BBH 302- L13

Provider-Level Factors that Impact Health Inequities

*Three main provider-level factors* were identified as potential sources of healthcare inequities including: *bias/prejudice against minorities greater clinical uncertainty beliefs/stereotypes about patient behaviors* Physicians work under intense time constraints and have to make recommendations for treatment or diagnostic tests based on limited information, and patients' symptoms and characteristics, which often involves making assumptions based on previous experiences with similar patients. Physicians need to combine their *medical knowledge with patient histories* (e.g., disease comorbidities) and patient characteristics (e.g., age, race, sex). Without time to fully assess patients' needs, beliefs, and behaviors, it can be difficult to provide the highest quality of care. *Bias/Prejudice Against Minorities* As we learned in the Lesson: Discrimination and Health, personally-mediated racism can be intentional and unintentional. Unintentional forms of discrimination include nonverbal behaviors that signal discomfort (e.g., reduced eye contact or a closed body posture, such as folded arms). Other forms of discrimination are more intentional, such as lower rates of referrals for diagnostic tests in minorities compared to Whites. In a study by Schulman et al. (1999), attendees at a medical conference were asked to participate in a study. They were shown video-recorded interviews of patients that varied by age (age 55 or 70), sex and race (Black or White). If you are interested in reading more about Schulman's research, including viewing the patient images that were used in the study, see The Effect of Race and Sex on Physicians' Recommendations for Cardiac Catheterization article published by The New England Journal of Medicine. *Patients' insurance, occupation, and health history* were identical across race and within age groups. Physicians at the conference were asked to use the information provided, including the subject's rating of chest pain, to make a decision about referring the patient for cardiac catheterization, which is a diagnostic test or treatment for cardiovascular diseases/problems. The results from the study showed that men and Whites were more likely than women and Blacks to be referred for cardiac catheterization; Black women were the least referred. The race of the physician did not matter; that is, even Black physicians were least likely to refer Black women. Whether intentional or not, physician biases impact the quality of health care, and clinical decisions are influenced by race. Perhaps *physicians' biases were truly based on their experiences with hundreds of patients*. If that is the case, would we see the same phenomena with medical students, who have no clinical practice experience to shape their perceptions? A similar study by Rathore et al. (2000) examined *medical students' *perceptions of patients' symptoms. Students viewed video vignettes of either a Black female or White male displaying identical angina symptoms. Overall, students assigned a lower quality of life rating to the Black female. Non-minority students rated the White male as higher in quality of life; minority students rated both patients similarly. Students were more likely to correctly diagnose the White male patients' symptoms as angina (74%), compared to less than half correctly diagnosing the Black female patients' symptoms as such (46%). A statement in the *Unequal Treatment report* asks, "How...could a well-meaning group of healthcare professionals, working in their usual circumstances with diverse populations of patients, create a pattern of care that appears...to be discriminatory?" Clinical uncertainty may play a role. *Greater Clinical Uncertainty* As mentioned earlier, physicians must make quick decisions about treatment based on the information they have at their disposal. The assumptions physicians make to determine a treatment path are referred to as priors; the hope is that physicians combine these priors with new, updated information that would guide them to provide the best course of treatment. *Priors Case Study* The following case study was used in the Unequal Treatment report to illustrate this point: A 50-yr-old Latino male and a 50-yr-old White male visit the same White physician. Both are healthy, experience the same symptoms, and describe these symptoms to the doctor. The physician believes that both patients have a low probability of developing cardiovascular problems based on priors. The Latino male has English as a second language, so there are some things that are lost in translation when he describes his symptoms to the physician. The White patient is referred for diagnostic tests, but the Latino patient is not. (Adapted from Unequal Treatment Brief Report) Why did the doctor arrive at different decisions, when both patients shared the same symptoms? Is the expression of pain and distress the same across cultures? Did the doctor feel more comfortable in the encounter with the White patient than with the Latino patient? What priors did the physician use to determine that the White patient needed to be referred for testing, but the Latino patient did not? This uncertainty is known as the uncertainty hypothesis; this hypothesis has been used to understand why errors of omission and commission occur. *Uncertainty impacts health disparities by:* leading to either too much or too little care for minorities (mismatch between CARE and NEED) leading to poorer health outcomes for minorities leading minorities to demand less healthcare leading minorities to seek care at lower rates leading minorities to have lower rates of compliance *Beliefs/Stereotypes about Patients and Patient Behaviors* Countless research studies on implicit bias show that participants often ascribe more negative descriptors to minorities than they do to Whites. Studies reviewed in the Unequal Treatment report show that Black patients are believed to be less compliant with treatment recommendations, and are more likely to be thought to have lifestyles that interfere with treatment, such as drug abuse. A study by *Cooper and colleagues* (2012), measured clinician's racial bias, and perceptions of patient compliance. Patients' perceptions of their clinicians were measured using a post-visit survey. The results showed that clinicians rated White patients as more cooperative than Black patients. Compared to clinicians with lower levels of racial bias, Black patients rated racially-biased clinicians as less respectful, rated that they liked them less, had lower confidence in them and would not recommend them. In contrast, White patients rated more racially biased clinicians as more respectful and perceiving that they are liked by them. A review by *FitzGerald and Hurst* (2017) concluded that there was *evidence of bias based on:* race and ethnicity gender socioeconomic status age disability mental illness weight status (i.e., overweight or obese) having AIDS having a brain injury (perceptions about patient responsibility for the injury) being an intravenous drug user social circumstances Bias was found to impact diagnoses, treatment recommendations, the number of questions the patients were asked, the number of tests that were ordered, among other things that impact the quality of care (FitzGerald and Hurst, 2017; pg. 13). Furthermore, biases may prevent physicians from recommending helpful and needed treatments or diagnostic tests for fear that they would be wasted on a non-compliant or destructive patient. *Some of the key findings from select studies* that were included in the FitzGerald and Hurst (2017) review include the following: Black patients were assumed to be of lower socioeconomic status (SES). Black patients presenting with challenging behaviors were more likely to be prescribed higher doses of opioids, compared to White patients exhibiting similar challenging behaviors. Middle-aged women were twice as likely to receive a mental health diagnosis than their middle-aged, male counterparts. Low-SES Black and Latina women were more likely to be recommended for an intrauterine contraceptive device (IUD) than low-SES White women. What do these findings tell you about inequities in the quality of care for women and minorities?

