Author + information
- Alexander M. Clark, PhD* (, )
- Lori Savard, BSc and
- David R. Thompson, PhD
- ↵*University of Alberta, Level 3 CSB, Clinical Sciences Building, Edmonton, Alberta T63 2G3, Canada
Dr. Ferrante and colleagues are correct to draw attention to the merits of the DIAL (Randomized Trial of Telephone Intervention in Chronic Heart Failure) because, unlike other trials, it was a large and comparatively well-described intervention delivered in a middle-income country. As such, DIAL is a welcome addition to the evidence base.
However, we disagree that reports of DIAL meet the requirements of the modifiedConsolidated Standards of Reporting Trials statement for nonpharmacological trials (1), because the original and companion reports (2) do not identify what usual care consisted of or how the care provided to the intervention group was standardized or monitored. These new standards are important, because they reflect principles of critical appraisal, and detailed information is necessary for rigorous systematic review (3).
We agree with the authors that the results of program trials are inconsistent. Current evidence does not adequately conceptualize or describe programs (3). However, it is becauseof these weaknesses that further meta-analysis of programs is inappropriate. As such, we did not seek to include “all” available evidence in our viewpoint but identified recent trials that did not find programs to be beneficial.
We drew attention to the lack of understanding of why program effects vary (4). These variations can be dismissed or attributed to biases or methodological weaknesses but may reflect actual differences in effects. Indeed, a very large recent trial (5) of another predominantly telephone-based program for patients (n = 30,000) with heart failure and diabetes reinforced the need to understand program effects better. This independent evaluation found no benefits from programs on hospitalization, mortality, patient satisfaction, self-care, or mental and physical functioning. Costs “far exceeded” savings. These negative results must be considered in the light of other positive findings, including those from DIAL.
In the face of these variations, proponents of different types of programs may continue to argue which type of program is “best,” but this reflects an overly simplistic approach to evidence-based health services. Different types of programs are likely to be suitable for different settings and populations with different resources. Developing a more nuanced and context-responsive evidence base is now vital.
- American College of Cardiology Foundation
- Clark A.M.,
- Savard L.A.,
- Thompson D.R.
- Kapp M.,
- MccCall N.,
- Cromwell J.,
- Urato C.,
- Rabiner D.