There are now a number of guidelines outlining the diagnosis and management of patients with chronic heart failure (CHF). The extent to which these guidelines are used and the effects on patient outcomes are not well known. The aim of this study was to examine the implementation of a heart failure guideline among cardiologist and non-cardiologist physicians in a university hospital setting. Case record data were examined from 400 patients with a primary diagnosis of CHF. Management of these patients was assessed using a systolic heart failure guideline (Scottish Intercollegiate Guideline Network, number 35) as a benchmark. Hospital admission data were examined contemporaneously over a 17-month period to assess associations between adherence to drug therapies and number of admissions. Overall, there was poor adherence to the guideline, with relatively high use of angiotensin-converting enzyme (ACE) inhibitors/angiotensin II receptor blockers (ARBs) (80%), low use of beta-blockers (32%) and digoxin (36%), and very low use of spironolactone (13%). Cardiologists used more beta-blockers (37 vs. 21%, P=0.003) and digoxin in sinus rhythm (18 vs. 5%, P<0.001) than non-cardiologists. Hospital admission rate was individually associated with increasing age, NYHA status, beta-blocker, diuretic and spironolactone prescription (all P<0.001). At multivariable analysis, only age, NYHA status and increased diuretic prescription were associated with more frequent admission (P<0.001, R(2)=0.15). Despite carefully designed guidelines, the implementation of evidence-based therapies for CHF remains inadequate, even in a university hospital environment. This may reflect a lack of organisational developments to facilitate the increasingly complex management of patients with CHF.