Prevalence of potentially inappropriate prescribing in an acutely ill population of older patients admitted to six European hospitals.

Purpose: Potentially inappropriate prescribing is common in older people presenting to hospital with acute illness in Ireland. The aim of this study was to determine if this phenomenon is unique to Ireland or whether it is a more widespread problem in hospitals across Europe.

Methods: Prospective data were collected from 900 consecutive older patients admitted to six university teaching hospitals (150 patients per centre) in Geneva (Switzerland), Madrid (Spain), Oostende (Belgium), Perugia (Italy), Prague (Czech Republic) and Cork (Ireland). Age, gender, comorbidity, cognitive status, prescription medicines taken before admission and baseline haematological, biochemical and electrocardiographic data were recorded. STOPP and Beers' criteria were applied to detect potentially inappropriate medicines (PIMs). START criteria were applied to detect potentially inappropriate prescribing omissions (PPOs).

Results: The overall PIM prevalence rate was 51.3% using STOPP criteria, varying from 34.7% in Prague to 77.3% in Geneva, and 30.4% using Beer's criteria, varying from 22.7% in Prague to 43.3% in Geneva. Using START criteria, the overall PPO prevalence rate was 59.4%, ranging from 51.3% in Cork to 72.7% in Perugia. Polypharmacy predicted the presence of PIMs using STOPP criteria [with >10 medications: odds ratio (OR)  7.22, 95% confidence interval (CI) 4.30-12.12, p < 0.001] and Beers' criteria (with >10 medications: OR 4.87, 95% CI 3.00-7.90, p < 0.001). Increasing co-morbidity (Charlson Index ≥2) and age ≥85 years significantly predicted PPOs.

Conclusion: Potentially inappropriate drug prescribing and the omission of beneficial drugs are highly prevalent in acutely ill hospitalized older people in six European centres.