Application of Beers 2012 and STOPP/START criteria : prevalence and health impact on inappropriate prescribing among older adults in Hong Kong

Abstract

The use of medicines has become an essential component in health care but can be harmful if they are inappropriately used. Most older adults suffer from multiple comorbidities and are prescribed with multiple medication. Together with age-related physiological changes, they are more vulnerable to negative health outcomes than younger adults. Use of appropriate medication has become a significant issue among older adults and screening tools such as the Beers and STOPP/START criteria have been used to identify potentially inappropriate prescribing (PIP) and provide guidance on safe medication practice. The aims of this research were to study the use of PIP, identified by the Beers and STOPP/START criteria, among Hong Kong (HK) older adults and investigate the impact of PIP on health outcomes. The objectives were to establish the content validity of the Beers 2012 and STOPP/START criteria according to HK setting; to determine the prevalence of potentially inappropriate medication (PIM)/potentially prescribing omission (PPO); to study the risk factors for PIM/PPO and to investigate the relationship between PIM/PPO, preventable drug-related admissions (PDRAs), hospital readmission and mortality. Using modified Delphi technique, a panel of 12 pharmacists rated that 84% and 94% of the indicators listed in the Beers 2012 and STOPP/START criteria respectively were relevant to HK practice. Together with the content validity index, this validation process has included 43 and 82 indicators of the Beers 2012 and STOPP/START criteria respectively in the HK-based criteria. Cross sectional data of 500 older adults who were admitted to a tertiary hospital were examined. The prevalence of PIM/PPO ranged from 25.0 to 50.2% when assessed by various criteria. A significant difference in PIM prevalence was observed between the Beers 2012 and STOPP criteria (p = 0.002). Moreover, fewer patients with PIM were identified by the HK-based criteria and significant differences in PIM prevalence were observed when compared to the original criteria. A positive, significant correlation between number of medication and PIM/PPO was observed, regardless of the criteria used. After adjusting for confounders, only polypharmacy was found to be the risk factor for PIM. A total of 53 admissions (10.6%) were considered as PDRAs. Among these PDRAs, 18 (34.0%) and 21 (39.6%) of them were involved with medication listed in the Beers 2012 and STOPP criteria respectively; however, no significant association was found between PDRAs and PIM. 44.5% of patients were readmitted within 30 days and over 80% of them were readmitted within 1 year. The readmission probabilities of patients with PIM were slightly higher than that of patients without PIM, although the difference was statistically insignificant. 212 patients (42.4%) and 356 patients (71.2%) died at 1-year and 3-year follow-up respectively. Again, no statistically significant difference was observed between patients with or without PIM for mortality. This study has shown that PIP among HK older adults was common and comparable to the rest of the world. However, it was not able to demonstrate the impact of PIP on outcome measures.