The Canary in the Mine

Before the advent of *electronic air monitoring devices*, *coal miners* used to bring a caged canary into the mines with them to estimate air quality. A sick or dead canary meant that there were likely high levels of carbon monoxide or some other toxin in the air that could harm the miners. Thus, the canary was a warning sign that something was amiss. Health disparities researchers often refer to health disparities as the canary in the mine. The canary is a metaphor for health disparities, and the mine is a metaphor for the healthcare system. That is, health disparities (canary) are the warning sign that the healthcare system (mine) is broken. Findings from the 2016 National Healthcare Quality and Disparities Report published by the Agency for Healthcare Research and Quality show that minorities continue to experience a lower quality of care than Whites. Figure 1, using data from the 2014 Behavioral Risk Factor Surveillance System, shows that minorities were more likely than Whites to report that they did not see a doctor, or went without care due to cost. FIGURE 1: *Percent of Nonelderly Adults who did not Receive or Delayed Care in the Past 12 Months* by Race/Ethnicity, 2014 Source: KFF. Disparities in Health and Health Care: Five Key Questions and Answers 2016. Figure 2 is a map showing *differences in the quality of care across the country*. Increasing quartiles indicate a greater difference, or lower quality care for minorities compared to Whites. Potential sources of these differences are addressed in this lesson in section, Healthcare Inequities: Unequal Treatment.

The Cliff Analogy

Let's return to the cliff analogy, from Jones (2014). Is it enough to simply move people back from the cliff? With that approach, the risk is substantially reduced, but conditions are not equitable. Jones (2014) suggested that instead, we make sure that all populations have equitable access to resources and services. Having *medical care and tertiary prevention* available to marginalized populations is not enough, particularly if those resources and services are poorer quality than those accessible to advantaged populations. Having *safety net programs *available in low-income communities is not enough, particularly when those programs are often poorly-funded and/or rely on charitable donations, and they may not provide comprehensive services that foster the best health outcomes. *Putting up a fence* is not enough because *primary prevention* mainly relies on individuals' ability (and motivation) to access those services (e.g., opportunities to be physically active, access to healthy foods). Why are some populations so close to the edge of the cliff? It is important to note that the *depth of the cliff differs for populations*; the distance from the top of the cliff to the bottom of the cliff is a metaphor for disparities in access to healthcare. For some highly disadvantaged populations, it is a quick fall into disease, given the absence of primary and secondary prevention. For others, it may be a slower process, with many opportunities to be rescued along the way. In an ideal world, all three levels of prevention would exist equally for all populations at the edge of the cliff. Until we dismantle institutionalized "isms" that create opportunities for some and inhibit or prevent opportunities for others, populations will always be at the edge of the cliff, and health disparities will persist.

