J Neurosurg Spine 30:405–413, 2019
The Patient-Reported Outcomes Measurement Information System (PROMIS) was developed to provide a standardized measure of clinical outcomes that is valid and reliable across a variety of patient populations. PROMIS has exhibited strong correlations with many legacy patient-reported outcome (PRO) measures. However, it is unclear to what extent PROMIS has been used within the spine literature. In this context, the purpose of this systematic review was to provide a comprehensive overview of the PROMIS literature for spine-specific populations that can be used to inform clinicians and guide future work. Specifically, the authors aimed to 1) evaluate publication trends of PROMIS in the spine literature, 2) assess how studies have used PROMIS, and 3) determine the correlations of PROMIS domains with legacy PROs as reported for spine populations.
METHODS Studies reporting PROMIS scores among spine populations were identified from PubMed/MEDLINE and a review of reference lists from obtained studies. Articles were excluded if they did not report original results, or if the study population was not evaluated or treated for spine-related complaints. Characteristics of each study and journal in which it was published were recorded. Correlation of PROMIS to legacy PROs was reported with 0.1 ≤ |r| < 0.3, 0.3 ≤ |r| < 0.5, and |r| ≥ 0.5 indicating weak, moderate, and strong correlations, respectively.
RESULTS Twenty-one articles were included in this analysis. Twelve studies assessed the validity of PROMIS whereas 9 used PROMIS as an outcome measure. The first study discussing PROMIS in patients with spine disorders was published in 2012, whereas the majority were published in 2017. The most common PROMIS domain used was Pain Interference. Assessments of PROMIS validity were most frequently performed with the Neck Disability Index. PROMIS domains demonstrated moderate to strong correlations with the legacy PROs that were evaluated. Studies assessing the validity of PROMIS exhibited substantial variability in PROMIS domains and legacy PROs used for comparisons.
CONCLUSIONS There has been a recent increase in the use of PROMIS within the spine literature. However, only a minority of studies have incorporated PROMIS for its intended use as an outcomes measure. Overall, PROMIS has exhibited moderate to strong correlations with a majority of legacy PROs used in the spine literature. These results suggest that PROMIS can be effective in the assessment and tracking of PROs among spine populations.