Aim: To establish the population-level predictive validity of the Framingham Heart Study risk equation used to derive cardiovascular disease risk-prediction charts for New Zealand management guidelines on raised blood pressure and dyslipidaemia.
Methods: During 1992-3, standard cardiovascular risk factors were measured in a cohort of 6354 people (4638 men and 1716 women) aged 35-74 years with no history of cardiovascular disease, who were either enrolled on Auckland general electoral rolls or employed by a nationwide multi-industry corporation. Five-year cardiovascular event rates were predicted for the cohort using a Framingham risk equation and were compared with observed five-year cardiovascular event rates for hospitalisation and mortality.
Results: Incident cardiovascular hospitalisations or death were reported for 411 cohort participants (325 men and 86 women) during five years of follow up. There was good agreement between observed and predicted five-year cardiovascular event rates in all five-year age/sex categories up to 69 years. The risk equation tended to underestimate risk slightly in men and overestimate risk slightly in women up to 69 years, while in 70- to 74-year-olds it was accurate for men. Its accuracy in older women is uncertain.
Conclusion: A Framingham Heart Study risk equation used to estimate incident cardiovascular events in individuals without previous cardiovascular disease accurately predicts five-year risk of hospitalisation or death from a first cardiovascular event in New Zealand men aged 35-74 and women aged 35-69 years at the population level.