Reliability and Validity of Momentary Pain and Disability Assessments for Lumbar Spine Surgery Patients.

Study Design: Prospective cohort study. Objective: To evaluate the validity and reliability of ecological momentary assessments (EMAs) of pain and disability in patients undergoing lumbar spine surgery. Summary of Background Data: Patient-reported outcome measures (PROMs) are used to evaluate disease severity and spine surgery outcomes. Traditional cross-sectional PROMs may be limited by recall-bias and inability to capture longitudinal symptom dynamics. Momentary symptom evaluations using brief mobile surveys (i.e., EMAs) can overcome these limitations by capturing symptom severity and its change over time in patients' natural environments. Methods: Adults undergoing lumbar/thoracolumbar spine surgery completed EMAs five-times daily for approximately 3-weeks preoperatively to assess momentary pain and disability. Participants also completed traditional PROMs, including Patient-Reported Outcomes Measurement Information System (PROMIS) pain-intensity, pain-interference, and Oswestry Disability Index (ODI). Several analyses were performed to evaluate reliability of EMA items, composites, and summary metrics, their variability within-patients over time, construct-validity, and convergent-validity. Results: A total of 123 patients, 46% males, and mean age 57-years (+/-13) were enrolled. EMA metrics demonstrated high item-reliability, composite-reliability, and temporal-stability of assessments over time. In validity analysis, EMA pain and disability showed strong correlation with PROMs, including PROMIS pain intensity, pain interference, and ODI. Nonetheless, there was substantial variability of EMAs in each severity category of PROMs. For example, among patients with high PROMIS pain-intensity, the average EMA pain score was 70 but ranged from 22 to 97 (+/-20). Several variables including sex, age, and ODI were associated with EMA symptom variability. In multivariable analysis, EMA pain and disability (mean, variability, and mean*variability interaction) explained substantial variability in PROMs. Conclusion: Momentary assessments of symptom severity using EMAs is a valid and reliable approach to evaluate spine-related pain and disability in everyday life.