378 Use of a Clinician’s Global Impression of Severity scale to measure insomnia severity in Alzheimer’s disease-dementia patients

Abstract

\n \n \n Full montage polysomnography (PSG) is the gold standard for the objective evaluation of sleep but is time consuming and inaccessible to most clinicians. A Clinician’s Global Impression of Severity (CGI-S) scale can be used in clinical practice to provide a subjective assessment of patients’ insomnia severity. However, the utility of a CGI-S scale for assessing insomnia in patients with Alzheimer’s disease (AD)-dementia is not well understood. In a recent Phase III randomized, placebo-controlled clinical trial (NCT02750306), patients on suvorexant with AD-dementia and insomnia showed improvements in both PSG total sleep time (TST) and CGI-S scores. We conducted additional analyses to examine the association between CGI-S and PSG-TST to inform on the possible use of a CGI-S scale to assess insomnia severity in patients with AD-dementia in real-world settings.\n \n \n \n Patients (N=285) met clinical diagnostic criteria for both probable mild-to-moderate AD-dementia and insomnia. The primary endpoint was change-from-baseline in overnight PSG-TST at Week-4. A single-item CGI-S rating of insomnia with responses of 1 (normal, not ill at all) to 7 (among the most extremely ill patients) was completed by a trained rater at baseline and after 2 and 4 weeks. CGI-S was an exploratory endpoint. Post-hoc correlational analyses and analyses of distribution of change-from-baseline to Week-4 in CGI-S response categories were performed.\n \n \n \n Pearson correlation indicated a significant association at baseline between PSG-TST and CGI-S (r=-0.18, nominal p=0.004). A correlation of change-from-baseline to Week-4 also indicated an association between PSG-TST and CGI-S (r=-0.24, nominal p<.0001). The distribution of change in CGI-S response category results at Week-4 showed that, compared to placebo, numerically less patients on suvorexant remained stable or worsened by >1 response category (21.8% vs. 29.4%, respectively) and numerically more improved by ≥1 response category (73.3% vs. 67.9%, respectively).\n \n \n \n Our findings suggest that a CGI-S scale may be a useful tool for assessing insomnia severity in mild-to-moderate AD-dementia patients. Future studies with these patients are needed to determine the utility of a CGI-S scale in real-world settings.\n \n \n \n This study was funded by Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc., Kenilworth, NJ, USA\n