Medication use by ethnicity in the regions covered by Waitemata District Health Board

Abstract

Ethnic disparities in health care have been extensively documented around the world with patients in the minority ethnic groups receiving poorer quality of health care across a broad range of clinical settings. Many literatures have identified disparities in the healthcare of New Zealand. It is understandable due to the fact that New Zealand is a multi-ethnic society and each cultural group has their own ways towards how their health should be managed. Although many literatures have suggested there could be an ethnic difference in the quality of health care delivered, many of the published literature have not been designed to specifically address the significance of ethnicity as a covariant of medication use. Although previously published New Zealand literature in hypertension and Type II diabetes did indicate that there could be a tendency for Maori and Pacific Islanders to have underuse prophylactic medications, results were inconsistent. A critically appraised report presented in this thesis suggested only a few studies have specifically addressed the issue of ethnic differences in medication use. All appraised studies indicated a mix of results with some studies identified a greater medication use amongst Europeans than minority ethnic groups; while some displayed the opposite. Aim The aim of this research project is to investigate the disparity of medication used by ethnicity across the areas covered by the Waitemata District Health Board in order to address the issues of under-utilisation of medication by some population groups such as the Maori and Pacific communities if there is any. Objectives The principle objectives of this thesis were to:  Illustrate the use of medication in primary care settings for different ethnic groups with a major focus on Maori and Pacific Island patients;  Determine whether disparities in medication use exist between different ethnic groups; while controlling for the confounding factors (age, gender, and socioeconomic status). Methods This study involved two types of study methods of data collection; a retrospective crosssectional audit and a retrospective cohort study. A retrospective cross-sectional audit was used for the Diabetes Care Support Service datasets where medication use in Type II diabetes patients was investigated; whereas a retrospective cohort design was employed when extracting data from both the National Minimum Database (NMDS) and Pharms Data Warehouse (Pharms). The NMDS/Pharms dataset was utilised in examining oral hypoglycaemic agents and antihypertensive medications dispensed within six months after being discharged from hospitals; meanwhile DCSS dataset examined only the use of oral hypoglycaemic agents in Type II diabetes patients obtained from general practices. Standardised medication use ratio (MUR) with 95% confidence interval in both datasets was calculated with the effect of age and gender being adjusted for. The effect of deprivation was only adjusted for in the NMDS/Pharms dataset. Results The results of the study indicated that the likelihood of using both diabetes and antihypertensive medication across Maori and Pacific Island patients were not significantly different to the European/Others group (confidence intervals inclusive of 1.0). Oral hypoglycaemic agents use The majority of patients (over 70%) in the DCSS dataset across all ethnic groups had been using some form of oral hypoglycaemic agents during the 12 months prior to the audit. Similar findings were reported in the NMDS/Pharms dataset where during the six months after discharge, almost two thirds of the patients were using medication in managing their diabetes and approximately 50% were using oral hypoglycaemic agents. The proportion of metformin use was different across the two dataset. More than two thirds of the patients in the DCSS dataset were using metformin in managing their Type II diabetes, with Pacific Island patients being reported of using the most (70.3%). Metformin use was much lower in the NMDS/Pharms dataset with only 42.3% of patients in the sample population were using it. fter controlling for the effect of potential confounders (age and gender), the overall medication use in the DCSS dataset suggested that the likelihood of Maori, Pacific Island, and Asian patients using oral hypoglycaemic agents were not significantly different to the European/Others. Similar findings were demonstrated in the NMDS/Pharms dataset where the overall likelihood of using oral hypoglycaemic agents in all ethnic groups were not statistically significant different to the European/Others population as all confidence intervals included 1.0. Antihypertensive medication use Just over half of all patients in the NMDS/Pharms dataset had received some form of agents affecting the rennin-angiotensin system for their hypertension during the first six months after being discharged from the hospital. The use of diuretics was also investigated but with a much smaller proportion of usage reported. About slightly more than a quarter of the patients with hypertension were using diuretics in the study population. After controlling for the effect of potential confounders (age and gender), the overall medication use in the NMDS/Pharms dataset suggested that the likelihood of Maori, Pacific Island, and Asian patients using agents affecting the rennin angiotensin system for hypertension were not significantly different to the European/Others. On the contrary, it was identified that the differences in the use of diuretics across all ethnic groups after controlling for the effects of confounding variables (gender and age) were significantly different to the standard European/Others population (all confidence intervals excluded 1.0). Both Pacific Islanders (MUR 0.79; 95% CI 0.98-0.63) and Asian (MUR 0.78; 95% CI 0.98-0.61) were reported to have underused diuretics; while Maori patients (MUR 1.25; 95% CI 1.53-1.01) had a greater likelihood of using diuretics in comparison to the European/Others. The likelihood of using both cilazapril and quinapril across ethnic groups were reported not statistically significant different to the European/Others. The only significant difference identified was the use of bendrofluazide between Asian ethnic group and the standard European/Others population. In general, Asian ethnic group (MUR 0.47; 95% CI 0.79-0.26) was reported to have under-used bendrofluazide; while statistically insignificant underuse was observed in both Maori and Pacifica Island ethnic groups. Conclusions The results strongly suggested no ethnic differences in medication use existed in both Maori and Pacific Island groups; however, further studies are required for this association to be fully established as the current study focused mainly on secondary care patients. There might be a problem of generalisation to the general population. Therefore, this study is only a partial representation of patients in the community. Future strategies should include the use of primary care based data and also direct at better understanding of cultural variation in medication use habits.