Self management education and good professional consultation skills for patients with diabetes

BMJ 2012; 344 doi: http://dx.doi.org/10.1136/bmj.e2673 (Published 26 April 2012) Cite this as: BMJ 2012;344:e2673
  1. Frank J Snoek, professor
  1. 1Department of Medical Psychology, VU University Medical Centre, 1081 BT Amsterdam, Netherlands
  1. fj.snoek{at}vumc.nl

Their effect on clinical outcomes in routine practice is time limited

It is well recognised that people with diabetes, both young ones with type 1 disease and adults with type 2 diabetes, need to develop skills in self management to manage their condition successfully. In the past few decades different strategies aimed at promoting self management have been developed and tested across a wide range of patient groups and settings. Two linked papers (doi:10.1136/bmj.e2333; doi:10.1136/bmj.e2359), which examine two different approaches delivered in routine practice, improve our understanding of the longitudinal dose-response relation between educational and consultation interventions and improvements in the self management behaviours and subsequent health outcomes of patients with diabetes.1 2

A systematic review found that group based programmes for self management strategies in type 2 diabetes were effective in terms of improving blood glucose, blood pressure, body weight, and requirement for antidiabetes drugs.3 The Diabetes Education and Self Management for Ongoing and Newly Diagnosed (DESMOND) programme was developed in the United Kingdom and tested in 824 patients who had been newly diagnosed with type 2 diabetes in a clustered randomised trial in primary care. At one year follow-up, the programme was shown to be effective with regard to weight loss, smoking cessation, and depressive symptoms but not glycaemic control.4 On the basis of these findings, the six hour structured DESMOND education programme was deemed to be cost effective.5 In the first of the linked studies, the DESMOND group reports on a uniquely long trial follow-up of three years.1 The authors were able to collect biomedical data from 83% of the original sample and psychosocial data by postal questionnaires from 70% of the participants. Participants who were lost to follow-up were younger, less healthy, and more depressed than those who were followed up. Over time, a non-significant small increase in glycated haemoglobin (HbA1c) was seen in both the usual care and the DESMOND group, with a mean HbA1c of 64 mmol/mol (8.0%), and no differences were found in other biomedical or lifestyle outcomes or use of drugs. The benefits in terms of non-smoking and weight loss seen at 12 months were not sustained.

From the data presented it is not clear at what point in time the improvements were lost. Interestingly, significant differences in illness beliefs, as measured by the illness perception questionnaire, were sustained, suggesting a more correct “mental model” of diabetes in people exposed to DESMOND. For example, DESMOND participants were more likely to agree on the chronic nature of diabetes and their personal responsibility in managing their symptoms. Social cognitive theories of behavioural change postulate that changes in a person’s illness beliefs or representations drive the adoption of health behaviour changes, with subsequent changes in medical outcomes. However, various illness beliefs do not have equal relevance for maintaining acquired behaviour changes at a later stage.6 Khunti and colleagues rightly point to the need for additional ongoing education and support in primary diabetes care to help sustain the initial benefits of the self management programme. For patients with poorly controlled diabetes, relatively short and targeted behavioural interventions are available and were shown to be effective.7

It is difficult to achieve strict glycaemic control in young people with type 1 diabetes, particularly during adolescence. Large differences are seen between centres,8 however, that are probably explained largely by differences in target setting by diabetes teams.9 Thus, providing help for healthcare teams to improve their ability to express and discuss goals for treatment could indirectly improve treatment outcomes for young people with type 1 diabetes. In the second linked study, Robling and colleagues present the 12 month results of a large well designed cluster-randomised trial conducted in 26 paediatric diabetes clinics in the UK where healthcare teams were offered the Talking Diabetes learning programme, which is informed by motivational interviewing and aimed at improving healthcare professionals’ consulting skills.2 Shared agenda setting, which ensures that concerns of patients and professionals are dealt with, is central to this approach. In Robling and colleagues’ trial, 79 professionals were trained, and 359 children with type 1 diabetes (mean age 10.6 years) and their care givers were included. No differences were found in HbA1c or quality of life at 12 months, which is disappointing, although trained teams were more skilful than controls in guiding, agenda setting, and consultation strategies, even though absolute levels of skilfulness were low. By contrast, a previous multicentre randomised controlled trial of motivational interviewing showed significant improvements in glycaemic control and psychosocial outcomes in adolescents with type 1 diabetes.10 In that study, the intervention was delivered outside the clinic by a skilled nurse specialist and with more frequent opportunities for contact. The difference in outcomes between the two trials might be explained by differences in the intensity and the quality of the intervention delivered.

That both the DESMOND and the Talking Diabetes trial were carried out in routine practice adds to the external validity of their findings. It is important to understand the association between the skills of practitioners, patients’ self management behaviours, and subsequent health outcomes for patients with diabetes, but we must not lose sight of the importance of glycaemic control. By international comparison the generally poor glycaemic control found in the Talking Diabetes study is worrying (mean HbA1c of 75 mmol/mol).8 11 Perhaps we should focus again on the setting of appropriate targets by professionals who care for patients with diabetes and the patients themselves. Is it time to raise the bar?


Cite this as: BMJ 2012;344:e2673


  • Research, doi:10.1136/bmj.e2333
  • Research, doi:10.1136/bmj.e2359
  • Competing interests: The author has completed the ICMJE uniform disclosure form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declares: no support from any organisation for the submitted work; no financial relationships with any organisations that might have an interest in the submitted work in the previous three years; no other relationships or activities that could appear to have influenced the submitted work.

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