Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
6 Policy Issues of Integration Data and measurement Romana Hasnain-Wynia, Ph.D. Northwestern University, Feinberg School of Medicine Communication in an effective, patient-centered manner is at the heart of providing high-quality health care. However, many chal- lenges exist to achieving high quality of care for low-literacy patients. First, few measures are available for either patient-centeredness or equity in health care, as there is currently little incentive to use such measures. Second, health care organizations do not collect data about patientsâ race, ethnicity, and primary language in patient- reported processes. Such data would allow for the assessment of whether implemented initiatives indeed made changes for the bet- ter. A third challenge is recognizing that providing patient-centered care within institutions is critical, but not sufficient. Fourth, health literacy is part of a larger, more complex component of providing effective communication with all populations. Many publications have demonstrated that low-literacy racial and ethnic minorities and patients with limited English proficiency have poor outcomes. The system must move beyond documenting disparities toward actions to eliminate them. As Lurie discussed, when deciding how best to use resources, it is important to use what is available and not wait for the ideal. 63
64 TOWARD HEALTH EQUITY AND PATIENT-CENTEREDNESS Current measures of equity are stratified measures of clinical effec- tiveness linked to race and ethnicity, limited English proficiency, and literacy. Challenges to developing better equity measures begin at the organizational level. Although 80 percent of hospitals and 45 percent of physician practices reported collecting patient race, eth- nicity, and language data, less than 20 percent use data for quality improvement. It simply is not part of the culture of most providers. Little incentive exists to stratify data, which would allow data to be used to support changes. Development of equity measures would be greatly helped by creating a uniform method for collecting race, ethnicity, and primary language data at the health care organiza- tion level, coupled with providing incentives for organizations to do so. The ideal measures of equity are measures of patient-centered- ness. Current patient-centered measures are self-reports of satisfac- tion, such as those used in CAHPS. These measures are important to capture, but there is reluctance to place weight on them as true measures of patient-centeredness because of their weak linkages to outcomes and the potential unintended consequences to providing care of reporting these data. To move beyond patient experiences of care, structural measures are needed, such as whether organizations know and understand their patient populations. Organizations need to be responsible for implementing ways to assess equity and dis- parities, suggesting that incentives to measure patient-centeredness must be developed. Health disparities are driven by a combination of who you are and where you get your care. Studies have shown that the magni- tude of disparities decreases substantially across specific quality measures when controlling for differences in care between hospitals, suggesting that where minority patients receive their care is a driver of disparities in health care. Understanding the balance between these drivers of disparities (who you are versus where you get your care) is critical when determining where interventions should be targeted. Policies to support elimination of disparities and improved patient communication should also be developed further. The poten- tial for performance incentives to improve quality but also to aug- ment the disparities gap must be recognized. It is important to focus on patient-centered care to reduce disparities nationally by targeting interventions both within hospitals as well as by developing policies that improve care across hospitals and other health care settings, Hasnain-Wynia concluded.
POLICY ISSUES OF INTEGRATION 65 National perspective Paul Schyve, M.D. The Joint Commission The Joint Commissionâs mission and focus is quality and safety of patient care. The Joint Commission, which accredits approxi- mately 15,000 providers on the basis of national standards, has had standards in place for respecting cultural differences for many years. These standards have increasingly included standards on cultural competence, linguistic competence, and health literacy. Currently, the patientâs primary language and barriers to communication are required to be recorded for every patient. A movement also exists to expand what is currently thought of as âcultureâ to include the cultures associated with visual, hearing, mobility, and cognitive dis- abilities. Addressing cultural differences in communication is critical to providing high-quality care, as the frequency and magnitude of adverse events increase with poor communication. Figure 6-1 shows the relationship between providing high- quality safe care to the patient at the âsharp endâ and the organiza- tional culture and infrastructure at the âblunt end.â While patient- Patient âSharp Endâ Patient- centered care Clinicians Supportive infrastructure âBlunt Endâ Organizational culture FIGURE 6-1â Providing high-quality safe care. Figure 6-1.eps
66 TOWARD HEALTH EQUITY AND PATIENT-CENTEREDNESS centered care occurs between clinicians and patients, it relies on the organizationâs supportive infrastructure, and the infrastructure (i.e., resources, policies, and procedures) is enabled, created, and sus- tained by the organizationâs culture. The ultimate goalâto provide patient-centered care for each patientâcan be achieved only if the infrastructure and culture are in place to enable providers to do so. For example, it is difficult for English-speaking clinicians to commu- nicate with Spanish-speaking patients if interpretive servicesâpart of the organizationâs infrastructureâare unavailable. This triangle can be used to make a number of points. First, the triangle itself can be applied to many levels of the health care sys- tem: the practitionerâs office, the provider organization (e.g., clinics and hospitals), the health plan, or society. Second, quality improve- ment interventions can be targeted at each of these levels: the pro- vider organization, health plan, and society. Third, within each level of the health care system, the quality improvement intervention can be targeted at each section of the triangle: organizational culture, infrastructure, and patient-centered care. The three sections in the triangle are interdependent, so an intervention in one section will reverberate through the others. Fourth, infrastructure and culture are necessary components if true patient-centered care is to be achieved. Fifth, culture is depen- dent on the leadership in each level of the health care systemâthe office, the organization, the health plan, and society. Improvement in the health care system is dependent, in part, on whether leaders can be engaged at these different levels. Finally, a challenge in creating a supportive infrastructure for patient-centered care is the potential for stereotyping. Being able to communicate effectively with a patient does not automatically trans- late into understanding that specific patient and his or her needs. Infrastructures that standardize the understanding of and response to patients in a particular cultural/linguistic/literacy group (i.e., ste- reotyping) may lead to the wrong understanding of or response to an individual within that group. Creating an infrastructure without stereotyping individual patients requires understanding the differ- ences within groups as well as between groups. While recognizing that disparities in processes of care will always exist because each patient and his or her preferences are dif- ferent, part of health care is managing these differences. Disparities in outcomes can be influenced by appropriate responses to differ- ing patient preferences, as well as unnecessary variability in care processes.
