Day-to-day variation in sleep duration is associated with increased all-cause mortality.

STUDY OBJECTIVES There is paucity of data on association between day-to-day variation in sleep pattern and all-cause mortality. We aim to investigate if day-to-day variation in sleep duration and onset of sleep are associated with cardiovascular and all-cause mortality. METHODS Data belonging to 388 unique subjects from the Midlife in the United States (MIDUS) 2 Biomarker study (2004-09) was used in our study. Sleep onset, duration, sleep-wake cycles were collected for 7 consecutive days using the Actiwatch device. Sleep irregularity was assessed using mean and standard deviations in sleep duration and time of onset of sleep over 7 days. Cox proportional regression analysis and Fine and Gray subdistribution method were used with all-cause and cardiovascular mortality, respectively. RESULTS Over a median of 8.6 years of follow up, 37 patients died including 10 deaths due to cardiovascular causes. There was no statistically significant increase in cardiovascular mortality with variation in sleep duration highest vs lowest tertile: HR 4.00(0.45-35.48, p 0.21). However, increased all-cause mortality was seen with the highest vs lowest tertile HR 3.99(1.33-11.94, p 0.01). Multivariable model adjusting for confounders had higher all-cause mortality with increased sleep duration variation, highest vs lowest tertile HR 4.85(1.52-15.49, P < 0.01). CONCLUSIONS Day to day variation in sleep duration is associated with increased all-cause mortality but not cardiovascular mortality after adjusting for mean sleep duration, inflammation, diabetes, age, BMI, renal function and blood pressure. Irregularity in the onset of sleep is not associated with all-cause mortality or cardiovascular mortality.