> Skip repeated content

Disparities in Treatment: A Decision Sciences Perspective


  1. Background
  2. Disparities vs. Inequities
  3. What Decision Sciences Tells Us
  4. EXCEED-An Ongoing Study
  5. Commentary




Disparities in the delivery of care to patients with rheumatologic and musculoskeletal (orthopedic) conditions have been widely studied by investigators, with a focus, respectively, on patterns of use for DMARDs for patients with rheumatoid arthritis, and total joint replacement surgery for patients with osteoarthritis. An overview of the findings shows common elements that affect treatment choices, including race, socioeconomic status, cost of therapy, insurance, age, sex, education, cultural beliefs, and geographic region[1-6]. Whether the patient is treated by an internist or by a rheumatologist can also have a significant influence on access to newer agents[7], with patients who do see a rheumatologist generally achieving a better functional outcome than those who do not[8].

Data presented at HSS by visiting professor Maria E. Suarez-Almazor, MD, of Baylor College of Medicine and Houston VA Medical Center, sheds additional light on these issues and how physicians make treatment decisions for their patients. She also raises interesting questions about relevant preconceptions that both physician and patient may bring to their interactions.   

Disparities vs. Inequities

When examining this complex issue, even the term disparities merit closer examination, according to Dr. Suarez-Almazor. Disparities are generally acknowledged to be differences among subgroups of individuals, while inequities suggest an inappropriate or even unethical difference in the delivery of care. But discriminating between the two may not always be easy. For example, one study of African-American patients-a population with a higher frequency of osteoarthritis than among Caucasians-showed a greater reliance on prayer for relief of symptoms[9]. While this finding appears to identify a cultural choice, another study of African-Americans has demonstrated that fewer members of this population know an African-American peer who has had a joint replacement[10]. Since patients often make decisions based on a trusted source who had a successful outcome, individuals without access to such peers may be less likely to choose surgery.

Other information about race and total joint replacement is more clear-cut. White patients are twice as likely to undergo this surgery as are African-Americans or Hispanics, irrespective of prevalence or severity of osteoarthritis[2]. And as Dr. Suarez-Almazor observed, historically, African-Americans report less confidence in physician recommendations and medical interventions than do other ethnic groups.

A Canadian study that looked at the role of gender also yielded results whose implications might not have been immediately apparent. In the study population equal numbers of men and women with arthritis underwent hip replacement, a finding that appears to indicate equity in treatment. However, because the prevalence of disease among women is considerably higher, an equal number of surgeries actually constitute a disparity. The study showed that men were more likely to have discussed total joint arthroplasty with their physicians than were women, even though women were equally willing to have the surgery. Moreover, those women who did undergo arthroplasty were more disabled by the time of their referral[3].
Physician preferences and biases come into play as well. In one study that focused on cardiovascular disease, physicians presented with patient scenarios were less likely to recommend certain procedures to African-American women than to white men, despite similar descriptions of clinical status[11]. 

On the issue of access to care, the literature yields more surprising information. In Canada where there is universal healthcare, patients with low socioeconomic status receive fewer total joint replacements despite greater need[3]. In the United States, differences were also observed among patients on Medicare. Despite being insured, these individuals still had a lower rate of total joint replacement[6]. 

However, Dr. Suarez-Almazor also points out that differences in health care do not necessarily imply inequity. Factors such as personal preferences, variations in health status and clinical need, and "differences in need under a social justice concept, e.g., need to work" also affect treatment choices.

When we look at variations in total joint replacement among the population, it is helpful to ask, "Are they clinically, socially, and ethically appropriate? Or, do they represent inequitable health care?"

What Decision Sciences Tells Us

Although we would all like to believe that our decisions are made rationally, based on the information at hand, evidence suggests that this is not always the case. Decisions about healthcare may involve a number of "non-rational" emotional responses. Moreover, issues of probability that are not always easy to assess come into play, for example, likelihood of success of surgery, incidence of side effects, and long-term prognosis.

Nobel Prize winners Amos Tversky and Daniel Kahneman[12] found that people "dislike" losses more than they "like" gains, that is, they are risk-averse for gains and are more likely to take risks to avoid losses. Thus patients who are presented with the two following scenarios are more likely to select the first:

· Intervention X has 5% chance of mortality, but if you don't undergo surgery you will get worse, your pain will increase by 20% (loss)
· Intervention X has 5% chance of mortality, if you undergo surgery you will get better, your pain will improve by 20% (gain)
EXCEED-An Ongoing Study

In an effort to contribute further to an understanding of disparities in care for patients with arthritis, researchers from the Baylor College of Medicine and the Houston VA Hospital have embarked on a study funded by the Agency for Health Research and Quality (AHRQ) as part of the EXCEED program (Excellence Centers to Eliminate Ethnic/Racial Disparities.) Dr. Suarez-Almazor and her colleagues are focusing on identification of determinants of ethnic variations in the use of total knee replacement. Using their findings they plan to develop aids to assist patients and physicians in this process. Dr. Suarez-Almazor and her colleagues are employing a range of information-gathering tools, including:

