BackgroundTo ensure that indicators for assessing prescribing quality are appropriate and relevant, physicians should be involved in their development. How general practitioners (GPs) rank these indicators is not fully understood. Aims(i) To determine how GPs in Ireland rank a set of evidence-based prescribing quality indicators in order of importance and relevance to their practice, and (ii) to compare the GPs' ranking of the defined set of indicators with actual prescribing practice using a prescription database. MethodsA postal questionnaire was sent to 105 GPs, who were asked to rank a set of 11 prescribing quality indicators, identified from the literature from most to least important. The results were aggregated and a weighted score for each indicator determined. These same prescribing indicators were then applied to a prescription database to compare the ranking provided with actual prescribing practice. ResultsEighty-six GPs (82%) returned the completed questionnaire. The higher ranks were for quality issues-use of inhaled corticosteroids, statins and benzodiazepines. Actual prescribing data showed prolonged use of benzodiazepines in over half of the prescriptions dispensed ( n = 18 171), 52.48% (95% confidence interval 51.95, 53.01) and low usage of generic drugs, 17.78% (17.70, 17.90) despite their high ranking by the GPs. ConclusionWhile GPs have diverse views about the value of different prescribing quality indicators, the results suggest that they do rank evidence-based guidelines on patient management highly, but those based on costs and less evidence the lowest. There was considerable divergence between theory and practice in the application of quality indices.

What is already known about this subject • Recommended treatment guidelines for hypertension have been developed to assist GPs' decisions about appropriate therapies. • The British Hypertension Society's (BHS) 2004 guidelines recommend initial drug choice based on age, and avoidance of β‐adrenoceptor blockers in diabetes. What this study adds • Prescribing of first‐line antihypertensives in Ireland appears guided by age, but mainly for those under 55 years. • Adherence to the guidelines was in part related to patient gender. • Presence of concomitant diabetes had a greater influence on the choice of therapy than age of patient. Aims To determine adherence to hypertension guidelines in relation to age and diabetes. Methods The Irish HSE‐PCRS prescribing database identified patients initiating antihypertensive monotherapy in 2005. Logistic regression predicted the likelihood of therapy according to guidelines. Results The odds ratio (OR) of receiving therapies according to the guideline recommendations in those <55 years vs. ≥55 years was 1.31 (95% CI 1.26, 1.37). Diabetics were more likely than nondiabetics to receive antihypertensives other than β‐adrenoceptor blockers (OR 2.97, 95% CI 2.74, 3.21). Conclusions Our findings show some adherence to the guidelines in relation to age but selective prescribing of antihypertensives for diabetics.

What is already known about this subject • Underuse of antithrombotic therapy in atrial fibrillation has been identified in hospital base and community surveys. • Whether this applies nationally and whether underuse is age or gender related is not established. What this study adds • Using a pharmacoepidemiological database with digoxin as a surrogate for atrial fibrillation, its prevalence and the concomitant use of antithrombotic therapy nationally mirrors that seen in smaller studies. • In addition, both prescribers' views and practice show that, based on age, those at highest risk and women are least likely to receive warfarin, which is predominantly used in the low‐risk population. • Such databases may be used nationally to examine adherence to guidelines. Aims To examine if appropriate antithrombotic therapy in atrial fibrillation is implemented nationally. Methods Using prescriptions for digoxin as a surrogate for atrial fibrillation, we identified its coprescription with antithrombotic therapy, aspirin or warfarin in a national prescribing database in 27 571 patients over 45 years old. Results Proportionately significantly more men were on warfarin, and use in those >75 years old was three times less than in those <65 years. Reluctance to use antithrombotics was confirmed in a postal survey. Conclusion Data suggest a missed opportunity to prevent stroke with women and those >75 years old least likely to receive warfarin.

We agree with the overall conclusions of Vegter et al. [1], whose data show a similar choice of first antihypertensives to that reported in Ireland.We now represent our data as requested, using likelihood to receive recommended therapy for 'young males vs. old males' , which shows an odds ratio (OR) = 1.75 [95% confidence interval (CI) 1.63, 1.88] and for 'young females vs. older females' , OR = 1.11 (95% CI 1.06, 1.17).We suggest that the reasons for similar findings in a number of different countries are not necessarily the same. For example, as distinct from the Netherlands, in Ireland we do not have pharmacotherapy audit meetings or strong national guidelines. However, we do agree that other factors may influence the choice of antihypertensive therapy. It is interesting to observe that, despite adverse publicity concerning b-blockers as first-line antihypertensive agents, in both countries, as seen in these studies, they have remained among the most popular antihypertensive groups [1,2].