Patient-Level Factors that Impact Health Inequities

Patient-level factors influence the quality of health care for racial and ethnic minority groups. Patients enter the clinical encounter, carrying healthcare needs with them, as well as preferences and expectations about the care they are going to receive. During the clinical encounter, physicians rely on patients being open, honest, and clear in the reporting of symptoms, health behaviors, and other attitudes/beliefs/practices that may help physicians provide the best care to their patients. *Patient-level factors* that may be a source of healthcare inequities include: patient preferences mistrust/treatment refusal care-seeking behaviors and attitudes *Patient Preferences* These may include patient preferences for a male or female physician or being seen by a physician of a certain race/ethnicity. Findings from a study by LaVeist and Nuru-Jeter (2002) showed that Black, Hispanic, Asian, and White patients with same-race physicians all reported a higher level of healthcare satisfaction than patients with a physician of the opposite race. Whites were most likely to have a same-race physician, followed by Asians, Blacks and Hispanics. In addition, Hispanics and Asians who did not speak English as a first language were more likely to have a physician of the same race. A low level of satisfaction with one's provider can lead to mistrust and treatment refusal. FIGURE 3: *Same-Race Physician* Source: LaVeist and Nuru-Jeter (2002) Number and percentage of quality measures for which members of selected groups experienced better, same, or worse quality of care compared with reference group (White) in 2013-2015 *Mistrust/Treatment Refusal* The authors of the Unequal Treatment report state that it is often difficult to separate patient preferences from treatment refusal and general mistrust of clinicians. Several studies report that minority populations report higher levels of discrimination in healthcare encounters, which can lead to mistrust. If patients do not believe their physicians have their best interest in mind, they are likely to refuse treatment or diagnostic procedures, and may not adhere to treatment plans or medication recommendations. Particularly for Blacks, and other populations with a history of medical transgressions (e.g., Tuskeegee Syphilis Study), mistrust is shaped by this history. Figure 4, using data from the National Healthcare Quality and Disparities Report, shows that while the majority of minorities received the same or better care than Whites, there is still a substantial percentage who receive worse care, particularly for Blacks and Hispanics. FIGURE 4: *Number and percentage of quality measures for which members of selected groups experienced better, same, or worse quality of care *compared with reference group (White) in 2013-2015 Source: AHRQ. 2016 National Healthcare Quality and Disparities Report *Rockville, MD: Agency for Healthcare Research and Quality; July 2017.* Number and percentage of quality measures for which members of selected groups experienced better, same, or worse quality of care compared with reference group (White) in 2013-2015 Figure 5 shows data from a survey that assessed *perceptions of health care quality for Whites, Blacks, and Latinos*. It shows that Whites were far more likely than Blacks or Latinos to perceive that Blacks and Latinos receive the same quality of care as Whites. Blacks and Latinos, however, were more likely to report that their racial and ethnic groups receive a lower quality of care than Whites. FIGURE 5: Perceptions of Disparities in Health Care Quality Source: Kaiser Family Foundation March/April 2006 Kaiser Health Poll Report Survey, April 2006 (Conducted April 2006) The Unequal Treatment report showed that Blacks were more likely than Whites to refuse diagnostic tests and life-saving procedures, although studies also showed that Whites were more likely than Blacks to be referred for these tests and procedures. Another study reviewed in the report showed that Blacks with end-stage renal disease were more likely to refuse a kidney transplant than White patients with end-stage renal disease. Overall, however, treatment refusal could not explain racial and ethnic differences in these health outcomes. *Care-Seeking Behaviors and Attitudes* Studies have shown that minorities tend to seek care at later stages of disease, which may play a role in health outcomes. The later the disease presentation, the more difficult it is to treat. Late stage disease diagnosis is associated with more negative health outcomes and higher mortality rates than those associated with early stage diagnosis. The reasons for seeking care late vary, but are influenced by limited access to care, and lack of a stable relationship with healthcare providers. Figure 6 shows data on mortality and early-stage (Stage 1) diagnosis of *breast cancer* from a study by Iqbal et al. (2015). The authors found that the smallest percentage of Blacks, Hispanics and South Asians (37%-40%) in the study were diagnosed in the early stages of the disease, compared to over 50% in Whites and Japanese participants. The greatest rate of breast cancer mortality was seen in Black women. Thus, a later stage of diagnosis may increase mortality. Again, however, these factors do not explain major racial/ethnic healthcare inequities.

