Test-Retest Reliability and Responsiveness of PROMIS Sleep Short Forms Within an RCT in Women With Fibromyalgia

Abstract

Background: Nonrestorative sleep is commonly reported by individuals with fibromyalgia, but there is limited information on the reliability and responsiveness of self-reported sleep measures in this population. Objectives: (1) Examine the reliability and validity of the Patient-Reported Outcomes Measurement Information System (PROMIS) sleep measures in women with fibromyalgia, and (2) Determine the responsiveness of the PROMIS sleep measures to a daily transcutaneous electrical nerve stimulation (TENS) intervention in women with fibromyalgia over 4 weeks compared with other measures of restorative sleep. Methods: In a double-blinded, dual-site clinical trial, 301 women with fibromyalgia were randomly assigned to utilize either Active-TENS, Placebo-TENS, or No-TENS at home. Measures were collected at baseline and after 4 weeks of treatment. To assess self-reported sleep, the participants completed three PROMIS short forms: Sleep Disturbance, Sleep-Related Impairment, Fatigue, and the Pittsburgh Sleep Quality Index (PSQI). To assess device-measured sleep, actigraphy was used to quantify total sleep time, wake after sleep onset, and sleep efficiency. Linear mixed models were used to examine the effects of treatment, time, and treatment*time interactions. Results: The PROMIS short forms had moderate test-retest reliability (ICC 0.62 to 0.71) and high internal consistency (Cronbach's alpha 0.89 to 0.92). The PROMIS sleep measures [mean change over 4 weeks, 95% confidence interval (CI)], Sleep Disturbance: -1.9 (-3.6 to -0.3), Sleep-Related Impairment: -3 (-4.6 to -1.4), and Fatigue: -2.4 (-3.9 to -0.9) were responsive to improvement in restorative sleep and specific to the Active-TENS group but not in the Placebo-TENS [Sleep Disturbance: -1.3 (-3 to 0.3), Sleep-Related Impairment: -1.2 (-2.8 to 0.4), Fatigue: -1.1 (-2.7 to 0.9)] or No-TENS [Sleep Disturbance: -0.1 (-1.6 to 1.5), Sleep-Related Impairment: -0.2 (-1.7 to 1.4), Fatigue: -.3 (-1.8 to 1.2)] groups. The PSQI was responsive but not specific with improvement detected in both the Active-TENS: -0.9 (-1.7 to -0.1) and Placebo-TENS: -0.9 (-1.7 to 0) groups but not in the No-TENS group: -0.3 (-1.1 to 0.5). Actigraphy was not sensitive to any changes in restorative sleep with Active-TENS [Sleep Efficiency: -1 (-2.8 to 0.9), Total Sleep Time: 3.3 (-19.8 to 26.4)]. Conclusion: The PROMIS sleep measures are reliable, valid, and responsive to improvement in restorative sleep in women with fibromyalgia. Clinical Trial Registration: www.ClinicalTrials.gov, identifier: NCT01888640.

Keywords: patient reported outcome measures; psychometrics; reproducibility of results; sleep wake disorders; transcutaneous electric nerve stimulation.

Conflict of interest statement

KS serves as a consultant for Pfizer Consumer Health and Novartis Consumer Healthcare/GSK Consumer Healthcare, had an active grant from the American Pain Society/Pfizer, and receives royalties from IASP Press. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Copyright © 2021 Chimenti, Rakel, Dailey, Vance, Zimmerman, Geasland, Williams, Crofford and Sluka.

Figures

Figure 1

Flow diagram for participants in the trial using the Consolidated Standards of Reporting Trials guidelines.

Figure 2

Effects of TENS on (A) PROMIS Sleep Disturbance, (B) Pittsburg Sleep Quality Index (PSQI), (C) PROMIS Sleep-Related Impairment, and (D) PROMIS Fatigue. Statistically significant effects of time (*0 to 4 weeks, and +0 to 8 weeks) and interaction effects of treatment*time(∧) are indicated. For all outcome measures, scores higher than the horizontal line are indicative of nonrestorative sleep.