UPM Institutional Repository

Prevalence and predictors of potentially inappropriate medications among elderly patients attending government primary care clinics


Chandra Sejara Rao, Hemah Devi (2019) Prevalence and predictors of potentially inappropriate medications among elderly patients attending government primary care clinics. Masters thesis, Universiti Putra Malaysia.


Elderly patient may become victims of potentially inappropriate medication (PIM) when their drug interactions and effect of the drug on other underlying diseases are not being properly investigated during follow ups with medical officers. PIM has been defined as medication which are not suitable for patients based on age, laboratory findings, and medical history which may lead to further complication of health. Issues on double prescriptions from clinics and hospitals, lack of communication among doctors in both facilities, patient understanding and adherence to medication to regimes, unavailability of drugs prescribed by specialists at primary care and cost are important factors which could contribute to PIM. Aim of the study is determine prevalence and predictors of PIM among elderly patients attending government primary care clinics in Seremban district. Elderly patients aged 60 and above attending health clinics in Seremban district were recruited in this cross sectional study by using random sampling method. Elderly patients’ sociodemographic and clinical characteristics were obtained from patients’ prescriptions and medical database as patients present at pharmacy to collect medication. Prescribed medications were analysed by using Screening Tool of Older Persons’ Potentially Inappropriate Prescription (STOPP) criteria and were identified as PIM if the medication were included in STOPP with similar description. STOPP is a screening tool to measure incidences of PIM based on physiological system. Sociodemographic data and clinical characteristics association with PIM were studied to determine predictors of PIM. Data analysis was conducted by using IBM Statistical Package for Social Sciences Software (SPSS) version 22. Chi square method was used to determine the association among variables. The confidence interval was set at 95% and level of significance as p<0.05. Simple logistic regression was applied to determine the crude odd ratio and variables with p<0.25 were entered into multivariate logistic regression model to determine predictors for PIM. Majority of elderly patients had two types of illnesses (50.4%). Most patients were taking five medication for their illnesses (24.1%). Elderly patients aged above 70 were more likely to have PIM (AOR=1.721, 95% CI 1.316-1.974) compared to patients below the age of 70. Patients who were taking more than five prescribed medication were more likely to have PIM (AOR=1.628, 95% CI 1.152 to 1.850) compared to patients taking less than 5 prescribed medication. Patient with more than three number of illnesses were more likely to have PIM compared to patients with less than three number of illnesses. Patients with endocrine disease, renal disease and urogenital disease were more likely to have PIM in their regime compared to patients without the disease. Predictors of PIM based on this study were age (>70 years) number of prescribed medication (>5), number of illnesses (>3), endocrine disease, renal disease and urogenital disease. Prevalence of PIM is found to be high, 37% based on STOPP criteria. The finding of the study can be used as a baseline study on PIM among elderly in Malaysia primary care setting. By identifying PIM, many health related issues and medication errors among elderly patients can be reduced and resolved.

Download File

[img] Text

Download (969kB)

Additional Metadata

Item Type: Thesis (Masters)
Subject: Medical Errors
Call Number: FPSK(m) 2020 31
Chairman Supervisor: Profesor Sherina Mohd Sidik, PhD
Divisions: Faculty of Medicine and Health Science
Depositing User: Ms. Rohana Alias
Date Deposited: 09 May 2023 08:00
Last Modified: 09 May 2023 08:00
URI: http://psasir.upm.edu.my/id/eprint/103834
Statistic Details: View Download Statistic

Actions (login required)

View Item View Item