Reply

We appreciate the comments from Tecer et al.¹ Our intent in reporting the results of a randomized trial of marine omega-3s and/or vitamin D supplementation and knee pain in a large community-based population of older adults (VITAL)², was that our findings would be generalizable to the broader knee osteoarthritis (OA) population and relevant to the practicing clinician. We recruited participants nationwide, reducing biases from focusing only on a local population, but precluding the ability to perform physical exams. We did not include fibromyalgia or mood disorders in the analysis. However, a strength of this trial is the large cohort and randomization such that these potential confounders would be equally distributed among the treatment groups and thus the results still valid.

We agree that pain in knee OA is complex, with intra-articular and peri-articular structures as well as neural processes likely contributing. However, in clinical practice, clinicians very rarely attempt to parse out the specific pain generators for each patient. Measures such as the PainDETECT are certainly of interest in OA research, but are not used in the clinical setting.³ Our goal was to inform practicing clinicians on the use of vitamin D and marine omega-3s for pain in the knee OA population as a whole where subtyping is not feasible.

The OA research community has grappled with how best to subset what is likely a heterogeneous disease, however, to date there are no widely accepted phenotype classifications.^4–6For potential disease-modifying osteoarthritis drugs, it is critical that the drug be targeted to the appropriate population as approval from institutions such as the Food and Drug Administration is based on evidence of patient-reported outcome measures such as pain and change in structural endpoints.⁷ We thus studied the supplements as they are currently being used in practice, for the general knee pain population. Our subgroup analyses were predicated on information a clinician would likely have on hand.

Thus, we agree that understanding medication efficacy in knee OA is nuanced in part due to the heterogeneity of disease and the complexity of pain perception itself. Certainly, future investigations could be specifically designed to investigate specific subtypes of knee OA participants, however, that was not the aim of this trial.