Potentially inappropriate medications among elderly patients in community healthcare institutions in Beijing, China

Abstract

PurposeTo evaluate potentially inappropriate medications (PIMs) prevalence and predictors in community healthcare institutions (CHIs) for the elderly.MethodsWe conducted a retrospective observational study, deriving data of patients aged ≥60 from 66 CHIs in Beijing, 2014‐2018. The system of Criteria of PIM for Older Adults in China was applied to identify PIMs. The primary outcome was the prevalence of visits with at least one PIM; secondary outcomes were the frequency and rate per thousand visits of specific PIMs. We used descriptive analysis and generalized linear models to analyzed PIMs and the predictors, and marginal effects methods were applied to estimate the mean adjusted PIMs prevalence.ResultsOverall, 4 528 884 elderly patient visits from 2014 to 2018 were eligible for inclusion. A total of 719 757 PIMs were detected, with 14.1% of the visits contained at least one PIM. PIM prevalence was significantly correlated with age, number of prescribed medications and number of diagnoses. Overall, 6.0 per thousand elderly patients in CHIs were exposed to at least one high‐risk PIM, while 117.5 per thousand were exposed to at least one low‐risk PIM. In 2018, 20% of GPs were responsible for more than half of overall PIM visits.ConclusionPrescribing of PIMs for older adults is common in CHIs in China, especially for patients who are aged, having multiple medications and diagnostic diseases. Strategies should be developed to enhance prescribing quality for geriatric patients, with special targeting of doctors responsible for a high number of PIMs.

Prevalence and determinants of potentially inappropriate medications prescribing in elderly patients in Chinese communities

“…It has also been reported that 14.1% of elderly patients visiting community healthcare institutions in Beijing were taking at least one PIM (19). In our study, the prevalence of PIMs was 37.07% in the three communities studied based on the Chinese criteria, which was between the other Chinese studies listed above.…”

Prevalence of Polypharmacy and Potentially Inappropriate Medications Use in Elderly Chinese Patients: A Systematic Review and Meta-Analysis

“…24 articles were excluded as they did not report the PIMs or polypharmacy, meeting abstracts, and the population was not elderly. Finally, 8 studies ( Fu et al, 2020 ; Huang et al, 2020 ; Ma et al, 2020 ; Zheng et al, 2020 ; Tian et al, 2021a ; He et al, 2021 ; Li et al, 2021 ; Zhao et al, 2021 ) were included in the study ( Figure 1 ).…”

Protocol of a tailored educational intervention for general practitioners on potentially inappropriate medications among older patients at community healthcare institutions in Beijing, China: a cluster-randomised controlled trial

“…Our previous study, evaluating the PIMs utilisation in the community healthcare institutions in Beijing, showed a prevalence of 16.9% in 2018. 22 Early detection of PIMs is of central importance to prevent adverse outcomes and improve geriatric care in elderly patients. 23 Interventions, including medication review, 24 pharmaceutical interventions, 20 computerised alerts, 20 25 educational training, 26 have been effective measures to improve appropriate drug use in many countries and some also resulted in clinically significant improvements.…”

“…Based on our previous study, the current PIMs prevalence in Dongcheng CHSs is ~20% 22 and this is, therefore, assumed for the control arm. Based on a conservative estimate from our review, we expect our intervention to lead to at least an 8% relative reduction in the PIMs prevalence within CHSs.…”

“…We calculate the intracluster correlation coefficient (ICC) using the prescribing data from the sampled CHSs in 2018, collected during our previous study. 22 In order to detect an 8% or greater reduction in the PIMs prevalence with 90% power at a 5% significance level using two-sided Z-test (unpooled), adjusting for ICC of 0.02, we estimate that we need 10 clusters per arm and a minimum of 350 patient visits per cluster. SAS V.14.2 is applied to calculate the ICC.…”