Development and Validation of the Hypertension Population Risk Tool (HTNPoRT)

BackgroundEffective hypertension prevention requires diagnostic tools applicable to broad populations, yet current models often rely on specialized laboratory testing. We aimed to develop and validate the Hypertension Population Risk Tool (HTNPoRT) - a diagnostic model derived with only readily available data, suitable for individual screening and population health planning. MethodsWe analyzed data from the Canadian Health Measures Survey (cycles 1-6, 2007-2019). The study included community-dwelling respondents aged 20-79 years. The primary outcome was hypertension, defined as measured systolic/diastolic blood pressure of 140/90 mm Hg or current antihypertensive medication use. Sex-specific logistic regression models were developed using 16 predictors, including 4 sociodemographic, 3 psychosocial, 2 health status, 5 health behavioural, and 2 chronic condition variables. The model was fully prespecified, including the stepdown procedure to derive parsimonious models. ResultsOf 19,643 participants, 5,152 (26.2%) had hypertension. The final models included age, body mass index, diabetes status, and family history of hypertension. Optimism-corrected c-statistics were 0.86 (95% CI 0.85-0.87) for men and 0.88 (0.87-0.88) for women. Calibration showed relative differences between observed and predicted risk of 1.02% (men) and 1.41% (women), and consistent performance across 179 of 181 policy-relevant subgroups. ConclusionsHTNPoRT is a well-performing diagnostic tool that relies only on minimal non-invasive, self-reported data. It is suitable for both individual risk screening and population-level surveillance to inform hypertension prevention strategies.