Agreement between sleep diary and actigraphy in a highly educated Brazilian population

This study evaluated the agreement between a sleep diary and actigraphy on the assessment of sleep parameters among school teachers from Brazil.

A total of 163 teachers (66.3% women; aged 45 ± 9 years) filled out a sleep diary and wore a wrist actigraph device for seven consecutive days. Data were collected from August 2014 to March 2015 in Londrina, a large city in southern Brazil. Intraclass correlation coefficients (ICC) and Pearson correlation coefficients (r) were used to compare self-reported and actigraphic data.

Self-reported total sleep time (TST), sleep onset latency (SOL), and sleep efficiency were higher than measured by actigraphy (mean difference: 22.6 ± 46.9 min, 2.6 ± 13.3 min, and 7.3± 5.7%, respectively). Subjective total time in bed (TIB) and wake-up time were lower than measured by actigraphy (mean difference: -10.7 ± 37.6 and -19.7 ± 29.6, respectively). Moderate or good agreement and correlation were found between the sleep diary and the actigraphic data for TST (ICC = 0.70; r = 0.60), TIB (ICC = 0.83; r = 0.73), bedtime (ICC = 0.95; r = 0.91), sleep start time (ICC = 0.94; r = 0.88), and wake-up time (ICC = 0.87; r = 0.78). However, SOL (ICC = 0.49; r = 0.38) and sleep efficiency (ICC = 0.16; r = 0.22) showed only fair or poor agreement and correlation.

In this highly educated population, the sleep diary and the actigraphy showed moderate or good agreement to assess several sleep parameters. However, these methods seemed to measure different dimensions of sleep regarding sleep onset latency and efficiency. These findings moderately varied according to the individual's subjective sleep quality.