Benzodiazepine and z-hypnotic prescribing for older people in primary care: a cross-sectional population-based study.

BACKGROUND Overall prescribing of benzodiazepines and z-hypnotics (B&Zs) has slowly reduced over the past 20 years. However, long-term prescribing still occurs, particularly among older people, and this is at odds with prescribing guidance. AIM To compare prescribing of B&Zs between care home and non-care home residents ≥65 years old. DESIGN AND SETTING Cross-sectional population-based study in primary care in Scotland. METHOD National patient-level B&Z prescribing data, for all adults aged ≥65 years, were extracted from the Prescribing Information System (PIS) for the calendar year 2011. The PIS gives access to data for all NHS prescriptions dispensed in primary care in Scotland. Data were stratified by health board, residential status, sex, and age (65-74, 75-84, and ≥85 years). To minimise disclosure risk, data from smaller health boards were amalgamated according to geography, thereby reducing the number from 14 to 10 areas. RESULTS A total of 17% (n = 879 492) of the Scottish population were aged ≥65 years, of which 3.7% (n = 32 368) were care home residents. In total, 12.1% (n = 106 412) of older people were prescribed one or more B&Z: 5.9% an anxiolytic, 7.5% a hypnotic, and 1.3% were prescribed both. B&Zs were prescribed to 28.4% (9199) of care home and 11.5% (97 213) of non-care home residents (relative risk = 2.88, 95% CI = 2.82 to 2.95, P<0.001). Estimated annual B&Z exposure reduced with increasing age of care home residents, whereas non-care home residents' exposure increased with age. CONCLUSION B&Zs were commonly prescribed for older people, with care home residents approximately three times more likely to be prescribed B&Zs than non-care home residents. However, overall B&Z exposure among non-care home residents was found to rise with increasing age.