P093\u2003Sleep characteristics and frailty in men: the influence of testosterone

Abstract

\n Background/Aims\u2002\n Previous studies have suggested an association between sleep disturbance and frailty. The mechanism is unknown, though it has been suggested that hormonal factors may play a role. The aim of this analysis was to determine the association between sleep duration, sleep quality and frailty in middle age and older European men and the influence of testosterone on these associations.\n Methods\u2002\n Men aged 40-79 years were recruited from population samples in eight European centres for participation in the European Male Age Study. Participants were interviewed and asked also to complete a questionnaire about sleep which included a question about average number of hours sleep per night and questions about sleep quality. Sleep quality was scored 0-20 with higher score indicating poor quality. A frailty index (FI) was constructed. Participants also had a single fasting morning venous blood sample taken and total testosterone measured using gas chromatography mass spectrometry. Using negative binomial regression we looked at the association between sleep duration, sleep quality and frailty (outcome; range 0-100) with the results adjusted for putative confounders and also testosterone level.\n Results\u2002\n 2,369 participants contributed data to the analysis. Mean sleep duration was 7.01 hours. 256 (10.7%) participants had sleep quality scores between 10 -14 and 101 (4.2%) participants’ scores were between 15 and 20. Mean (standard deviation) testosterone level was 16.4 (6.2) pmol/l and 14.7% of subjects had a low testosterone level (<10.5pmol/l). After adjustment for age, centre, body mass index, depression, pain, alcohol intake and smoking, compared to those with a sleep score of 0-4, those with scores of 10-14 and 15-20 were more likely to have a higher frailty index (RR = 1.37; 95% CI 1.25,1.49) and (RR = 1.57; 95% CI 1.38, 1.78) respectively. Further adjustment for testosterone did not influence the strength of these associations. Compared to those with sleep 6-8 hours, those with less than 5 hours had a significantly higher FI (RR = 1.16; 95% CI 1.06, 1.28) and also those with a longer sleep duration (9 hours or more) had a significantly increased FI (RR = 1.11; 95% CI 1.00, 1.23). Further adjustment for testosterone did not influence the strength of these associations.\n Conclusion\u2002\n Frailty is associated with impaired sleep quality and also sleep duration. The mechanism is unknown though cannot be explained by testosterone level.\n Disclosure\u2002\n S.D. Sharma: None. M.J. Cook: None. D. Vanderschueren: None. L. Antoniou: None. E. Gielen: None. F.C. Wu: None. T.W. O’Neill: None.