Inappropriate prescribing defined by STOPP and START criteria and its association with adverse drug events among hospitalized older patients: A multicentre, prospective study

OBJECTIVES: To provide baseline information on inappropriate prescribing (IP), and to evaluate whether potentially inappropriate medications (PIMs), as defined by STOPP (Screening Tool of Older Persons' potentially inappropriate Prescriptions) criteria, were associated with preventable adverse drug events (ADEs) and/or hospitalization.

METHODS: We prospectively studied older patients (n = 301) admitted to three urban, public-funded hospitals. We scrutinized their medical records and used STOPP-START (Screening Tool to Alert Prescribers to Right Treatment) criteria to determine PIM and potential prescribing omissions (PPO) respectively- together these constitute IP. Prescriptions with PIM(s) were subjected to a pharmacist medication review, aimed at detecting cases of ADE(s). The vetted cases were further assessed by an expert consensus panel to ascertain: i) causality between the ADE and hospitalization, using, the World Health Organization Uppsala Monitoring Centre criteria, and, ii) whether the ADEs were avoidable (using Hallas criteria). Finally, percentages of PIM-associated ADEs that were both preventable and linked to hospitalization were calculated.

RESULTS: IP prevalence was 58.5% (n = 176). A majority (49.5%, n = 150) had moderate to severe degree of comorbidities (Charlson Comorbidity Index score ≥ 3). Median age was 72 years. Median number of medications was 6 and 30.9% (n = 93) had ≥8 medications. PIM prevalence was 34.9% (117 PIMs, n = 105) and PPO 37.9% (191 PPOs, n = 114). Most PIMs and PPOs involved overuse of aspirin and underuse of both antiplatelets and statins respectively. With every increase in the number of medications prescribed, the likelihood of PIM occurrence increased by 20%, i.e.1.2 fold (OR 1.20, 95% CI: 1.1-1.3). Among the 105 patients with PIMs, 33 ADEs (n = 33); 31 ADEs (n = 31) considered "causal" or "contributory" to hospitalization; 27 ADEs (n = 27) deemed "avoidable" or "potentially avoidable"; and 25 PIM-associated ADEs, preventable, and that induced hospitalization (n = 25), were identified: these equated to prevalence of 31.4%, 29.5%, 25.7%, and 23.8% respectively. The most common ADEs were masked hypoglycemia and gastrointestinal bleed. With every additional PIM prescribed, the odds for ADE occurrence increased by 12 folds (OR 11.8, 95% CI 5.20-25.3).

CONCLUSION: The majority of the older patients who were admitted to secondary care for acute illnesses were potentially exposed to IP. Approximately a quarter of the patients were prescribed with PIMs, which were plausibly linked with preventable ADEs that directly caused or contributed to hospitalization.