Eliminating Healthcare Inequities

Soon after the 2012 U.S. elections, *Craig Froehle *posted an image on Google+ to illustrate his frustrations about the differences between equal opportunity and fairness or equity. Using an image of the *Great American Ball Park in Cincinnati, Ohio, *he pulled together a graphic of children of differing heights trying to see into the ball park over a fence. On the left side of the image is the "to a *conservative*" approach. In this illustration, equal treatment would mean providing each child with a crate to stand on, so that they could see over the fence. The problem with this conservative approach is that the crate alone does not do much to help the smallest child...the tallest and medium-height child can see over the fence, but the smallest child cannot. In fact, the tallest child has the most advantage. The smallest child is a metaphor for marginalized populations (e.g., poor, unemployed, segregated, etc.), and the tallest child is a metaphor for privileged populations. Instead, *equity* - as shown on the left side of the image labeled "to a *liberal*" - would mean providing each child with an equitable opportunity to see over the fence, which means providing as many crates as needed to achieve the final goal. Here, the tallest child needs no crates, the medium-height child only needs one crate, and the smallest child needs two crates. This illustration has seen many iterations and has been widely used across many disciplines to discuss issues of fairness, equity, and justice. *Problem with Equality* So, what is the problem with equality, as communicated in this illustration? Think about this question prior to viewing the answer. Problem with Equality *Equity as the Solution* Is the "liberal" approach within the illustration the answer to these issues? Think about this question prior to viewing the answer. The Equitable Approach, as shown on the Liberal Side