POLICY ISSUES OF INTEGRATION 67 Expectations of how the system should function are necessary to drive change. Expectations about the culture of health care systems include the following: ⢠Organizational culture: Improving care requires awareness, sensitivity, and humility from health care organizations. Rec- ognition that disparities and literacy are relevant to all people is critical because everyone has a language, cultural/ethnic background, and literacy level and is likely to become dis- abled in some way with aging (e.g., limited vision, hearing, mobility). Organizations should understand the risks that result from ignoring these issues. ⢠Education: Physicians, administrators, and patients and fami- lies should be expected to recognize their roles in reducing disparities and improving literacy. All actors should also sup- port each other in understanding and behaving as expected in these roles. ⢠Diverse workforce: The workforce should be diverse to help health care organizations become more innovative in com- municating with patients and in making the system better at many levels. To complement cultural expectations, expectations of the infra- structure must also be addressed, including the following: ⢠Supportive infrastructure: Resource staff, interpretation and translation equipment, and information technology should all be used as necessary enablers of communication, data col- lection, and analysis. ⢠Continuous improvement: Continuous improvement is a fun- damental and ethical responsibility in health care. Providers should pay attention to cultural competency, literacy, and lan- guage when working on projects targeting specific diseases or procedures, such as cardiac or diabetes care, not just those projects aimed at culture, literacy, and language. ⢠Measurement: Data are necessary to identify targets for improvement. As mentioned many times, what is not mea- sured cannot be improved. How these variables are measured and how measures are used and stratified should be identi- fied to optimize care delivery. Evaluation of where we are with respect to these expectations is critical, requiring a sequence of actions to take place. First, there
68 TOWARD HEALTH EQUITY AND PATIENT-CENTEREDNESS must be an understanding of how evaluation tools should be used, requiring the use of comparative data. Next, there must be exter- nal evaluations through accreditation of education and health care. These evaluations require valid measures of care processes and out- comes. The final step of evaluation is third-party public disclosure to validate the reliability of data. Schyve concluded by discussing the reimbursement system. In health care, all stakeholders should be driven by altruistic incen- tives to provide higher-quality, safer patient care to meet patientsâ needs. A secondary incentive for producing better care is reducing the waste of health care resources that result from adverse events and system failures. The reimbursement policy and infrastructure must be transformed to support delivery of better care by reduc- ing disincentives to providing high-quality, safe care and providing incentives to do so. Discussion Incentives Agreeing with the notion that few incentives exist to stratify data, a question arose about the ability of many hospitals to stratify, as the numbers of minority patients may be too low to do so. It was noted that a significant minority of hospitals could in fact stratify even after combining all care into broad groups for specific condi- tions such as cardiac care. In response, Hasnain-Wynia said that it is important to understand the level at which the intervention is targeted. In this case, the goal is to link data to quality metrics and to make changes through quality improvement. Even when dealing with small patient populations (e.g., a Vietnamese patient popula- tion of 20 people), it is important to know what level of care these populations are receiving. The bigger challenge, Hasnain-Wynia said, is to aggregate and report differences in care in the context of performance incentive programs. It is also important to begin development of methodolo- gies to fill voids in the literature, such as whether certain groups of people or geographic regions can be clustered together. Communication In health care, communication with the public is a large problem that needs to be addressed, said Carolyn Clancy of the Agency for Healthcare Research and Quality (AHRQ). Special outreach efforts
POLICY ISSUES OF INTEGRATION 69 are needed to communicate with hospital CEOs, especially in the area of health literacy. Through her research, Hasnain-Wynia noted that the largest com- munications challenges are with hospital administrators. Although a handful of organizations and leaders understand the issues and are working on them, this understanding and corresponding action are not widespread. In addition, clinicians often do not believe that disparities occur in their own practices or hospitals. Adverse Effects of Data Collection Responding to a comment about the patient fear generated in response to data collection, aggregation, and disclosure, Schyve rec- ognized the courage it takes for people to collect and share data that can sometimes uncover unexpected, embarrassing results. Unex- pected findings must nevertheless be communicated in order to embolden others to take risks in collecting and using data. Public disclosure of organization-specific data should occur only when appropriate, as premature disclosure may decrease the likelihood that people will collect and use the data for improvement. Data that are used in the aggregate can often demonstrate where a problem exists. Disparities or Patient-Centered Care? It is important for individual organizations to understand what data are revealing, Hasnain-Wynia said. For example, in some places it takes longer than recommended standards for black patients with acute myocardial infarction to get from the emergency department to the catheterization lab. Further qualitative analysis of data showed that the reason for this disparity was that many black patients wanted to consult with their primary care physicians before under- going another service. As another example, Hispanic women often have longer visits because they want to talk with male family mem- bers. Are these examples of health disparities or patient-centered care? It is a fine, and sometimes difficult, line to identify. Data tell part of the story, not the whole story, but they help in beginning to reveal what that story is, Hasnain-Wynia said. Schyve emphasized that it is very common in health care to jump directly from recognizing a problem to choosing a solution, sometimes because that solution was successful in another insti- tution that faced the same problem. This is often inappropriate,
70 TOWARD HEALTH EQUITY AND PATIENT-CENTEREDNESS though; only data can reveal the specific causes of the problem, which may differ among institutions. Applying the same âsolutionâ to a different cause is likely to waste resources, lead to disappointing results, and leave the problem unresolved.