Impact of health practice index and cardiovascular health metrics on incident cardiovascular disease according to glucose tolerance status

Aims To evaluate and compare the association of incident cardiovascular disease (CVD) with the Health Practice Index (HPI) reflecting only lifestyle habits and Ideal Cardiovascular Health Metrics (ICVHMs) consisting of lifestyle habits and factors targeted for control in the same population according to glucose status. Methods This retrospective cohort study included 1,28,162 participants aged 18–72 years with no history of CVD followed for ≥ 3 years between 2008 and 2016. Participants were classified according to normal glucose tolerance (86,174), prediabetes (36,096), or diabetes (5892). HPI and ICVHMs scores were classified into three groups (high/medium/low). Multivariate Cox regression hazard analysis examined CVD risk. Results During a mean follow-up of 5.2 years, 1057 CVD events occurred. In prediabetes, CVD risk was significantly higher in groups with both medium and low HPI scores and ICVHMs scores compared to high scores for normal glucose tolerance (hazard ratios [HRs] for high/medium/low HPI scores were 0.95 [0.78–1.17], 1.56 [1.29–1.89], and 2.41 [1.74–3.34] and for ICVHMs scores were 0.74 [0.50–1.11], 1.58 [1.26–1.98], and 2.63 [2.10–3.31], respectively). Regarding diabetes, compared with high HPI/ICVHMs scores in the normal glucose tolerance group, a significantly increased CVD risk was observed in the high-score HPI group, but not in the high-score ICVHMs group (HPI high/medium/low HR, 1.63 [1.22–2.18], 2.19 [1.69–2.83], and 2.26 [1.34 -3.83]; ICVHMs high/medium/low HR, 1.14 [0.47–2.81], 2.38 [1.75–3.23], and 3.31 [2.50–4.38], respectively). Conclusions In diabetes, ideal lifestyle practices alone were insufficient for primary prevention of CVD but had a greater impact on primary prevention of CVD in prediabetes.