Health System-Level Factors that Impact Health Inequities

System-level factors include those that govern the ways in which healthcare systems are organized and financed, as well as the ease with which patients can access services. These factors may disproportionately impact poor, minority, and immigrant patients. *Four main factors* were identified as potential sources of healthcare inequities, including: cultural and language barriers lack of stable relationships with primary care providers financial incentives to limit services fragmentation of healthcare financing and delivery *Cultural and Language Barriers* In a 2001 survey of healthcare quality published by the Commonwealth Fund, minorities reported greater difficulty in communicating with physicians; Hispanics and Asian Americans reported the greatest communication problems (Figure 7). Those whose primary language was not English reported greater difficulty understanding instructions from doctors (Figure 8). In a 2014 study on low-income Americans and the Affordable Care Act, a larger percentage of Hispanics with Spanish as their primary language reported greater difficulty understanding their health plan, including costs and services provided, compared to Hispanics with English as their primary language (Figure 9). Thus, language and cultural barriers may not only impact the clinical encounter and patients' satisfaction with their physician, but it may also impact their ability to navigate healthcare access and coverage. FIGURE 7: Hispanics and Asian Americans Had More Communication Difficulties During Doctor Visits Source: The Commonwealth Fund 2001 Health Care Quality Survey. FIGURE 8: Non-English Speakers Have More Difficulty Understanding Instructions from Doctor's Office Source: The Commonwealth Fund 2001 Health Care Quality Survey. FIGURE 9: Health Plan Rating, Understanding, and Gaps among Nonelderly Insured Hispanic Adults by Language Source: KFF. "How do Health Care Access, Use, and Experiences Vary by Language for Insured Hispanic Adults? *Lack of Stable Relationships with Providers* Minorities have been found to be more likely to seek care at local, community-based clinics and emergency departments, or some location other than a doctor's office. Figure 10 shows the usual sources of care for Blacks, Whites, and Hispanics, by insurance status. Whether uninsured, privately insured, or a Medicaid recipient, Whites were most likely to list a doctor's office as their usual source of care. A greater percentage of uninsured Blacks and Hispanics reported being seen at a clinic, than at a doctor's office or some other place. Compared to Whites, a greater percentage of Hispanics who were privately insured or on Medicaid reported receiving care at a clinic or some other locations. The overall conclusion from Figure 11 is that even when insured at the same level as Whites, minorities are less likely to be cared for by a physician in a doctor's office. Having a usual source of care can increase the use of preventative care (e.g., wellness checks and screenings), and can reduce patient mistrust if the relationship with the provider is respectful and trustworthy. FIGURE 10: Type of Usual Source of Care by Coverage Type and Race/Ethnicity Source: KFF. "Racial and Ethnic Disparities in Access to and Utilization of Care among Insured Adults FIGURE 11: Type of Usual Source of Care among Nonelderly Insured Hispanic Adults by Language Source: KFF. "How do Health Care Access, Use, and Experiences Vary by Language for Insured Hispanic Adults? *Financial Incentives to Limit Services* Physicians, nurses, and healthcare organizations are often incentivized for a number of "achievements," including positive patient health outcomes (i.e., lower rates of disease), and limited service use. In other words, physicians will be paid more when their patients have better health outcomes, when they see more patients (which usually means reduced time spent with each patient), and when they limit treatments and services that increase cost. *The problem with this approach is that:* by increasing the patient load (in order to get more payments), physicians may take on a larger load than they can manage, which reduces their quality of care physicians may focus more time and attention on healthier patients in order to reap the benefit of having greater positive patient outcomes clinicians and healthcare systems that serve a large minority and/or low-income population will likely have patients with a high burden of disease, reducing financial incentives that could be used to improve services and facilities there is lack of oversight of physicians, making it hard to determine whether they are being ethical and equitable in their care these practices may disproportionately and negatively impact minorities The *authors of the IOM Unequal Treatment report *suggest that a better approach would be to provide financial incentives to physicians for greater patient satisfaction, or for better clinical outcomes, such as better disease management and control (e.g., blood pressure control). *Fragmentation of Healthcare Systems* Minorities are often enrolled in health care coverage plans with higher copays, higher deductibles, and lower-quality coverage than non-minorities. Quality may be impacted by limiting the policy-holder's freedom to see a clinician of their choice or their ability to visit the care setting of their choice. In addition, lower quality plans cover fewer services, which limits the quality of care for those who may need it most (likely patients living in or near poverty).

Healthcare Inequities: Unequal Treatment

The *Institute of Medicine (IOM) *published a report in 2002 that upended healthcare professionals, health researchers, and policy makers. In this report, entitled *Unequal Treatment*: Confronting Racial and Ethnic Disparities in Health Care, results from scores of empirical papers were reviewed, and discussions from meetings with clinicians, health researchers, tribal leaders, and other professionals were synthesized and reduced to these major findings: —Racial and ethnic disparities were consistently found across a wide range of healthcare settings (managed care, public/private hospitals, teaching or community), disease areas (cardiovascular disease, cancer, diabetes, HIV), and clinical services, even when various factors that partially explain health disparities are accounted for (i.e. socioeconomic status, stage of disease presentation, comorbidities). That is, *racial and ethnic minorities received inferior care* and had worse health outcomes compared to Whites, *regardless of*: where they were seen, what disease they were seen for, how advanced their disease, the burden of multiple diseases, or whether or not they were living in poverty. *Alan Nelson *is a retired physician and former president of the *American Medical Association*. He was the chairperson of the committee that published the *IOM Unequal Treatment report*, and stated that "disparities in health care delivered to racial and ethnic minorities are real and are associated with worse outcomes in many cases, which is unacceptable." Let's take a look at some of the individual findings from the report. *Three major levels of influence *were reviewed as potential factors that may explain healthcare inequities: *Patient-level Provider-level System-level* Each factor, and examples of how these factors manifest to impact health disparities are reviewed in the following sections. It is important to note, that while the Unequal Treatment report reviews these various factors as potential sources of healthcare disparities, they conclude that they do not fully explain disparities in health between minorities and Non-Hispanic Whites. The major conclusion is that minorities receive inferior treatment. We do not want this message to get lost in the studies that will be reviewed in the following sections.