Op0295-hpr the effect of group-based cognitive behavioural therapy for insomnia in patients with rheumatoid arthritis: a randomised controlled trial

BackgroundInsomnia is highly prevalent in patients with rheumatoid arthritis (RA) and may exacerbate symptoms and burdens, such as fatigue, depressive symptoms, and pain1. Cognitive behavioural therapy for insomnia (CBT-I) has been shown to produce positive effects on sleep in other clinical populations2,3. However, CBT-I has not previously been investigated in patients with RA.ObjectivesThe primary objective was to compare the effect of nurse-led group-based CBT-I to usual care on sleep efficiency, measured by polysomnography (PSG) immediately after the intervention (i.e. seven weeks after baseline) in patients with RA. Secondary objectives included comparing the longer-term effect of CBT-I on sleep and RA-related outcomes at 26 weeks’ follow-up.MethodsIn a randomised controlled trial, using a parallel group design, the experimental intervention was six weeks’ CBT-I; the control comparator was usual care. CBT-I was delivered face-to-face by a CBT-I trained nurse. The primary analyses were based on the intention-to-treat (ITT) population; missing data were statistically handled using repeated-measures linear mixed effects models adjusted for the level at baseline.ResultsThe ITT population consisted of 62 patients (89% women), with an average age of 58 years (SD 10), DAS28-CRP of 3.4 (SD 1.0), Insomnia Severity Index (ISI) score of 18.9 (SD 4.4) and median Patient Global Assessment score of 55 (IQR 28;71).When primary outcome was measured by PSG at week seven, sleep efficiency was 88.7% in the CBT-I group, compared to 83.7% in the control group (difference: 5.0 [95% CI -0.4 to 10.4]; p=0.068) (See Table 1). Secondary outcomes measured by PSG had not improved at week 26 either. However, for all secondary sleep and RA-related patient-reported outcomes, there were statistically highly significant differences between CBT-I and usual care e.g. insomnia (ISI: -9.8 [95% CI -11.8 to -7.9]), RA impact of disease (RAID: -1.4 [95% CI-1.9 to -0.80]) and Patient Global Assessment (-13.0 [95% CI -20.9 to -5.1]) at 26 weeks’ follow-up.Table 1.Primary and key secondary outcomes at week 7 and week 26, and differences between treatment groups (based on the ITT population)CBT-I n=31Usual care n=31Difference between groups (95% CI)P-valueAt week 7Sleep efficiency (PSG, %)188.7 (1.8)83.7 (2.0)5.0 (-0.4 to 10.4)0.068At week 26Sleep efficiency (PSG, %)84.8 (1.9)86.3 (2.0)-1.5 (-7.0 to 3.9)0.577Total sleep time (PSG, minutes)376.5 (11.8)394.6 (12.8)-18.1 (-52.5 to 16.4)0.302Sleep onset latency (PSG, minutes)14.2 (2.2)10.0 (2.4)4.2 (-2.2 to 10.7)0.197Wake after sleep onset (PSG, minutes)52.1 (10.7)41.5 (11.6)10.6 (-20.7 to 41.9)0.505Insomnia severity (ISI 0-28)27.6 (0.7)17.4 (0.7)-9.8 (-11.8 to -7.9)<0.0001Sleep quality global (PSQI 0-21)35.9 (0.5)11.1 (0.5)-5.2 (-6.6 to -3.8)<0.0001Fatigue (BRAF-MDQ 0-70)424.0 (1.4)36.4 (1.5)-12.4 (-16.5 to -8.4)<0.0001RA impact of disease (RAID 0-10)54.2 (0.20)5.5 (0.20)-1.4 (-1.9 to -0.80)<0.0001Depressive symptoms (HADS-D 0-21)63.8 (0.5)6.5 (0.5)-2.7 (-4.1 to -1.3)<0.0001Values are reported as least squares means (standard errors) by group, while the differences between groups are reported with 95% confidence intervals.1Polysomnography, 2Insomnia Severity Index, 3Pittsburgh Sleep Quality Index, 4Bristol Rheumatoid Arthritis Fatigue - Multidimensional Questionnaire, 5Rheumatoid Arthritis Impact of Disease, 6Hospital Anxiety and Depression Scale - Depression.ConclusionNurse-led, group-based CBT-I for two hours per week for six weeks, did not improve objectively measured sleep efficiency or any other outcomes measured by PSG. However, CBT-I showed long-term improvement on patient-reported outcomes such as fatigue, impact of disease, depression, pain, and Patient Global Assessment – a finding that could have important clinical implications.References[1]PMID: 25620673[2]PMID: 16804151[3]PMID: 26434673AcknowledgementsWe thank the participants for their time and commitment and the patient research partners for valuable insight into the process and content of the trial.Disclosure of InterestsNone declared