Cardiac rehabilitation program attendance after coronary artery bypass surgery: overcoming the barriers

In Australia, cardiac rehabilitation (CR) is available for all patients after an acute cardiac event. CR usually comprises low-to-moderate-intensity exercise and education, and runs weekly for 6–8 weeks, commencing in early convalescence. CR has many benefits, including reduced mortality and improved risk-factor profiles and physical and psychological functioning.1-3 Therefore, international guidelines recommend all cardiac patients be referred to CR.1,4,5

However, many patients do not attend CR programs,3,6,7 due to a range of barriers.1,6,8-10 While coronary artery bypass graft surgery (CABGS) patients are more likely than other cardiac patients to attend CR,3,6 overall attendance in Australia remains less than optimal, ranging from 37%3 to 66%.6

A major factor in non-attendance at CR is the absence of effective referral procedures.1,7 The Victorian Government’s Best practice guidelines for cardiac rehabilitation and secondary prevention recommend specific referral strategies to maximise CR attendance (Box 1).1 Royal Melbourne Hospital (RMH) in Victoria has adopted these recommended referral procedures for all CABGS patients (regardless of the CR program to which they are referred), and has also implemented additional recruitment procedures consistent with the best practice guidelines (Box 2) to further encourage attendance among patients referred to the RMH CR program.

We investigated the CR attendance rates of RMH CABGS patients and the impact on attendance of sociodemographic, medical, cognitive, psychosocial and geographical barriers, to determine if adherence to best practice guidelines for CR referral and recruitment minimised the influence of individual patient factors on attendance.

The average age of patients was 66 years (SD, 10 years; range, 42–88 years). Other patient characteristics are shown in Box 3.

Of the 170 included patients for whom CR attendance was tracked, 123 (72%) attended at least one CR session. The number of sessions offered across programs ranged from six to 16, with a mean (SD) of 6.9 (1.5). Most patients (90/123, 73%) who attended one session completed the full program. Of the 123 attendees, 30 were referred to and attended the program at RMH, while the remaining 93 were referred to and attended the program closest to their home. There was no significant difference in CR attendance between the included and non-responding patients (χ21 = 1.60; P = 0.28).

Characteristics of CR attendees and non-attendees are shown in Box 3. Compared with non-attendees, patients who attended CR had a significantly shorter travel time (mean difference, 5.31 min [95% CI, 0.81–9.81 min]; F1,159 = 5.42; P = 0.021), lived closer to the program venue (mean difference, 5.53 km [95% CI, − 0.22 to 11.27 km]; F1,159 = 3.61; P = 0.059) and were more likely to have been referred to the program at RMH than elsewhere (91% v 68%; n = 169; χ21 = 6.8; P = 0.006). There were no other significant differences between attendees and non-attendees.

When these three significant variables were entered into the logistic regression model, referral to RMH (Wald, 5.10; odds ratio [OR], 4.36 [95% CI, 1.21–15.26]; P = 0.024) and travel time (Wald, 4.25; OR, 0.86 [95% CI, 0.75–0.99]; P = 0.039) emerged as significant, independent predictors of attendance. Patients referred to RMH were over four times more likely to attend CR than patients referred elsewhere. Patients were less likely to attend CR as travel time increased: 1 min of extra travel time was associated with a 14% reduction in the likelihood of attendance, and 10 min of extra travel time corresponded to a 77% reduction. The model explained more than 10% of the variance in CR attendance (Nagelkerke’s R2 = 0.124).

The implementation of best practice guidelines at RMH appears to overcome the multiple barriers to CR referral and attendance.1,6,8-10 To our knowledge, no other studies have reported a higher CR attendance rate for CABGS patients, either within Australia or internationally. Patients traditionally at risk of non-attendance1,6,8-10 were no less likely to attend CR, and the attendance rate for patients referred to CR at RMH itself was particularly high. Thus, when routine inclusive referral practices are in place, along with tenacious follow-up, attendance rates are high and the impact of individual patient characteristics on attendance is minimised.

RMH has a policy of automatic referral to CR as part of the discharge plan for all CABGS patients. This process is facilitated by the ward clerks, who ensure that referrals are completed before discharge. In many Australian hospitals, the referral process is often ad hoc, being the responsibility of busy ward nurses or a single staff member, and can be vulnerable to time pressures.16 The higher overall CR attendance rates at RMH appear to have resulted from efforts to overcome systemic problems in the referral process. Further, the practices followed by the cardiothoracic unit at RMH aid patient decision making about CR attendance by ensuring that patients’ perceptions of CR are accurate and positive.

Attendance was affected by travel time. This is an important consideration in planning service provision for cardiac patients, highlighting the importance of a network of CR programs to ensure that travel times are minimised and access maximised.

Some limitations of our study should be noted. First, while the 46% response rate is consistent with that for previous postal surveys for this patient group,17 it could reduce the external validity of the study findings, although participant and non-participant characteristics were comparable. Second, CR attendance could not be tracked for all patients due to loss of records in two hospitals. Finally, while adoption of systematic procedures for routine referral and follow-up appears to enhance attendance for CABGS patients, these findings do not necessarily apply to patients who have had other acute cardiac events.

Our findings reinforce the importance of attention to procedures as well as policy. It is recommended that, where possible, the best practice guidelines1 for both referral and follow-up should be adopted by all hospitals. Future research could then investigate the impact of widespread implementation of these practices.