Skip to main content
. 2006 Oct 17;66(3):377–388. doi: 10.1136/ard.2006.062091

Table 3 Evidence of efficacy—pooled effect size (ES) and number needed to treat (NNT) for hand OA.

Intervention* Studies ESpain (95% CI) ESfunction (95% CI) NNT (95% CI) References
Level* No of studies (patients) Duration
Education+exercise v OA information Ib 1 (40) 3 months 2 (1 to 6) 19
Splint (full v half) Ia 2 (47) 1 week 0.64 (0.02 to 1.26) 4 (2 to 13) 21, 22
NSAIDs Ia 2 (654) 2–4 weeks 0.40 (0.20 to 0.60) 0.17 (−0.03 to 0.36) 3 (2 to 6) 23, 24
Topical NSAIDs Ia 2 (131) 2–3 hours 0.77 (0.32 to 1.22) NS 25
Topical capsaicin Ia 2 (318) 4 weeks 3 (2 to 5) 26
Chondroitin sulphate Ib 1 (92) 3 years NS 27
Chondroitin polysulphate Ib 1 (130) 3 years 8 (4 to 166) 27
IA corticosteroid Ib 1 (40) 24 weeks NS NS NS 28
Surgery Ia 7 (384) 3–66 months
 T+LRTI/IA v T −0.17 (−0.57 to 0.24) 0.03 (−0.37 to 0.44) NS 29
 TJR v T+IA −0.3 (−1.07 to 0.47) 29

*Compared with placebo, unless otherwise stated; †see table 1 for definitions

No, number of studies; ES, effect size of treatment compared with placebo unless otherwise stated; NNT, number needed to treat to obtain moderate to excellent (more than 50%) pain relief or symptomatic improvement; –, not available; NS, not significant; T, trapeziectomy; LRTI, ligament reconstruction and tendon interposition; IA, interposition arthroplasty; TJR, total joint replacement.