PURPOSE: The Western Ontario McMaster Universities (WOMAC) Osteoarthritis Index is a patient-reported outcome measure (PROM). It contains three domains: pain (5 items), physical function (17 items), and stiffness (2 items). The domains are often used as endpoints in clinical trials for knee osteoarthritis (KOA) therapies This study demonstrated a process to adapt the WOMAC into attributes for use in a discrete choice experiment (DCE) study sponsored by FDA to evaluate how patients with KOA make trade-offs between relevant endpoints, which may be used in regulatory decision-making.
METHODS: The adaptation process comprised four steps (see Figure 1). First, we focused on the pain and function domains due to their frequent use as endpoints in KOA trials. Second, both were framed as treatment effica cy attributes and were presented as improvement in scores from baseline. We created two versions of the DCE that presented the attributes using: 1) domain scores and 2) item scores for “walking on a flat surface.” Third, we determined the relevant improvement levels for both attributes: 0%, 30%, 50% and 100%. Lastly, we combined these attributes with other attributes and developed the DCE designs and surveys. We pre-tested both versions and the pretest interviews evaluated participants’ ability to understand the adaptations of the WOMAC, and to tradeoff the pain and function improvements.
RESULTS: Twenty-six participants (mean age: 60 years, 58% female, 67% White or Caucasian, 62% with ≥ 5 years since KOA diagnosis) pre-tested the surveys. Overall, they found the DCE survey comprehensible and were willing and able to consider tradeoffs between independ ent improvements in pain and function. Some with low baseline pain scores did not find the presented improvements meaningful. Therefore, we set an eligibility criterion of baseline WOMAC pain score ≥ 40 for the subsequent implementation of the DCE. We also restructured the surveys for easier comprehension and revised the attributes for better cognitive feasibility and relevance. The final DCEs included these attributes: 1) improvement in pain, 2) improvement in function, 3) duration of improvement and 4) risk of developing scar tissue in the knee. Each choice task includes a status quo option.
CONCLUSIONS: Using the WOMAC as an example, our study demonstrated how PROMs can be adapted for use in patient preference studies to inform regulatory decision-making, contributing to the limited literature in this area.
