Aim The present study examined relationships among older adults' sleep duration (1–5 h and ≥10 h vs 6–9 h), sleep medication use, sleep quality and healthcare use. Methods The 2013 USA National Health Interview Survey provided data on a sample of 7196 community‐dwelling individuals aged ≥65 years. Bivariate analyses and multivariate regression analyses were used to examine study questions. Results Of the sample, 85% slept 6–9 h on average, 8% slept 1–5 h and 7% slept ≥10 h. One‐sixth of optimal sleepers, one‐quarter of short sleepers and one‐fifth of long sleepers used sleep medications at least once in the preceding week. Medication intake of one to two times, as opposed to non‐use, was associated with a greater risk of short than optimal sleep (RRR 1.60, 95% CI 1.03–2.48). Short sleep was positively, but long sleep was negatively associated with trouble falling asleep and staying asleep. Any medication use was positively associated with trouble falling asleep and staying asleep. Short sleep was associated with greater odds of emergency department visits (OR 1.36, 95% CI 1.06–1.77), whereas long sleep was associated with greater odds of an overnight hospital stay (OR 1.48, 95% CI 1.04–2.11) and home care use (OR 2.46, 95% CI 1.37–4.41). Medication intake one to two times was associated with greater odds of emergency department visits (OR 1.48, 95% CI 1.11–1.98). Conclusions Sleep medication use does not appear to promote sleep health. Both short and long sleep durations are associated with a higher likelihood of healthcare use. Clinicians need to examine older adults' sleep health and medication use, and their effects on healthcare use. Geriatr Gerontol Int 2017; 17: 545–553.