A comparative analysis of cardiovascular disease risk profiles of five Pacific ethnic groups assessed in New Zealand primary care practice: PREDICT CVD-13

Abstract

Background Data on the cardiovascular disease risk profiles of Pacific peoples in New Zealand is usually aggregated and treated as a single entity. Little is known about the comparability or otherwise of cardiovascular disease (CVD) risk between different Pacific groups. Aim To compare CVD risk profiles for the main Pacific ethnic groups assessed in New Zealand primary care practice to determine if it is reasonable to aggregate these data, or if significant differences exist. Methods A web-based clinical decision support system for CVD risk assessment and management (PREDICT) has been implemented in primary care practices in nine PHOs throughout Auckland and Northland since 2002, covering approximately 65% of the population of these regions. Between 2002 and January 2009, baseline CVD risk assessments were carried out on 11,642 patients aged 35–74 years identifying with one or more Pacific ethnic groups (4933 Samoans, 1724 Tongans, 1366 Cook Island Māori, 880 Niueans, 1341 Fijians and 1398 people identified as Other Pacific or Pacific Not Further Defined). Fijians were subsequently excluded from the analyses because of a probable misclassification error that appears to combine Fijian Indians with ethnic Fijians. Prevalences of smoking, diabetes and prior history of CVD, as well as mean total cholesterol/HDL ratio, systolic and diastolic blood pressures, and Framingham 5-year CVD risk were calculated for each Pacific group. Age-adjusted risk ratios and mean differences stratified by gender were calculated using Samoans as the reference group. Results Cook Island women were almost 60% more likely to smoke than Samoan women. While Tongan men had the highest proportion of smoking (29%) among Pacific men, Tongan women had the lowest smoking proportion (10%) among Pacific women. Tongan women and Niuean men and women had a higher burden of diabetes than other Pacific ethnic groups, which were 20–30% higher than their Samoan counterparts. Niuean men and women had lower blood pressure levels than all other Pacific groups while Tongan men and women had the highest total cholesterol to HDL ratios. Tongan men and women had higher absolute 5-year CVD risk scores, as estimated by the Framingham equation, than their Samoan counterparts (Age-adjusted mean differences 0.71% [95% CI 0.36% to 1.06%] for Tongan men and 0.52% [95% CI 0.17% to 0.86%] for Tongan women) although these risk differences were only about 10% higher in relative terms. Conclusion The validity of the analyses depend on the assumption that the selection of participants for CVD risk assessment in primary care is similar between Pacific groups. The ethnic-specific CVD risk profiles presented do not represent estimates of population prevalence. Almost all previous Pacific data has been aggregated with Pacific peoples treated as a single entity because of small sample sizes. We have analysed data from the largest study to date measuring CVD risk factors in Pacific peoples living in New Zealand. Our findings suggest that aggregating Pacific population data appears to be reasonable in terms of assessing absolute CVD risk, however there are differences for specific CVD risk factors between Pacific ethnic groups that may be important for targeting community level interventions.