Abstract
Introduction
Actigraphy and sleep diaries are recommended by both the ICSD-3 and AASM. However, actigraphy provides a significant operational burden. We aimed to evaluate concordance between these tests.
Methods
60 sleep studies with concurrent Actigraphy and sleep diaries were sampled. Total sleep times (TST) and lights out/on times were compared using paired means t-tests and Welch t-tests. Diaries were defined as concordant or non-concordant based on clinician interpretation.
Results
47/60 had complete sleep diaries. The TST derived from sleep diaries were statistically different to actigraphy (7.67hrs;SD ± 1.78 vs 7.01hrs;SD ± 1.35hrs, p<.05). Differences in lights-out (55.8min;SD ± 56.4min) and lights-on times (42min;SD ± 42min) were not significant. 32 (68%) diaries were interpreted as concordant with a mean TST difference of 0.91hrs (SD ± 1.0hrs). Non-concordant diaries (NCD) had a mean TST difference of 2.51hrs (SD ± 1.39hrs). NCD TSTs were greater then actigraphy, except patients identified with insomnia who universally had lower diary sleep times. 31% of NCD had physician assessed delayed sleep phase disorder (DSPD), whilst restricted sleep was described in 62% of cases. Despite inconsistencies between sleep diary and actigraphy, patients in NCD cohort were able to accurately perceive TST on their polysomnogram (mean difference = 0.71hrs;SD ± 0.57hrs) with no difference to concordant diaries (p>.5).
Discussion
Sleep diaries demonstrated overall good concordance. Patients with NCD were more likely to be diagnosed with insomnia, DSPD or sleep restriction. To reduce operational burden, standalone sleep diaries could be used more routinely in clinic for initial evaluation. While concurrent actigraphy may be reserved for cases where the identified conditions could better inform clinical judgement.