Ticking one more heart-health box linked to 11% lower death risk

Cardiovascular health is usually discussed one risk factor at a time, which makes it hard to know what a whole-package approach is worth. Composite scores that bundle smoking, diet, physical activity, body weight, cholesterol, blood glucose and blood pressure were designed to answer exactly that. But individual cohort studies looking at these scores against death and disease had produced inconsistent results, some finding large protective associations and others much weaker ones. Pooling them was needed to see whether the composite really tracks with the outcomes people care about.

The authors searched PubMed and Web of Science from inception to February 2017, plus the reference lists of retrieved papers, for prospective cohort studies relating ideal cardiovascular health metrics to cardiovascular events and mortality. Thirteen prospective studies with a combined 193 126 cohort members met the criteria. They compared the highest against the lowest categories of ideal metrics and calculated pooled relative risks for all-cause mortality, cardiovascular mortality, cardiovascular disease and stroke. They also tested whether risk fell in a steady line as the number of ideal metrics increased.

Comparing the most to the least ideal categories, the pooled relative risk was 0.54 for all-cause mortality, 0.30 for cardiovascular mortality, 0.22 for cardiovascular disease and 0.33 for stroke. The relationship for all-cause and cardiovascular mortality was linear, meaning risk fell steadily rather than only at the extremes: each additional ideal metric corresponded to 11% lower all-cause mortality and 19% lower cardiovascular mortality. When individual components were examined, significant results for mortality came from smoking, diet, physical activity, plasma glucose levels and blood pressure.

This consolidates a scattered set of cohort findings into one consistent picture and supports the case for composite scoring rather than chasing single risk factors. The linear dose-response is the most useful part, because it suggests partial improvement is not wasted. What it cannot settle is causation: these are people followed over time, not people randomly assigned to change their habits, and those who score well tend to differ in income, education and access to care. Whether deliberately moving a metric into the ideal range delivers the same benefit remains untested here.

The practical message is that cardiovascular health appears to behave as a gradient rather than a pass or fail. Improving one component while others lag still sat alongside meaningfully lower risk in these cohorts. The components carrying the clearest signal for mortality were smoking, diet, physical activity, blood glucose and blood pressure, which is a reasonable indication of where the weight of the association lies. Since this comes from following populations rather than intervening on them, treat the numbers as a strong pattern worth knowing rather than a guaranteed personal return.