Variation in reported prevalences of hypertension in The Netherlands: the impact of methodological variables.

STUDY OBJECTIVE To estimate the prevalence of hypertension in The Netherlands and to quantify the influence of methodological variables on the reported prevalences. DESIGN A pooled analysis was performed based on reported age specific prevalences of hypertension. A logistic model was used to estimate the probability of hypertension. MAIN RESULTS The age standardised prevalence of hypertension varies more than fivefold between studies carried out in The Netherlands. The probability of having hypertension was lower if blood pressure was measured at more than one point in time (Odds ratio 0.44 (OR) (95% confidence interval (95% CI) 0.38, 0.51) for men and 0.47 (0.41, 0.54) for women, and if the study was carried out more recently (OR 0.92 (0.91, 0.93) per year). The probability was higher if the study was carried out in a general practice (OR 1.14 (1.03, 1.27) for men and 1.52 (1.36, 1.69) for women). The inclusion of treated people as hypertensive yields contradictory results for men and women. CONCLUSIONS The strong variation in prevalence is explained by methodology and by a period effect indicating a decrease of the prevalence of hypertension over time. Whether this decrease is true or caused by confounding due to unknown or unreported methodological variation over time is unknown. For future studies, a standardised method could reduce the influence of methodological variables and thereby the variation in reported prevalences.