Assessment of potentially inappropriate medicine prescribing for elderly patients in the South African private health sector
Abstract
The aim of this study was to investigate the medicine prescribing patterns for elderly patients in the private health sector of South Africa utilising different screening tools for potentially inappropriate prescribing.
The study consisted of a literature review and an empirical investigation. The literature review entailed an overview of the concept and prevalence of potentially inappropriate prescribing (PIP), factors influencing prescribing for the elderly and a comparison of different screening tools for PIP in elderly patients. The empirical investigation followed a quantitative, descriptive, cross-sectional design to study inappropriate prescribing for elderly patients’ ≥65 years using medicine claims data from a one-year period (1 January – 31 December 2013). PIP was assessed by applying the 2012-Beers criteria list for potentially inappropriate medicine items (PIMs) and the Mimica Matanović and Vlahović-Palčevski comprehensive protocol for potentially clinically important drug-drug interactions (DDIs).
Variables were characterised using 95% confidence intervals, descriptive statistics such as proportions/ratios for categorical variables, and means and standard deviations for continuous variables. Independent two-sample t-tests were used to assess the statistical significance of mean differences between men and women, whereas one-way between-group analysis of variance (ANOVA) with post-hoc comparisons using the Tukey HSD test was conducted to determine significant differences between the means of two or more independent groups. The Chi-square test was done to determine the association between categorical variables. Practical significance was assessed using Cohen's d-value (for mean differences between groups), and Cramér’s V statistic (for associations between categorical variables). Statistical analyses were performed using SAS Software, version 9.3. Medicine claims data for 103 420 patients, mean age 74.1 ± 6.7 years, 57.1% female, were included in the analyses. Only 102 of the 143 2012-Beers criteria list items were available in South Africa at the time of the study. Application of the criteria list identified 562 852 PIMs in 71 206 patients (68.9%). The most common PIMs were: oestrogen (12.4%), meloxicam (7.3%), amitriptyline and combinations thereof (6.5%), diclofenac (6.4%), ibuprofen (6.1%), alprazolam (5.3%), meprobamate and combinations thereof (5.0%), insulin (3.3%), amiodarone (3.1%) and doxazosin (2.6%). There were statistically significantly more PIMs prescribed to women than to men (1.9:1; P<0.0001), although this difference was not practically significant (Cramér’s V = 0.0559). General practitioners were the group that prescribed the highest number of PIMs.
Only 65 of the 70 Mimica Matanović and Vlahović-Palčevski comprehensive protocol for potentially clinically important drug-drug interactions were available in South Africa at the time of the study. Application of the protocol identified a total of 331 655 (7.8%) DDIs among 912,712 prescriptions. Women experienced proportionally more DDIs than men (62.6% vs. 37.4%) (P < 0.0001, Cramér’s V = 0.0491). A mean 0.36 ± 0.7 (95% CI, 0.36-0.37) DDIs were encountered per prescription. There was no practically significant difference between sexes (P < 0.0001, Cohen’s d-value = 0.05) or age groups (F(3, 235 869) = 112.38, P < 0.0001; Cohen’s d-value < 0.05) with regard to the mean number of DDIs per prescription. The most common DDIs were between several drugs acting on the central nervous system (30.6%), antihypertensives and nonsteroidal anti-inflammatory drugs (23.5%), diuretics and nonsteroidal anti-inflammatory drugs (8.3%), angiotensin-converting enzyme inhibitors and potassium supplements (4.9%), and nonsteroidal anti-inflammatory drugs/aspirin and corticosteroids (4.8%).
Inappropriate medicine item prescribing and drug-drug interactions were common among the patients in the study population. Female patients received more PIMs and had a higher prevalence of potentially serious DDIs than their male counterparts. The prevalence of PIMs was the highest among patients between the ages of 72 and 78 years, whereas the prevalence of potentially serious DDIs was the highest in patients older than 78 years. Drug groups of concern included sedative hypnotics, antihypertensives, antidepressants and NSAIDs. The Beers criteria list and the Mimica Matanović and Vlahović-Palčevski drug-drug interaction list were useful to determine inappropriate prescribing patterns; however, there is a clear need for custom-made screening instruments for determining potentially inappropriate prescribing in South African elderly patients.
Collections
- Health Sciences [2061]