Racial disparities in treatment preferences for rheumatoid arthritis.

Racial disparities in the delivery of healthcare have been well documented across many disorders (1). Current efforts are now focused on understanding the reasons why minority patients often receive less aggressive care compared to Whites. While unwanted variability in healthcare utilization may be due to both system and provider factors, data suggest that differences in patient preferences may account for some of the differential use of healthcare services across persons of different racial backgrounds. For example, both Byrne et al. (2) and Ibrahim et al. (3) found that Black patients with moderate to severe osteoarthritis were less willing to consider total joint arthroplasty compared to White patients with similar disease severity. Similarly, Whittle et al. (4) found that White patients were more likely to be willing to undergo coronary artery bypass grafting compared to Black patients. Among seriously ill hospitalized patients, preferences for discussions related to resuscitation efforts also differs by race (5). These studies suggest that racial disparities in the use of relatively high risk medical interventions may be partially explained by patient preferences. In contrast, less is known regarding whether variability in patient preferences influences racial disparities in chronic disease. Cooper et al. (6) found that treatment preferences differ significantly among White, Black and Hispanic patients meeting criteria for major depression. Other studies examining chronic diseases including osteoporosis (7), osteoarthritis (8), diabetes (9) and hypertension (10) have failed to find an association between sociodemographic characteristics and treatment preferences; however, these studies were not designed, nor powered, to examine the impact of race or ethnicity on outcomes. Rheumatoid arthritis (RA) is the most common type of inflammatory arthritis affecting 1% of the world's population. This disorder results in significant disability in most patients within two decades from symptom onset and is associated with two-fold increased mortality rate (11). The economic impact of RA is comparable to that of coronary artery disease in large part due to the loss of work productivity (12). Some studies suggest that minority RA patients have worse outcomes compared to their counterparts. Specifically, greater levels of pain (p<0.05) and higher rates of disability have been reported in Black RA patients compared to their White counterparts (13,14). The care of patients with RA has changed dramatically over the past two decades, and now emphasizes the early introduction of aggressive therapies to suppress disease activity. This shift in treatment paradigm is supported by studies indicating that early suppression of disease activity improves both short and long-term clinical outcomes (15-17). Emerging data suggest that minority RA patients, with access to care and insurance, may be less likely to receive aggressive therapy compared to White patients. In a large retrospective cohort study of over 44,000 patients, Berrios-Rivera et al. (18) found that Black patients were about half as likely to use a biologic agent (the newest class of disease modifying agents) than were White patients of similar disease severity. Similarly, using data abstracted from a large national prospective cohort study of community-based RA patients, Head et al. (19) also found that Blacks were less likely to have been prescribed a biologic agent compared to White patients after adjusting for sociodemographic characteristics, disease severity, prior medication use and current health status. One explanation for these results is that patient preferences for aggressive treatment of RA differ by race. In order to examine this hypothesis we administered a conjoint analysis survey to RA patients under the care of a rheumatologist. Conjoint analysis is a well-validated tool originally developed to understand consumer preferences and predict market shares of innovative products (20-22). This method is strongly based on seminal work in mathematical psychology (23). It has a strong theoretical basis, obtains high levels of internal consistency, is able to predict future choices, and works in real world settings (20, 21, 24, 25). This approach has been used across diverse clinical settings in patients from varied sociodemographic backgrounds, including those with lower levels of education, to elicit preferences for healthcare (7, 8, 26-30). When faced with complex decisions, people typically evaluate a number of attributes and then make trade-offs to arrive at a final choice. Conjoint analysis evaluates these trade-offs to determine which combination of attributes are most preferred. Using this information, preferences for specific options can be calculated. Conjoint analysis is a decompositional technique that is based on the premise that respondents' preferences can be calculated based on the value that respondents attach to the specific attributes of the product under consideration. For example, consider having to choose from four insurance plans which differ on four attributes: co-pays, access to subspecialists, drug coverage, and deductibles. By asking subjects to evaluate these characteristics, using for example rating and paired comparison tasks (described in detail below), conjoint analysis can determine which plan is preferred by each individual subject. This method minimizes the influences associated with the context in which choices are presented, eliminates ordering effects by presenting treatment characteristics in random order, and makes trade-offs between competing options explicit. Careful consideration of the trade-offs involved in complex decisions has been shown to improve the quality of decision making (31). Because conjoint analysis elicits individual patient preferences based on how they value treatment characteristics, it is not biased by physicians' preferences, recognition of a treatment name, or personal experiences with specific medications.