"Initial Nonoperative Treatment Choices for Thumb Carpometacarpal Osteoarthritis".

Background: We evaluated whether baseline patient-reported outcome measures (PROMs) are associated with initial treatment choice of hand therapy (HT) or corticosteroid injection (CI). We hypothesized patients with worse baseline PROMs would choose CI.

Methods: We have an ongoing prospective study of patients aged 35-85 with thumb carpometacarpal (CMC) osteoarthritis (OA). We track initial treatments and PROMs including visual analog/numerical rating sub-scales (VA/NRS), brief Michigan Hand Questionnaire (bMHQ), and Patient-Reported Outcomes Measurement Information System (PROMIS) Pain Interference (PI) and Global Health (GH) modules. Global Mental Health (GMH), Global Physical Health (GPH), and EuroQol are calculated. We retrospectively identified patients from 2019-2022 who initially chose HT or CI. We performed pairwise comparisons of independent variables between HT and CI groups, then exploratory bivariate logistic regression (LR) analyses for independent variables. We then performed more stringent multivariable LR analyses.

Results: We included 156 patients with 191 thumbs (average age 61 years (SD 9); 77% female). 158 thumbs received initial HT and 33 CI. For exploratory bivariate LR analyses, baseline and activity-related VA/NRS pain and bMHQ scores were worse for patients pursuing CI. Each multivariable LR model demonstrated higher VA/NRS activity-related pain levels and was still independently associated with increased odds of choosing CI.

Conclusions: Adjusting for covariates, baseline activity-related pain was associated with increased odds for choosing CI. Other baseline PROMs/demographics were not associated with initial treatments. Our findings can inform providers, patients, and investigators considering HT and CI, as well as their potential comparative outcomes, for initiating nonoperative thumb CMC OA care.