Table 2. Primary Outcomes at 18-Month Follow-up in a Study of the Effect of High-Intensity Strength Training on Knee Pain and Knee Joint Compressive Forces Among Patients With Knee Osteoarthritis.
| Outcome | Intensity | Control | Mean difference (95% CI) | P valuea | |
|---|---|---|---|---|---|
| High | Low | ||||
| Primary outcomes | |||||
| WOMAC pain (0-20) | |||||
| Baseline mean (SD) | 7.0 (2.7) | 7.4 (2.6) | 7.2 (2.7) | ||
| 18-mo follow-up, unadjusted mean (SD) | 4.9 (3.2) | 4.5 (3.4) | 4.8 (2.9) | ||
| 18-mo absolute change (95% CI) | −2.0 (−2.6 to −1.4) | −2.8 (−3.5 to −2.2) | −2.3 (−2.9 to −1.6) | ||
| No. | 108 | 104 | 99 | ||
| High intensity vs controlb | 0.3 (−0.6 to 1.2) | .56 | |||
| High intensity vs low intensity | 0.8 (−0.1 to 1.7) | .07 | |||
| Low intensity vs control | −0.6 (−1.5 to 0.3) | .22 | |||
| 18-mo adjusted means (95% CI)c | 5.1 (4.5 to 5.7)d | 4.4 (3.8 to 4.9) | 4.9 (4.3 to 5.5) | ||
| No. | 91 | 86 | 84 | ||
| High intensity vs control | 0.2 (−0.6 to 1.1) | .61 | |||
| High intensity vs low intensity | 0.7 (−0.1 to 1.6) | .08 | |||
| Low intensity vs control | −0.5 (−1.4 to 0.3) | .22 | |||
| Knee joint compressive force, Ne | |||||
| Baseline, mean (SD) | 2326 (727) | 2325 (782) | 2261 (691) | ||
| 18-mo follow-up, unadjusted mean (SD) | 2448 (783) | 2498 (813) | 2456 (844) | ||
| 18-mo absolute change (95% CI) | 89 (−59 to 237)f | 116 (−6.2 to 238) | 162 (10 to 314) | ||
| No. | 65 | 64 | 63 | ||
| High intensity vs controlb | −73 (−271 to 124) | .47 | |||
| High intensity vs low intensity | −27 (−223 to 170) | .79 | |||
| Low intensity vs control | −46 (−245 to 152) | .65 | |||
| 18-mo, adjusted means (95% CI) | 2453 (2296 to 2611) | 2475 (2325 to 2625) | 2512 (2352 to 2672) | ||
| No. | 50 | 55 | 48 | ||
| High intensity vs control | −58 (−282 to 165) | .61 | |||
| High intensity vs low intensity | −21 (−235 to 193) | .85 | |||
| Low intensity vs control | −37 (−256 to 181) | .74 | |||
Abbreviation: WOMAC, Western Ontario and McMaster Universities Osteoarthritic Index.
Pairwise comparisons were deemed significant at P ≤ .05/6 = .0083 from the adjusted model.
Unadjusted pairwise comparisons based on 18-month changes.
The model-adjusted outcomes used only participants with complete follow-up data. Sensitivity analyses were conducted to multiply imputed data for participants with missing data under the assumption that data were missing at random (eTable 2 in Supplement 2).
The difference in the number of completers between the unadjusted and adjusted analyses was due to the model-adjusted outcomes using only participants with complete follow-up data.
One pound of force is equal to 4.45 N of force.
The difference in the number of completers in the unadjusted analysis for the 2 primary outcomes was due to missing baseline and/or a follow-up biomechanical gait analysis.