Over the years, dozens of patient-reported outcome measures (PROMs) have been used in orthopedics practice and research. Considered as a group, these questionnaires have certain limitations:
- They are validated only for particular diagnoses or patient populations
- Patients are often asked to fill out several questionnaires whose content overlaps, which can lead to survey fatigue and lower completion rates
- Researchers have their choice of several different PROMs to assess the same anatomic location, which makes it difficult to standardize studies
For these reasons, the National Institutes of Health created the PROMIS (Patient-Reported Outcomes Measurement Information System) set of assessment instruments. These questionnaires are specific to domains (e.g., physical function, mental health, social health) rather than diseases.
Most PROMIS physical function (PF) questionnaires can be administered as a computer-adaptive test (CAT), in which question delivery is customized in real-time according to the patient’s answers. This allows for a higher level of precision while requiring fewer questions.
Melvin C. Makhni, MD, MBA, director of Complex Spine Surgery in the Department of Orthopaedic Surgery at Brigham and Women’s Hospital, and colleagues conducted the first systematic review to thoroughly compare PROMIS PF questionnaires with “legacy” PROMs in spine surgery. They describe favorable results for PROMIS in The Spine Journal.
One of the authors searched PubMed and Embase for English-language articles related to physical function in spine surgery published by June 21, 2021. By consensus of all authors, 37 full-text articles were included in the final analysis (combined sample size of 10,296 patients).
The research team compared four PROMIS PF forms (Short Form 10a, PF CAT, PF–Upper Extremity CAT, and Mobility CAT) with 34 unique legacy PROMs relevant to spine surgery. Frequently studied legacy PROMs were the Neck Disability Index, Oswestry Disability Index, Short Form–12 item physical component summary, and visual analog scales of back and leg pain.
The PROMIS PF questionnaires correlated well with the legacy questionnaires and outperformed them in terms of administrative burden. They were similar with regard to quality criteria:
- Correlation—Correlation coefficients were converted to an absolute value for 36 of the 37 studies; the weighted total of 26 Pearson correlations was 0.59 (indicating moderately strong correlation) and the weighted total of 10 Spearman correlations was 0.70 (indicating strong correlation)
- Average time to completion—48 seconds for PROMIS PF questionnaires vs. 175 seconds for legacy questionnaires (P<0.001)
- Average number of questions presented—4.2 for PROMIS PF questionnaires vs. 9.5 for legacy questionnaires (P<0.001) (P=0.015)
- Reliability (whether the results on subsequent completion of the questionnaire will be altered by a patient’s background characteristics)—No significant difference between the two groups of questionnaires
- Floor and ceiling effects (whether a questionnaire adequately captures patients whose physical function is very poor or very good, respectively)—No significant differences between the two groups
Guidance for Patient Care
These data show PROMIS PF measures can be administered efficiently and to a broad patient population while providing high-quality results. As another advantage, scores from different PROMIS PF questionnaires can be compared and interpreted easily because they’re drawn from a common item bank.
In some patients with spine problems, however, intervention is more focused than the PROMIS PF questionnaires address. In those cases, legacy PROMs may still be needed to provide anatomic-specific and condition-specific measurements of physical function.