Background/Purpose:

Total knee replacement (TKR) surgery is common with > 700,000 surgeries performed annually in the United States. Racial variation in TKR utilization has been documented in population-based studies, however, there is a paucity of prospective data on disparities in TKR use in patients with osteoarthritis (OA). We aimed to evaluate racial disparity in incident TKRs in a cohort of patients with physician-diagnosed OA followed in a current clinical trial. 

Methods:

Vitamin D and Omega-3 Trial (VITAL) is an ongoing nationwide, racially diverse, community-based, randomized controlled trial of 25,874 males age ≥50 and females age ≥55. We identified a knee pain subgroup with probable knee OA based on: a) self-reported knee pain symptoms in walking 2-3 blocks; b) pain > 1 day/week for > 1 year; and c) a doctor’s diagnosis of OA. This subgroup completed a modified Western Ontario and McMaster’s Universities Osteoarthritis Index (WOMAC) questionnaire, and reported prior TKR and laterality (103 people with bilateral TKR at baseline were excluded.) Participants reported new TKR, laterality and date, over the 2 year follow-up. We compared baseline WOMAC scores by race using t-tests. In multivariable-adjusted logistic regression analyses, we investigated the association of race with TKR risk, adjusting for potential confounders including age, sex, race, geographic location, WOMAC pain, function and stiffness, body mass index, income, education level, and self-reported depression. 

Results:

At baseline, 1508 (5.8%) of trial participants were eligible for the knee pain cohort. Of those, 1241 (94.7% of mailed) returned a baseline knee pain questionnaire and 1019 (82%) returned a follow-up questionnaire, a mean of 27 (±4) months after baseline. Within this group, 133 (13%) reported TKR in follow-up. Among those who underwent TKR, mean age was 68 (±7) years, 65% were female, and 13% were Black. Among those who did not have TKR, mean age was 67 (±7) years, 66% were female and 24% Black. All baseline WOMAC score means were significantly higher among Black than White patients. (Table)In multivariable logistic regression analyses, only race and WOMAC pain score were related to TKR incidence. The odds ratio of TKR among Black compared to White patients was 0.40 (95% CI 0.22, 0.74). After adjustment for baseline WOMAC pain and stiffness scores, the odds of TKR among Black patients was further reduced to 0.32 (0.18, 0.58). 

Conclusion:

Despite having worse self-reported worse knee pain, function and stiffness at baseline, and having demonstrated a high level of engagement by virtue of volunteering for a large randomized trial, Black patients had much reduced odds of undergoing TKR compared to White patients. This community-based, racially diverse cohort is unique as subjects have physician-diagnosed OA, and have been followed prospectively. Our data highlights the need for further efforts to correct this large racial disparity.