Application Of Promis Instruments To Sub-Populations Of Interest

Purpose of study The Patient-Reported Outcomes Measurement Information System (PROMIS) is a series of metrics that were introduced by the NIH in 2004. By utilising item-response theory with computer-adaptive testing, PROMIS instruments can assess patient-reported outcomes in dozens of domains with fewer questions, minimal floor and ceiling effects, and minimal loss of precision. PROMIS data are reported separately for each domain on a scale of 0 to 100, with a mean of 50. PROMIS scores were calibrated from Medicare and disability databases and it is unknown if these scores are representative of sub-populations of interest to the orthopaedic provider. This study aims to test the hypothesis that mean PROMIS scores for these sub-populations will not differ from NIH population mean values. Methods used Participants from two state university institutions were recruited to complete a voluntary one-time survey administered through REDCap during a three-month period starting in May 2017. The survey included the following PROMIS domains: Pain Interference, Physical Function, Mobility, Social Functioning, Depression, and Global Health. Respondents were grouped into one of three categories:Elite Athletes were defined as students-athletes on an NCAA Division 1 roster (n=38);Medical Students (n=15); andResidents/Fellows (n=76). Data were analysed using SAS 9.4 using Tukey-Kramer pairwise comparisons and one-sample t-tests, where appropriate. A p-value<0.05 was considered statistically significant. Summary of results Mean PROMIS scores for both Elite Athletes and Residents/Fellows differed from the population mean across all PROMIS domains (p<0.01). Mean PROMIS scores for Medical Students differed from the population mean for Physical Functioning, Pain Interference, Mobility, Social Functioning, and Global Health (Physical) (p<0.01). Mean PROMIS scores for Medical Students did not differ from the population mean for Depression (p=0.4780) or Global Health (Mental) (p=0.6623). Conclusions These data suggest that NIH mean PROMIS domain scores may not be generalizable to subpopulations of interest. This demonstrates the importance of caution with PROMIS score interpretation in the clinical setting and highlights the need for more research.