Recipients of hip replacement for arthritis are less likely to be Hispanic, independent of access to health care and socioeconomic status.

OBJECTIVE: To compare the proportion of Hispanics among recipients of hip replacements for primary articular disorders, recipients of knee replacements for the same reason, and persons hospitalized for other reasons.

METHODS: Twelve of the 17 accredited hospitals in Bexar County, Texas, in which hip or knee replacement surgery is performed permitted us to review their medical records. From 1993 through 1995, 3,100 elective, non-fracture-related, hip or knee replacements were performed. These individuals were matched by age, sex, hospital, and month of admission with 4,604 persons hospitalized for other reasons. Age, sex, ethnic background, type of medical insurance, median household income by zip code of residence, joint replaced, and surgical diagnosis were abstracted from the medical records. The validity of variables abstracted from the medical records was tested by comparison with self-report data in 115 patients interviewed prior to elective hip or knee replacement surgery.

RESULTS: During the study period, 2,275 subjects had a total knee replacement and 825 had a total hip replacement. Recipients of hip replacements were significantly less likely to be Hispanic than were recipients of knee replacements (19.5% versus 29.9%; odds ratio [OR] 0.57, 95% confidence interval [95% CI] 0.46-0.71; P < or = 0.001) or persons hospitalized for other reasons (29.4% Hispanic; OR 0.67, 95% CI 0.55-0.81). The under-representation of Hispanics was more pronounced among persons undergoing hip replacement for osteoarthritis compared with recipients of knee replacements for the same disease (OR 0.48, 95% CI 0.37-0.62). This pattern persisted after adjusting for age, sex, type of medical insurance, and median household income by the zip code of residence. Concordance between medical records and self-report data on ethnic background was high (kappa = 0.93).

CONCLUSION: Recipients of hip replacement are less likely to be Hispanic than are other hospitalized persons with a similar level of access to care. The reasons for this under-representation probably involve factors in addition to lack of access to health care and low socioeconomic status. Further research is needed to understand the nature of such factors.