Assessing sleep in fibromyalgia: investigation of an alternative scoring method for the Jenkins Sleep Scale based on data from randomized controlled studies

Objectives. To investigate the validity of a rescored version of the Jenkins Sleep Scale (JSS) to assess the extent of possible bias of a 4-week recall period in assessing sleep in patients with fibromyalgia. Methods. A rescoring algorithm of the JSS was developed. The psychometric properties of the rescored JSS were examined using blinded, observed data from a Phase 2 trial (n=195) in subjects with fibromyalgia. In addition, data from two Phase 3, randomised, controlled trials (n=1,121) in subjects with fibromyalgia were used to further validate the rescored JSS by conducting correlation analyses with other assessments expected to correlate with sleep. These included fatigue and tiredness items from the Fibromyalgia Impact Questionnaire (FIQ), the Functional Outcomes of Sleep Questionnaire (FOSQ), and the Short Form-36 (SF-36 (TM)) Vitality scale. Results. Construct validity of the rescored JSS was found to be acceptable, with an internal consistency reliability of alpha=0.70. Test-retest reliability on stable subjects, defined using the FIQ total score, was also acceptable (ICC=0.70). Moderate to high correlations (Pearson r>0.66) were found with two FIQ items, addressing fatigue and non-restorative sleep, and the SF-36 (TM) Vitality scale; correlations with the original JSS were similar. Both JSS versions were found to be responsive (p<0.0001), and the rescored version accounted for 90% of the variance captured in the original version. Conclusion. These results showed the rescored JSS performed similarly to the original scale, suggesting the original scale's 4-week recall period did not introduce substantial bias in capturing the experience of fibromyalgia-related sleep disturbances.