Predicting cardiac rehabilitation enrollment: the role of automatic physician referral

BACKGROUND: Despite the established benefits of cardiac rehabilitation, evidence suggests referral to, and subsequent enrollment in, cardiac rehabilitation following a coronary event remains low (10-25%). The aim of this study was to identify predictors of attendance to cardiac rehabilitation intake and subsequent enrollment in rehabilitation after coronary artery bypass graft surgery within the framework of an automatic referral system. DESIGN AND METHODS: We conducted a historic prospective study of patients who underwent coronary artery bypass graft surgery between 1 April 1996 and 31 March 2000 and lived within the geographic referral area of a multi-disciplinary cardiac rehabilitation center in central-south Ontario, Canada. Coronary artery bypass graft surgery patients are automatically referred to cardiac rehabilitation at the time of hospital discharge. Consecutive health records of eligible patients were reviewed for medical history, cardiac risk factor profiles, and evidence of cardiac rehabilitation intake attendance and enrolment. RESULTS: A total of 3536 patients met eligibility criteria. Patients were predominantly male (79.1%), approximately 64 years of age, living with a spouse or a partner, English-speaking, retired and had multiple cardiac risk factors. Of eligible patients, 2121 (60.0%) attended the cardiac rehabilitation intake appointment. Of patients who attended cardiac rehabilitation intake 1463 (69%) enrolled in at least one cardiac rehabilitation service, based on their risk factor profile. Selected cardiac rehabilitation services were exercise training (n=1287; 88%), nutrition counseling (n=571; 39.0%), nursing care (n=546; 37.3%), and psychological intervention (n=223; 15.2%). CONCLUSIONS: An institutionalized, physician-endorsed system of automatic referral to cardiac rehabilitation resulted in higher rates of cardiac rehabilitation intake and enrollment following coronary artery bypass graft surgery than previously reported and should be adopted for all cardiac populations.