· focus groups of ethnically diverse patients with osteoarthritis who have not undergone total joint replacement,
· telephone surveys of general public and patients with osteoarthritis
· surveys of physicians to assess preferences and biases
· analysis of medical interaction in total joint replacement clinics

While the study is ongoing, investigators have gathered some helpful data from the focus groups. Findings show that patients are strongly influenced by relatives, friends and peers who have undergone joint replacement. Patient perception of medical interaction and trust in their physician was also influential. Those considering joint replacement were sometimes unsure about the trade-offs between benefits and risks from surgery, undergoing "decisional conflict". The concept of "readiness" to undergo surgery or surgery as "last resort" was also often expressed by patients in relation to their views about joint replacement.

The telephone survey of 193 individuals in the general population showed that whites were generally wealthier and better-educated, Hispanic respondents were younger, and African-Americans were in worse health[13].  The patient survey of 198 patients showed that:

· African-Americans had worse WOMAC scores than other groups
· Total knee replacement (TKR) was recommended more often to African-Americans than to whites or Hispanics-owing to more severe symptoms (WOMAC)-however,
· More whites would consider TKR if their condition worsened and it was recommended by a physician
· White respondents were more likely to consider TKR as beneficial
· White respondents were more likely to know someone with TKR
· Overall, patients with TKR acquaintances more likely to undergo TKR (p=0.001)

Future study phases include investigating preferences with conjoint pair analysis, applying utility and prospect theory to patient preferences, and studying the physician-patient interactions at the time of the medical encounters 


Findings from the EXCEED study and others clearly demonstrate that variation-disparities and inequities-in total joint replacement and access to and use of DMARDs exist. Multiple factors contribute to this phenomenon. 

While many questions remain about how best to respond, healthcare providers can start by seeking to minimize disparities among groups of individuals at similar risk or with similar burden of illness. By raising awareness among physicians and patients about the complexity of this issue, we come closer to helping patients in all groups achieve comparable health with access to equivalent procedures.

At Hospital for Special Surgery, resources are already available to address some of the patient concerns described  in this article. For example, individuals who don't know anyone who has undergone a particularly procedure or therapy, in some cases may benefit from peer counseling. HSS also has an educational program in place for patients on managed care, who may need help in clarifying options available to them, in order to maximize their access to specialty care.


[1] Skinner et al. Racial, ethnic, and geographic disparities in rates of knee arthroplasty among Medicare patients. N Engl J Med. 2003 Oct 2;349(14):1350-9.

[2] Anderson JJ, Felson DT. Factors associated with OA of the knee in the first national Health and Nutrition Examination Survey (HANES I). Evidence for an association with overweight, race, and physical demands of work. Am J Epidemiol 1988;128:179-189. 

[3] Hawker G et al. Sex differences in arthroplasty provision. Arthritis Rheum September 1999 (abstract) 42 Suppl.

[4] Hawker GA. The quest for explanations for race/ethnic disparity in rates of use of total joint arthroplasty. J Rheumatol. 2004 Sep;31(9):1683-5.

[5] Hawker GA, Wright JG, Glazier RH, et al. The effect of education and income on need and willingness to undergo total joint arthroplasty. Arthritis Rheum. 2002 Dec;46(12):3331-9.

[6] Escalante A, Barrett J, del Rincon I, et al. Disparity in total hip replacement affecting Hispanic Medicare beneficiaries. Med Care. 2002 Jun;40(6):451-460. 

[7] Kwoh C, Kent et al. Variations in Treatment of RA by Rheumatologists and Non-Rheumatologists. Arthritis Rheum 2001; Vol. 44 Suppl.

[8] Yelin EH, Criswell LA, Feigenbaum PG. Health care utilization and outcomes among persons with rheumatoid arthritis in fee-for-service and prepaid group practice settings. JAMA 1996 276: 1048-1053.

[9] Ang DC, Ibrahim SA, Burant CJ, et al. Ethnic differences in the perception of prayer and consideration of joint arthroplasty. Med Care. 2002 Jun;40(6) 471-6.  

[10] Blake et al. Racial differences in social network experience and perceptions of benefit of arthritis treatments among New York City Medicare beneficiaries with self-reported hip and knee pain. Arthritis Rheum. 2002 Aug 15;47(4) 366-71.

[11] Schulman KA, Berlin JA, Harless W et al. The effect of race and sex on physicians' recommendations for cardiac catheterization. N Engl J Med 1999;340:618-626.

[12] Kahneman D, Tversky A, eds. Choices, Values, and Frames. Cambridge, UK: Cambridge University Press, 2000.


Summary prepared by Nancy Novick


Roberta Horton, ACSW, LCSW
Director, Department of Social Work Programs
Hospital for Special Surgery

Related Professional Content