OBJECTIVE: Hypertension is often poorly controlled, despite its importance and despite the availability of very effective treatments. An under-recognized problem is the failure of consensus guidelines to acknowledge the important difference between efficacy in clinical trials and effectiveness in clinical practice. The present survey was designed to prospectively assess what is the target blood pressure (BP) goal defined by a general practitioner (GP) for an individual patient, and what are the reasons for not modifying an antihypertensive drug regimen, when pre-defined individual BP goals are not achieved. DESIGN: Family practice based, open intervention survey. SUBJECTS: Participating GPs enrolled 2621 hypertensive patients. At the first visit each physician was required to assess the cardiovascular risk profile of each patient and to define individual BP targets. INTERVENTIONS: Treatment was started with irbesartan alone or in fixed combination with hydrochlorothiazide. Follow-up visits were suggested after 1 month, 2 months and 4 months. Physicians were asked to report BP values under the new treatment regimen and to indicate whether in their opinion pre-defined BP targets set at baseline were achieved or not and whether the antihypertensive regimen was modified or maintained in relation to whether target BP was reached or not. MAIN OUTCOME MEASURE: To provide reasons for not changing the treatment even though BP goals were missed. RESULTS: Average target BP values defined by the physicians at baseline were 138 +/- 8 mmHg for systolic and 84 +/- 5 mmHg for diastolic BP. Among GPs, defined target BP values did not depend on individual risk stratification, but rather depended on baseline BP values. At baseline systolic and diastolic BP averaged 165/97 +/- 17/10 mmHg, while at the last visit achieved BP averaged 140/84 +/- 14/8 mmHg. There were three main reasons for not intensifying antihypertensive treatment when BP targets were not achieved. These reasons were: (1). the assumption that the time after starting the new drug was too short to appreciate its full effect (44% at first, 14% at last follow-up), (2). that there was a clear improvement or the target BP was almost reached (24% at first, 34% at last follow-up) or (3). that self-measurements were considered satisfactorily (10% at the last visit). CONCLUSIONS: Failure of physicians to follow guidelines is apparently dependent on the belief that baseline BP dictates the target, that a clear improvement in BP might be sufficient and that the full drug effect may take up to 4 months or more to be attained.