Background Poor sleep quality in older people is common and positively associated with morbidity and mortality. In particular, recent studies have suggested that sleep disturbance may be associated with impaired physical performance (PP) in American older adults. We attempted to replicate these findings in a population of UK community-dwelling older adults.

Methods The study population comprised 443 men and women aged 71–80 years from the Hertfordshire Cohort Study, UK. Sleep quality was assessed using the validated Pittsburgh Sleep Quality Index (PSQI). The PSQI is a 10-item questionnaire that assesses a variety of factors related to sleep quality, and results in a global score where >5 is indicative of poor sleep quality. PP was assessed using the Guralnik short physical performance battery (SPPB), a timed up and go (TUG) test and grip strength.

Results Completed questionnaires were returned by 373 (84%) participants. 153 (41%) of these participants had a PSQI score of >5, indicating poor sleep. Men, but not women, who reported poor sleep had a significantly lower SPPB score (≤9) after adjustment (age, smoking, alcohol, social class and BMI) (OR 2.42, 95% CI 1.14, 5.14, p = 0.022). This association was robust to additional adjustment for number of co-morbidities and number of medications. Sleep quality was associated with TUG time in men but not women. Men who had poorer sleep quality (PSQI >5) had an average increase (poorer performance) in TUG z-score of 0.30 compared with those with good sleep quality (PSQI ≤5) (95% CI 0.01, 0.59, p < 0.042). This association was attenuated by adjustment for number of co-morbidities and number of medications. Sleep quality was associated with grip strength in women but not men; women with poor sleep quality had, on average, a grip strength of 2.14kg less than those with good sleep quality (95% CI -3.89, -0.38, p = 0.017). This association became non-significant after adjustment for number of medications.

Conclusion We found further evidence in support of an association between poor sleep quality and worse PP in older adults. However, differences were observed between men and women. Our results suggest that co-morbidities and medication use may explain some of these observations. Further research into the association between sleep and PP is warranted.