The GAP QI intervention was previously shown to be successful in increasing physician adherence to guidelines in-hospital, and to reduce mortality.3,9,29
We evaluated an added telephone counseling intervention post-discharge. Over a period of 8 months after hospitalization, we found the telephone counseling intervention to add minimally to self-reported behavior change during the intervention, with no significant changes in health status or quality of life. While we found a small difference in physical activity in the intervention group, our results are largely negative. We conclude that a telephone counseling intervention added to hospital QI during the 3-month period post-hospitalization failed to produce a meaningful advantage in terms of health status and quality of life.
What is the explanation for our negative results? One explanation could be that the intervention approach itself was not a good match for these patients. While our follow-up intervals exceeded the usual three-month interval found in most studies, it is possible that the increased physical activity reported, if maintained for an even longer time period, could result in health status and quality-of-life benefits. However, our previous success in smoking cessation in primary care did not carry over to multiple risk factor intervention post-hospitalization for ACS.30–32
This approach did not produce gains in physical activity, weight, functional status or quality of life beyond what patients accomplished spontaneously with medical management.
We believe the main explanation for the failure of the telephone intervention to show additional benefits is that it came on top of an ongoing QI program in which patients consistently received standard in-hospital counseling. This suggests that for the majority of patients, instruction in hospital appears to have been important and effective, and that additional counseling outside the context of follow-up office care added only a little benefit. It may well be that, at least following ACS, patients largely followed the discharge advice, including relatively high medication adherence. The QI protocol required in-hospital counseling and a discharge patient contract that provided the patient with numerical values for ejection fraction and cholesterol, and made medication and behavior change recommendations.
Do we conclude that telephone-delivered health behavior change is not effective in accomplishing secondary prevention? In part, the answer is yes. We show that in our study, telephone counseling for patients added little when the clinicians were participating in rigorous quality improvement. In other settings that do not have active QI protocols, telephone counseling may act as a “reminder” to a patient to raise guideline-specific issues with a clinician, though we are not aware of research testing this approach. Putting our results in a broader context, they are consistent with a recently published Cochrane review of telephone follow-up following discharge from hospital. The review included 33 randomized trials or quasi-randomized trials that followed a total of 5,110 patients.33
The review authors found the studies to be of generally low methodological quality that varied widely in terms of type of health professional involved. However, they concluded that overall, clinically equivalent results were found in the telephone follow-up and control groups. Our study is more rigorous, followed patients for eight months rather than three, and followed a known standardized quality improvement protocol in the five hospitals. Our findings corroborate the more tentative ones of Mistiaen and Poot in the Cochrane review33
. This suggests that where resource allocation choices are being made, higher payoff for secondary prevention post-hospital may be found in consistent delivery of guideline-consistent care.
What is the impact of other patient-oriented interventions? Several reviews of related standardized, written patient information show clear improvement in patient knowledge.34,35
Reviews of patient-centered care, and coaching show improvements in patient satisfaction and question asking.36,37
Free-standing patient-oriented behavior change interventions, however, show limited impact on behavior and health outcomes, including layperson led chronic disease self-management courses.38,39
At the same time, some disease-specific interventions that provide information to both clinician and patient at the point of care do show impact on health outcomes in at least some patient-oriented chronic disease interventions.40,41
We conclude that patient-oriented interventions, to be both effective and efficient must be integrated into on-going medical care. Whether or not brief behavior change interventions can contribute substantially to clinical care beyond the clear success of smoking cessation requires further testing. Future research should investigate both brief and intensive behavior change interventions that are well-integrated into care delivery systems.