Zopiclone Use and Risk of Fractures in Older People: Population-Based Study

• Published in: Journal of the American Medical Directors Association Vol. 18; no. 4; p. 368.e1
• Main Authors: Nishtala, Prasad S, Chyou, Te-Yuan
• Format: Journal Article
• Language: English
• Published: United States 01.04.2017
• Subjects: Aged; Aged, 80 and over; Azabicyclo Compounds - administration & dosage; Cohort Studies; Female; Hip Fractures - epidemiology; Hip Fractures - etiology; Humans; Hypnotics and Sedatives - administration & dosage; Male; New Zealand - epidemiology; Piperazines - administration & dosage; Population Surveillance; hypnotics; older people; case-crossover; Benzodiazepine; fracture; pharmacoepidemiology
• ISSN: 1538-9375
• DOI: 10.1016/j.jamda.2016.12.085

The primary objective was to evaluate the risk of fractures associated with use of zopiclone, a new Z-hypnotic, in a population-based cohort of older people in New Zealand (NZ) in a real-world setting. The secondary objective was to examine a nonlinear relationship with age and risk for fractures in zopiclone users. Population-based cohort study with a case-crossover design. A nationwide representative sample of New Zealanders aged 65 years and older sourced from the pharmaceutical collections and hospital discharges. 74,787 older individuals with a first-time fracture between January 1, 2005, and December 31, 2015, were analyzed using a case-crossover design for fracture risk with zopiclone use. Prescription records (2005-2014) of zopiclone were sourced from NZ Pharmaceutical Collections (Pharms). The first-time coded diagnosis of fracture was extracted from the National Minimal Datasets. Datasets were linked by a unique patient identifier to set up case-crossover designs. Relative risks (RRs) of fracture associated with zopiclone was calculated using conditional logistic regression. A varying-coefficient conditional logistic model was employed to examine the influence of age as a risk factor for fractures. The risk of fracture associated with zopiclone is higher [RR = 1.45, 95% confidence interval (CI) = 1.37-1.55], compared to non-use. The increased risk of fracture associated with zopiclone [adjusted relative risk (ARR) = 1.36, 95% CI = 1.28-1.45] remained significant after adjusting for concomitant use of alpha blockers, antipsychotics, beta blockers, benzodiazepines, and tricyclic antidepressants. The varying coefficient model showed that the risk of fracture increases significantly and monotonically with age. The results support that the magnitude of the risk of fracture is higher with use of zopiclone compared to non-use. Prescribers must be aware that the relationship between age and fracture risk is nonlinear, and the oldest old are highly vulnerable to fractures with zopiclone use.