In patients with first metatarsophalangeal joint osteoarthritis (first MTPJ OA), the most responsive outcome measures were the pain subscales of the Foot Health Status Questionnaire (FHSQ) and the Foot Function Index Revised Short Form (FFI-RS), as well as pain severity while walking, according to a study published in Arthritis Care & Research.
With worsening severity, first MTPJ OA leads to increased pain, disability, and deformity, with reduced range of motion. Despite considerable clinical impact, there has been little research into standard treatment efficacies or the sensitivities of outcome measures used to evaluate therapies.
In a sub-analysis of a larger parallel-group randomized trial (Australian New Zealand Clinical Trials Registry ID: ACTRN12613001245785) in which the efficacy of rocker-sole footwear vs prefabricated foot orthoses was examined, a total of 88 participants (mean age, 57.2; 58.0% women) with first MTPJ OA were analyzed regarding the responsiveness of several outcome measures over 3 months. Patients were asked to complete the FHSQ, FFI-RS, and report pain and stiffness levels using a visual analog scale. Paired t-tests, Cohen’s d, standardized response means (SRM), and the Guyatt Index (GI) were used to assess outcome measure subscale responsiveness, and minimally important differences (MID) were used to estimate required sample sizes. The FHSQ pain score was the study’s primary outcome.
Although all measures were sensitive to change, with medium or greater effect sizes, there were 3 sub-scores that had large or very large effect sizes across measures. At 12 weeks, compared with baseline, the FHSQ pain subscale (73.4 vs 53.3, respectively; P <.001), the FFI-RS pain subscale (23.3 vs 42.2, respectively; P <.001), and the pain severity during walking (21.2 vs 47.8, respectively; P <.001) were found to have excellent responsiveness and were deemed suitable for assessing post-intervention sensitivity to change. It was estimated that 20 to 33 patients in each group would be needed for MID detection at 80% power, and that 27 to 44 patients were required to reach 90% power.
Study limitations include the lack of a gold standard for assessing responsiveness, a lack of consensus regarding MID definition parameters, and alteration of the FFI-RS wording on some questions that may affect psychometric validity and reliability.
“These findings provide useful information to guide researchers in selecting appropriate outcome measures for use in future clinical trials and assist in determining sample size requirements,” noted the authors.
Reference
Menz HB, Auhl M, Tan JM, Levinger P, Roddy E, Munteanu SE. Comparative responsiveness of outcome measures for the assessment of pain and function in first metatarsophalangeal joint osteoarthritis. Arthritis Care Res (Hoboken). March 2019. doi:10.1002/acr.23883