Psychometric evaluation of the PROMIS SD-SF-8b instrument in individuals experiencing vasomotor symptoms due to menopause

Abstract

BackgroundWomen with vasomotor symptoms (VMS) due to menopause frequently experience poor sleep quality. The Patient-Reported Outcomes Measurement Information System Sleep Disturbance – Short Form 8b (PROMIS SD-SF-8b) has been developed to assess sleep disturbance. The study objective was to use data from the fezolinetant SKYLIGHT 1 and 2 studies in individuals with VMS to assess the psychometric properties of the PROMIS SD-SF-8b.MethodsIndividuals (aged ≥ 40–≤65 years) with moderate-to-severe VMS (≥ 7 hot flashes/day) were enrolled. Besides PROMIS SD-SF-8b, eight other patient-reported outcome (PRO) measures were used for the psychometric evaluation. All the PRO assessments were completed at weeks 4 and 12 during the treatment period and most were completed at baseline. Psychometric analyses included factor analysis and reliability, construct validity, and sensitivity to change assessments. The within-patient threshold for a clinically meaningful change in sleep disturbance was derived.ResultsOverall, 1022 individuals were included from the SKYLIGHT 1 and 2 studies. Mean PROMIS SD-SF-8b total score at baseline was 26.80, which decreased to 22.68 at week 12, reflecting improved sleep disturbance. The confirmatory factor analysis supported the proposed PROMIS SD-SF-8b domain structure. Internal consistency was excellent, with Cronbach’s alpha values of 0.915 and 0.935 and a McDonald’s omega of 0.917. Item-to-item and item-total correlations were sufficient and moderate test-retest reliability was noted. The construct validity assessments showed that moderate Spearman rank correlations (r: 0.608 to 0.651) were observed between PROMIS SD-SF-8b total scores and measures of sleep disturbance and sleep-related impairment, and that significant differences were noted in the total scores across PRO categories. The responsiveness of PROMIS SD-SF-8b total scores was supported by the results from the correlations in change scores and comparisons of mean change scores by PRO categories. Statistically significant differences in mean scores were observed between responder and non-responder PRO groups. A PROMIS SD-SF-8b total score of 8 points was identified as the within-patient threshold to use to confirm a meaningful change in sleep disturbance.ConclusionsThe psychometric properties of the PROMIS SD-SF-8b support its use to measure sleep disturbance in women with VMS due to menopause.Trial registrationClinicalTrials.gov numbers: NCT04003155 and NCT04003142.