Abstract
PurposeTo investigate prevalence, independent associations, and variation over time of potentially inappropriate prescriptions in a population of older hospitalized patients.MethodsA longitudinal study using a large dataset of hospital admissions of older patients (≥ 70 years) based on an electronic health records cohort including data from 2015 to 2019. Potentially inappropriate medication (PIM) and potential prescribing omission (PPO) prevalence during hospital stay were identified based on the Dutch STOPP/START criteria v2. Univariate and multivariate logistic regression were used for analyzing associations and trends over time.ResultsThe data included 16,687 admissions. Of all admissions, 56% had ≥ 1 PIM and 58% had ≥ 1 PPO. Gender, age, number of medications, number of diagnoses, Charlson score, and length of stay were independently associated with both PIMs and PPOs. Additionally, number of departments and number of prescribing specialties were independently associated with PIMs. Over the years, the PIM prevalence did not change (OR = 1.00, p = .95), whereas PPO prevalence increased (OR = 1.08, p < .001). However, when corrected for changes in patient characteristics such as number of diagnoses, the PIM (aOR = 0.91, p < .001) and PPO prevalence (aOR = 0.94, p < .001) decreased over the years.ConclusionWe found potentially inappropriate prescriptions in the majority of admissions of older patients. Prescribing relatively improved over time when considering complexity of the admissions. Nevertheless, the high prevalence shows a clear need to better address this issue in clinical practice. Studies seeking effective (re)prescribing interventions are warranted.
Highlights
Prescription of medications for older people is complex due to multimorbidity, polypharmacy, and age-specific changes in pharmacokinetics and pharmacodynamics [1, 2]
In the present study, we aimed to investigate potentially inappropriate prescribing in older patients during hospital stay according to the STOPP/START criteria v2, including prevalence, independent associations, and trends over time, using a large dataset based on an electronic health records (EHR) cohort
Our results showed that number of medications, number of diagnoses, Charlson score, and age were independently associated with a higher prevalence of Potentially inappropriate medication (PIM) and potential prescribing omission (PPO)
Summary
Prescription of medications for older people is complex due to multimorbidity, polypharmacy, and age-specific changes in pharmacokinetics and pharmacodynamics [1, 2]. Most clinical guidelines contain treatment advice for single diseases and provide insufficient support for prescription of multiple drugs in older populations with multiple diseases [3, 4]. Inappropriate prescribing is common in older hospitalized patients with a PIM prevalence varying from 23 to 77% and PPO prevalence varying from. A longitudinal study of 44 general practices showed that the likelihood of potentially inappropriate prescribing is higher after hospitalization compared with before [10]. PIMs are associated with adverse patient-related outcomes, such as hospitalization and adverse drug events (ADEs) [11, 12]. In a meta-analysis pooling data from studies with a “new-user design” (implying that patients did not receive the treatment prior start of the study), PIMs were associated with increased mortality (RR = 1.59) [13]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
