The Patient-Reported Outcomes Measurement Information System (PROMIS) is an adaptive, self-reported outcomes assessment tool that utilizes item response theory and computer adaptive testing to efficiently and precisely evaluate symptoms and perceived health status. Efforts to implement and report PROMIS outcomes in spine clinical practice remain limited. The objective of this retrospective cohort study is to evaluate the performance and psychometric properties of PROMIS physical function (PF) and pain interference (PI) among patients undergoing spine surgery.
The authors identified all patients who underwent spine surgery at their institution between 2016 and 2018, and for whom there was retrievable PROMIS data. Descriptive statistics were calculated to summarize demographics, operative characteristics, and patient-reported outcomes. Assessments were evaluated preoperatively, and postoperatively within 2 months (early), 6 months (intermediate), and up to 2 years (late). Pairwise change scores were calculated to evaluate within-subjects differences and construct responsiveness over time. Pearson’s correlation coefficients were used to evaluate the association between PROMIS PF and PI domains. Subgroup analysis was performed based on the primary diagnoses of cervical radiculopathy, cervical myelopathy, or lumbar degenerative disease.
A total of 2770 patients (1395 males, 50.4%) were included in the analysis. The mean age at the time of surgery was 57.3 ± 14.4 years. Mean postoperative follow-up duration was 7.6 ± 6.2 months. Preoperatively, patients scored an average 15.1 ± 7.4 points below the normative population (mean 50 ± 10 points) in PF, and 15.8 ± 6.8 points above the mean in PI. PROMIS PF required a mean of 4.1 ± 0.6 questions and median 40 seconds (interquartile range [IQR] 29–58 seconds) to be completed, which was similar to PI (median 4.3 ± 1.1 questions and 38 seconds [IQR 27–59 seconds]). Patients experienced clinically meaningful improvements in PF and PI, which were sustained throughout the postoperative course. PROMIS instruments were able to capture anticipated changes in PF and PI, although to a lesser degree in PF early postoperatively. There was a strong negative correlation between PROMIS PF and PI scores at baseline (Pearson’s r = −0.72) and during follow-up appointments (early, intermediate, and late |r| > 0.6, each). Subgroup analysis demonstrated similar results within diagnostic groups compared to the overall cohort. However, the burden of PF limitations and PI was greater within the lumbar spine disease subgroup, compared to patients with cervical radiculopathy and myelopathy.
Patients receiving care at a tertiary spine surgery outpatient clinic experience significant overall disability and PI, as measured by PROMIS PF and PI computer adaptive tests. PROMIS PF and PI health domains are strongly correlated, responsive to changes over time, and facilitate time-efficient evaluations of perceived health status outcomes in patients undergoing spine surgery.