Diuretics for hypertension – an inconsistency in primary care prescribing behaviour

Abstract

Objective:Internationally there is an ongoing debate on diuretics as first-line therapy for most patients with hypertension. In spite of many arguments against them in antihypertensive monotherapy, the authors of the present study perceived them to be regularly prescribed in combination therapy in Germany. The study objective was to look for this discrepancy in prescribing reality as a contribution from clinical practice to an academic debate.Methods:A descriptive cross-sectional study in a yearly contact group (YCG; 1.7.2007–31.06.2008) was conducted based on data from a scientific network of 22 general practitioners in Germany. All patients with hypertension as diagnosed by their general practitioner were included. Antihypertensives were grouped according to the ATC classification. To assess for potential design effects by the given two-level setting, 95% confidence intervals (CI) were adjusted for clustering.Results:Hypertension had been diagnosed in 9.3% of the 58 852 patients. Of these, 21.6% received no antihypertensives. Of those who were treated, 30.6% (CI [28.6–32.6]) had monotherapy. In monotherapy, 8.6% (CI [7.1–10.2]) were prescribed some diuretic, 1.5% (CI [0.5–3.0]) received hydrochlorothiazide (HCT). Combination therapy was prescribed to 69.4% (CI [67.2–71.6]). These patients received some diuretic in 79.0% (CI [76.9–81.0]) of the cases, of which 80.8% (CI [78.5–83.1]) had a combination with HCT. HCT was prescribed in 76.2% (CI [73.5–78.9]) in fixed-dose formulations.Conclusion:In spite to the many arguments against them, leading to their almost complete disregard in monotherapy, thiazide-diuretics seem to be standard in combination therapy in Germany. This inconsistency can not be explained by the arguments of the current debate. Key limitations of the present study include the lack of ability to tell whether a given monotherapy is the first-line medication, the small sample size and the possible selection bias.