Relationship, differences, and agreement between objective and subjective sleep measures in chronic spinal pain patients with comorbid insomnia: a cross-sectional study.

Abstract

Sleep disturbances are one of the most frequent reported problems in people with nonspecific chronic spinal pain (nCSP) and presents an additional treatment challenge. Interventions targeting sleep problems are mainly based on subjective sleep complaints and do not take objective sleep into consideration. The aim of this cross-sectional study was to evaluate the relationship and conformity between self-reported and objectively measured sleep parameters (ie, questionnaire vs polysomnography and actigraphy). The baseline data of 123 people with nCSP and comorbid insomnia who are participating in a randomized controlled trial were analyzed. Pearson correlations were used to investigate the relationship between objective and subjective sleep parameters. Differences between objective and subjective sleep parameters were analyzed using t tests. Bland-Altman analyses were performed to quantify and visualize agreement between the different measurement methods. Except for the significant moderate correlation between perceived time in bed (TIB) and actigraphic TIB ( r = 0.667, P < 0.001), all other associations between subjective and objective measures were rather weak ( r < 0.400). Participants underestimated their total sleep time (TST) (mean difference [MD] = -52.37 [-67.94, -36.81], P < 0.001) and overestimated sleep onset latency (SOL) (MD = 13.76 [8.33, 19.20], P < 0.001) in general. The results of this study suggest a discrepancy (differences and lack of agreement) between subjective and objective sleep parameters in people with nCSP and comorbid insomnia. No or weak associations were found between self-reported sleep and objectively measured sleep. Findings suggest that people with nCSP and comorbid insomnia tend to underestimate TST and overestimate SOL. Future studies are necessary to confirm our results.