Intended for healthcare professionals


Interventions to enhance self management support

BMJ 2013; 346 doi: (Published 19 June 2013) Cite this as: BMJ 2013;346:f3949
  1. Xin Sun, senior researcher12,
  2. Gordon H Guyatt, professor23
  1. 1Chinese Evidence-based Medicine Center, West China Hospital, Sichuan University, Chengdu 610041, China
  2. 2Department of Clinical Epidemiology and Biostatistics, McMaster University, Hamilton, Ontario, Canada
  3. 3Department of Medicine, McMaster University, Hamilton, Ontario, Canada
  1. sunx26{at}

Add no noticeable value to the benefits of existing care for chronic conditions

Clinical trialists, interested in ensuring that their interventions are effective in the real world of clinical practice, have long called for study designs—termed practical, pragmatic, or effectiveness—that reflect practice as closely as possible.1 These practical trials are particularly important in health services research, which can involve apparently effective interventions that are resource intensive and implemented by charismatic enthusiasts. Such interventions are likely to fail when introduced into settings with fewer resources by doctors preoccupied with the considerable stresses of everyday practice. In a linked paper by Kennedy and colleagues (doi:10.1136/bmj.f2882), we learn that effective interventions in health services are often not feasible and—as revealed by practical trials—feasible interventions are often not effective.2

In a cluster randomised controlled trial, Kennedy and colleagues investigated whether self management interventions targeting care providers could improve outcomes over 12 months in patients with diabetes, chronic obstructive pulmonary diseases, and irritable bowel syndrome. The investigators went to great lengths to implement a practical trial design. Their intervention was brief, consisting of only two sessions with staff from primary care clinics. Attendance at the two training sessions was high (90% and 82%, respectively), and most attendees rated the sessions at least moderately positively (mean score >2.5 on a five point scale). However, 42% of physicians reported no use of a tool to assess patient support needs and priorities, which was at the heart of the intervention. Patient reports confirmed failure of implementation. Inevitably, the intervention had no effect on health related quality of life, self efficacy, resource use, or many secondary outcomes.

Does this report represent an isolated failure of interventions for self management support directed as care providers? Unfortunately, it does not. Previous randomised trials have focused largely on educational and telehealth interventions directed at patients, with the ultimate goal of enhancing health outcomes and reducing costs.3 4 5 Many such trials have investigated various interventions for self management of patients with diabetes, asthma, heart failure, irritable bowel disease, depression and pain. Some studies suggested benefits on outcomes such as quality of life, symptoms (for example, pain), social function, and psychological wellbeing,3 4 5 6 7 8 although others did not.

The few studies testing interventions of self management support that focus on care providers provide even less encouragement. Their results, in keeping with the current study, have been completely negative. Among these trials,9 10 11 not only did the outcomes of interest not differ between intervention and control groups, but investigators were—again, as in the current study—unable to document significant differences in the implementation of self management support.

Had doctors implemented the self management supports as planned, would these trials have resulted in important benefits for patients? Although the results of patient level trials suggest that they might, the failure to implement leaves considerable doubt. Implementation is not an all or nothing phenomenon (although what has happened thus far seems quite close to nothing), and the degree of implementation that might realistically be achieved through more intensive interventions remains uncertain. Furthermore, whether doctors have the skills and training to do a good job is also uncertain.5

The difficulties in implementing self management support in pragmatic clinical trials is best understood in the context of the broader literature on changing doctors’ behaviour 12 Many randomised trials have left us with a sobering realisation of the enormous challenges of such behavioural change. One important contextual problem is the competing demands on the care of long term conditions that doctors face in the trial. Another is the understanding and motivation of the healthcare team. In the current trial, one might seriously question doctors’ awareness of the importance of patient self management of chronic conditions. After all, they not only refused to participate in additional training beyond the first two sessions, but also refused to allow monitoring of the fidelity of the intervention.

It is evident that the failure of care directed interventions to enhance patient self management results from inadequate consideration of the relevant attitudes and possibly skills of doctors and the obstacles of time, inertia, and competing priorities. Considerable incentives will probably be needed to change doctors’ behaviours with respect to self management support practices. The authors’ ultimate conclusion is even more pessimistic: “perhaps we should abandon current models of both provider and patient based self management support for innovative interventions.” Although they may be right, experience with current models mandates careful study of all possible obstacles, and even then a high level of skepticism regarding all future efforts in this area.


Cite this as: BMJ 2013;346:f3949


  • Research, doi:10.1136/bmj.f2882
  • Competing interests: I have read and understood the BMJ Group policy on declaration of interests and declare the following interests: none.

  • Provenance and peer review: Commissioned; not externally peer reviewed.


View Abstract