BRIEF COMMUNICATION Potentially inappropriate prescribing among Australian veterans and war widows/widowers E. E. Roughead, B. Anderson and A. L. Gilbert Sansom Institute, School of Pharmacy and Medical Sciences, University of South Australia, Adelaide, Australia Key words Beers criteria, adverse drug event, potentially inappropriate medicine, elderly, prescribing. Correspondence E. E. Roughead, Quality use of Medicines and Pharmacy Research Centre; Sansom Institute, School of Pharmacy and Medical Sciences, University of South Australia, GPO Box 2471, Adelaide 5001, South Australia, Australia. Email: libby.roughead@unisa.edu.au Received 30 April 2006; accepted 15 July 2006. doi:10.1111/j.1445-5994.2007.01316.x Abstract This study examined the extent of potentially inappropriate medicine, as de- fined by explicit criteria, dispensed to Australian veterans using the Repatria- tion Pharmaceutical Benefits Scheme Pharmacy Claims database. Twenty-one per cent of the 192 363 veterans aged 70 years, with an eligible gold card, were dispensed at least one potentially inappropriate medicine in the first 6 months of 2005. Long-acting benzodiazepines, amitriptyline, amiodarone, oxybutynin and doxepin were the medicines most commonly implicated. Strategies to support quality prescribing of medicines to the elderly must include a focus on these medicines. Several of international studies have now documented high levels of potentially inappropriate prescribing in the elderly, ranging from 17 to 25%. 1–3 Potentially inappro- priate medication prescribing in nursing home residents or frail elderly is even higher, with most estimates ranging from 20 to 50%. 4,5 There is some variability in prevalence estimates, however, in general studies with lower esti- mates have used stricter criteria, 1,2 shorter time frames 3 or patient self-completed surveys. 3 The use of medicines identified as potentially inappro- priate in the elderly does appear to have the potential for increased harm. An observational study found that expo- sure to potentially inappropriate medicines in individuals is associated with increased hospitalization (odds ratio (OR) = 1.27, P = 0.002) and death (OR = 1.28, P = 0.01) compared with those not on these medicines after controlling for age, gender and nursing home stay. 4 Another study showed higher mean numbers of inpatient, outpatient and emergency department visits after control- ling for gender, comorbidities and numbers of medicines taken. 4 Self-reported poorer health status was also found for those on potentially inappropriate medicines after controlling for age, gender, comorbidities and income. 5 Intermittent exposure to potentially inappropriate medi- cines was also associated with a higher odds of death (OR = 1.89, P < 0.001). 4 In identifying potentially inappropriate medication use, the majority of studies have used explicit criteria, most commonly the Beers criteria, which have been developed using a consensus method by a panel of US clinicians with expertise in psychopharmacology, pharmacoepidemiol- ogy, clinical geriatric pharmacology and clinical geriatric medicine. 6 The criteria identify medications that should be generally avoided or doses, frequencies or durations of medication, that should generally be avoided in the elderly. 7–9 A Canadian set of criteria, developed by McLeod et al., has also been used. 5,10 The extent of potentially inappropriate medication use in the elderly has had only limited study in Australia. A study of 1000 persons at high risk of medication misad- venture found one in four was considered to be using the wrong or an inappropriate medicine. 11 Another Australian Funding: This research was supported by as research grant from the Australian Government, Department of Veteran Affairs. Potential conflicts of interest: None 402 ª 2007 The Authors Journal compilation ª 2007 Royal Australasian College of Physicians Internal Medicine Journal 37 (2007) 402–405