epidemiology Project. Mayo Clinic Proceedings 71(3):266–274.
National Heart, Lung, and Blood Institute. 2013. Dr. Michael Lauer co-publishes perspective piece on randomized registry trials. http://www.nhlbi.nih.gov/news/spotlight/fact-sheet/dr-michael-lauer-co-publishes-perspective-piece-on-randomized-registry-trials.html (accessed November 20, 2013).
NCHS (National Center for Health Statistics). 2006. Health, United States, 2006, with chartbook on trends in the health of Americans. Hyattsville, MD: National Center for Health Statistics.
NICE (National Institute for Health and Care Excellence). 2009. Depression: The treatment and management of depression in adults (partial update of NICE Clinical Guideline 23). NICE Public Health Guidance 90. United Kingdom: National Institute for Health and Care Excellence. http://www.nice.org.uk/nicemedia/pdf/CG90NICEguideline.pdf (accessed February 24, 2014).
NICE. 2010. Alcohol-use disorders: Preventing harmful drinking. NICE Public Health Guidance 24. United Kingdom: National Institute for Health and Care Excellence. http://www.nice.org.uk/nicemedia/live/13001/48984/48984.pdf (accessed February 24, 2014).
NICE. 2013a. Physical activity: Brief advice for adults in primary care. NICE Public Health Guidance 44. United Kingdom: National Institute for Health and Care Excellence. http://www.nice.org.uk/nicemedia/live/14176/63945/63945.pdf (accessed February 24, 2014).
NICE. 2013b. Tobacco: Harm-reduction approaches to smoking. NICE Public Health Guidance 45. United Kingdom: National Institute for Health and Care Excellence. http://www.nice.org.uk/nicemedia/live/14178/63996/63996.pdf (accessed February 24, 2014).
Noble, D. J., and L. P. Casalino. 2013. Can accountable care organizations improve population health?: Should they try? JAMA 309(11):1119–1120.
NRC (National Research Council). 1997. The computer-based patient record: An essential technology for health care. R. S. Dick and E. B. Steen, eds. Washington, DC: National Academy Press.
NRC. 2011. Toward precision medicine: Building a knowledge network for biomedical research and a new taxonomy of disease. Washington, DC: The National Academies Press.
NRC. 2014. Proposed revisions to the Common Rule for the protection of human subjects in the behavioral and social sciences. Washington, DC: The National Academies Press.
NRC and IOM. 2013. U.S. health in international perspective: Shorter lives, poorer health. Washington, DC: The National Academies Press.
OECD (Organisation for Economic Co-operation and Development). 2011a. Child outcomes. CO1.1: Infant mortality. In OECD family database. Geneva, Switzerland: Organisation for Economic Co-operation and Development.
OECD. 2011b. Health: Spending continues to outpace economic growth in most OECD countries. http://www.oecd.org/newsroom/healthspendingcontinuestooutpaceeconomicgrowthinmostoecdcountries.htm (accessed December 12, 2013).
OECD. 2013. Strengthening health information infrastructure for health care quality governance: Good practices, new opportunities and data privacy protection challenges. Geneva, Switzerland: Organisation for Economic Co-operation and Development.
Office of Behavioral and Social Science Research. 2010. Behavioral and social sciences research (BSSR) definition. Bethesda, MD: National Institutes of Health, Office of Behavioral and Social Sciences Research. http://obssr.od.nih.gov/about_obssr/BSSR_CC/BSSR_definition/definition.aspx#bfr (accessed February 18, 2014).
ONC (Office of the National Coordinator for Health Information Technology). 2013. Understanding the impact of health IT in underserved communities and those with health disparities. Prepared by NORC (National Opinion Research Center) at the University of Chicago. Washington, DC: U.S. Department of Health and Human Services.
ONC. 2014. Medicare EHR incentive program: Attestation patterns among professionals who first attested to Meaningful Use in 2011. Health IT quick-stat, no. 31. Washington, DC: Office of the National Coordinator for Health Information Technology. http://dashboard.healthit.gov/quickstats/pages/FIG-Medicare-Professionals-StageOne-Meaningful-Use-Attestation-Cohort-2011.html (accessed August 25, 2014).
Openchowski, M. W. 1925. The effect of the unit record system and improved organization on hospital economy and efficiency. Archives of Surgery 10(3):925–934.
Pagliari, C., D. Detmer, and P. Singleton. 2007. Potential of electronic personal health records. British Medical Journal 335(7615):330–333.
PwC (PricewaterhouseCoopers) Health Research Institute. 2010. Designing the health IT backbone for ACOs. Part I: Hospitals look to meaningful use and health information exchanges to guide them. Dallas, TX: PricewaterhouseCoopers LLP.
Reiser, S. J. 1991. The clinical record in medicine. Part 1. Learning from cases. Annals of Internal Medicine 114(10):902–907.
Rosenbloom, S. T., F. E. Harrell, C. U. Lehmann, J. H. Schneider, S. A. Spooner, and K. B. Johnson. 2006. Perceived increase in mortality after process and policy changes implemented with computerized physician order entry. Pediatrics 117(4):1452–1455.
RWJF (Robert Wood Johnson Foundation), Mathematica Policy Research, and Harvard School of Public Health. 2013. Health information technology in the United States: Better information systems for better care, 2013. Princeton, NJ: RWJF.
Stange, K. C., P. A. Nutting, W. L. Miller, C. R. Jaen, B. F. Crabtree, S. A. Flocke, and J. M. Gill. 2010. Defining and measuring the patient-centered medical home. Journal of General Internal Medicine 25(6):601–612.
Stead, W. W. 1989. A quarter-century of computer-based medical records. M.D. Computing: Computers in Medical Practice 6(2):74–81.
United Nations. 2009. World population prospects: The 2008 revision, highlights. Working paper no. Esa/p/wp.210. Geneva, Switzerland: Department of Economic and Social Affairs, Populations Division, United Nations.
Wagner, E. H., K. Coleman, R. J. Reid, K. Phillips, M. K. Abrams, and J. R. Sugarman. 2012. The changes involved in patient-centered medical home transformation. Primary Care 39(2):241–259.