Kripke · BMJ open 2012 · matched cohort study · n=34,205

Hypnotics' association with mortality or cancer: a matched cohort study.

Cited 405 times in the scientific literature.

Level 3 - non-randomized controlled study

Non-randomized matched cohort study using electronic medical records

PubMed 22371848 · doi:10.1136/bmjopen-2012-000850 · record verified 2026-08-31

What was done

A 1:2 matched cohort survival analysis was conducted using longitudinal electronic medical records from a large integrated US health system (2002–2007). The cohort included 10,529 patients (mean age 54 years) who received hypnotic prescriptions and 23,676 matched controls with no hypnotic prescriptions, followed for an average of 2.5 years. Cox proportional hazards models adjusted for age, gender, smoking, BMI, ethnicity, marital status, alcohol use, and prior cancer, with exact matching across up to 116 strata for 12 comorbidity classes.

What was found

Patients prescribed hypnotics had elevated mortality hazards compared to unprescribed controls in a dose-response fashion: - 0.4–18 doses/year: HR 3.60 (95% CI 2.92 to 4.44) - 18–132 doses/year: HR 4.43 (95% CI 3.67 to 5.36) - >132 doses/year: HR 5.32 (95% CI 4.50 to 6.30) Hazard ratios remained elevated in separate analyses for zolpidem, temazepam, eszopiclone, zaleplon, other benzodiazepines, barbiturates, and sedative antihistamines. Hypnotic use in the upper third was associated with elevated incident cancer (HR 1.35, 95% CI 1.18 to 1.55).

Why it matters

The study reports a substantial statistical association between hypnotic prescriptions and both all-cause mortality and incident cancer, even at low prescription volumes.

Limits

The observational design cannot establish causality, leaving substantial risk of confounding by indication or residual confounding from unmeasured disease severity and sleep disorder characteristics. Prescription records do not verify actual consumption of medications, and mean follow-up was relatively short (2.5 years).

Cited by