The impact of New Zealand CVD risk chart adjustments for family history and ethnicity on eligibility for treatment (PREDICT CVD-5).

AIMS

Current New Zealand cardiovascular (CVD) risk management guidelines advocate targeting treatment to patients with a high 5-year CVD risk assessed using a calculator derived from the Framingham Heart Study. For some high-risk population subgroups, a 5% upward adjustment to their calculated 5-year CVD risk is recommended. We estimated the impact of these adjustments on eligibility for treatment in a primary care setting.

METHODS

Between 2002 and 2006, 23,709 patients visiting their primary care provider in Auckland, New Zealand had CVD risk assessments as part of an opportunistic screening programme using PREDICT, a web-based clinical decision support system. We calculated their baseline CVD risk with and without the 5% upward adjustment for family history of premature ischaemic CVD or for being of Maori, Pacific or Indian subcontinent ethnicity.

RESULTS

A baseline CVD risk could be calculated for 23,693 (99.9%) patients of whom 90% were between ages 35 and 74 years. Unadjusted risk scores classified the majority (70%) below the 10% 5-year risk threshold for specific individualised treatment. A further 11% were between 10 to 15% risk (recommended to receive individualised lifestyle counselling in general practice) and 19% had a greater than 15% risk ( recommended for drug treatment and referral to a dietician in addition to individualised lifestyle counselling). Over a quarter of patients were recorded as having a premature family history of CVD; 21% were Maori, Pacific, or Indian subcontinent and thus met the criteria for a single 5% upward adjustment. This increased the number of people eligible for drug treatment, intensive lifestyle management, and dietician referral by approximately 20% and individualised lifestyle assessment and counselling by 50%.

CONCLUSIONS

The upward adjustments to the calculated CVD risk recommended by the New Zealand CVD risk management guidelines has the potential to substantially increase resource requirements for CVD preventive services in primary care. Moreover the true impact is likely to be underestimated given other adjustment factors related to diabetes risk that were not available in this dataset. Given the impact of these relatively small changes to the CVD risk calculator, locally developed and validated risk equations are urgently needed.