. Author manuscript; available in PMC: 2009 Dec 7.

Published in final edited form as: Arthritis Rheum. 2009 Nov 15;61(11):1539–1544. doi: 10.1002/art.24824

Table 4.

Effect estimates for the association of SBA and cartilage loss using 3 different methodologic approaches^*

Methodologic approach	OR (95% CI)^†
M:N matched case–control approach (n = 459 knees, 4,446 subregions)	7.5 (5.6–9.9)
Subregion-based approach (n = 968 knees, 9,680 subregions)	5.4 (4.1–7.0)^‡
Knee-based approach (n = 968 knees)	3.0 (2.2–4.2)^‡

SBA = subchondral bone attrition; OR = odds ratio; 95% CI = 95% confidence interval; M:N = cases:controls.

^†

OR for the association of SBA with cartilage loss.

^‡

Adjusted for potential confounders (age, sex, body mass index) because these analyses are across individuals. The M:N matched case–control approach does not require such adjustments since the analyses are within knee.