Rapid Responses to:

RESEARCH: Kirsty Winkley, Sabine Landau, Ivan Eisler, and Khalida Ismail Psychological interventions to improve glycaemic control in patients with type 1 diabetes: systematic review and meta-analysis of randomised controlled trials BMJ 2006; 333: 65 [Abstract] [Full text]

Rapid Responses: Submit a response to this article

Rapid Responses published:

Not the magic bullet?

Frank J Snoek   (29 June 2006)

What works for whom and why?

Khalida Ismail   (24 July 2006)

Not the magic bullet? 29 June 2006
Frank J Snoek, Professor of Medical Psychology VU University Medical Centre, Van der Boechorststraat 7, 1081 BT Amsterdam, Netherlands Send response to journal: Re: Not the magic bullet?	Winkley and colleagues are to be commended for their thourough review of psychological interventions in Type 1 diabetes, both in youth and adults (1). It is well recognized that psychological factors play a key role in diabetes management, but it is less certain which strategies can effectively assist patients in achieving strict glycaemic control. On the basis of their review and meta-analysis Winkley et al. conclude that so far psychological interventions have shown to have modest effects with regard to lowering HbA1c, particularly in adults. The strenght of this well-written review lies in bringing together the evidence on the effectiveness of psychological interventions in people with diabetes. There are however also weaknesses,that limit the generalizability of the findings and are not discussed by the authors. First, and most importantly, the question is not so much if psychologial counseling is effective in lowering HbA1c in patients with Type 1 diabetes per se. Rather, the question is whether psychological counseling helps to improve glycaemic control in the patients with longstanding poor control. One could debate the definition of poor control criterion, but most clinicians would probably agree on repeated HbA1c's over 8% as a cut-off. Data on baseline HbA1c's for the various studies are not reported, nor if the interventions evaluated were designed specifically for poorly-controlled patients. Often intervcentions are offered to a mixed patient population, too small to allow for subgroup analyses. This brings us to the second problem, namely the variety of interventions reviewed in the meta-analysis - the old problem of apples and pears. Interestingly, one of our own studies evaluating the effects of a nurse-led monitoring of well-being procedure in an outpatient setting, using a computerised assessment and feedback procedure, was included in the review and categorsised under 'counselling' (2). Our study was not designed as a counselling programme,and certainly not for poorly-controlled specifically. Rather it was designed to test the effectivenss of implementing systematic monitoring of well-being as part of clincial routine, as recommendend by the International Diabetes Federation and WHO, that had in fact never had been tested in a randomised controlled trial. This study should probably not have been included in this review. Moreover, and related to the first comment on glycaemic control, mean HbA1c in this population was below 8% and thus left not much room for improvement. Still, the intervention was effective in promoting emotional well-being and increasing satisfaction with care. Whether these psychological benefits are helpful in maintaining good diabetes control in due course is not known and warrants further research. Indeed, more sophisticated research is needed to enhance our understanding of the complex interaction between medical and psychological factors in producing health outcomes. The review by Winkley and associates is timely and offers much food for thought. We should however be cautious in oversimplifying reality by councluding that psychology does not work. Complex problems warrant complex solutions and psychology by no means is a magic bullet. Nor are education or medical technology on their own. The challenge is to bring together these elements in comprehensive care programmes, that effectively address the different problems of patients who have difficulty managing their diabetes. 1. Winkley K, Landau J, Eisler I, Ismail K. Psychological interventions to improve glycaemic control in patients with type 1 diabetes: systematic review and meta-analysis of randomised controlled trials. BMJ/doi 10.113/bmj.38874.652569.55 2. Pouwer F, Snoek FJ, Ploeg van der HM, Ader HJ, Heine RJ. Monitoring of psychological well-being in outpatients with diabetes: effects on mood, HbA1c and the patients' evaluation of the quality of diabetes care, A randomised controlled trial. Diabetes Care 2001, 24: 1929-1935. Competing interests: None declared
What works for whom and why? 24 July 2006
Khalida Ismail, Senior Lecturer in Liaison Psychiatry Institute of Psychiatry, London, SE5 9RS Send response to journal: Re: What works for whom and why?	Dear Sir We thank Professor Snoek for raising several important points and for opening the debate as to which psychological treatments are effective for which subgroups of people with diabetes, and perhaps even why. We would like to emphasise that our review found that, on the basis of current evidence, overall psychological treatments had a modest but significant effect in lowering HbA1c in children and adolescents but not in adults. We are in full agreement with Professor Snoek that one of the key research questions, certainly now for future studies, is whether psychological treatments improve glycaemic control in people with persistent or longstanding suboptimal glycaemic control. Our systematic review showed that only 2 children and adolescent and 4 adult studies specified that the intervention was targeted for people with suboptimal glycaemic control but we noted that the mean baseline HbA1c for each study, which we did indeed report, was suboptimal. As more randomised controlled trials addressing this question are published it will be possible to synthesise the evidence in the years to come. Professor Snoek also highlights another problem we encountered; the descriptions of the psychological intervention of many studies were meagre and we had to exclude those studies that were too ambiguous. Snoek and his team have pioneered interventions in mental health and diabetes and their valuable work, including the 2 studies incorporated in our review, contributes to ensuring that the standards and methods that are used to evaluate psychological treatments in mental health settings, such as the description and integrity of psychotherapy, therapist factors and manual development,(1) are also applied to the evaluation of complex non-pharmacological interventions in diabetes. It was our interpretation that monitoring of wellbeing, discussing and giving feedback in an ‘explorative/nonjudgmental way’ and ‘using active listening and exploration of feelings’ as described by the authors themselves (2) is a psychotherapeutic technique and a counselling style. It is standard Cochrane Collaboration procedure to use secondary outcomes in the systematic synthesis of evidence for an intervention so it was legitimate for us to use HbA1c data when it was reported even if this was not a primary endpoint of the intervention in that study(3). 1. Waltz J, Addis M, Koerner K, Jacobson N. Testing the integrity of a psychotherapy protocol: assessment of adherence and competence. Journal of Clinical and Consulting Psychology 1993;61:620-630. 2. Pouwer F, Snoek F, van der Ploeg H, Ader H, Heine R. Monitoring of psychological well-being in outpatients with diabetes. Effects on mood, HbA1c and the patient's evaluation of the quality of diabetes care: a randomized controlled trial. Diabetes Care 2001;24:1929-35. 3. Alderson P, Green S, Higgins Je. Cochrane Reviewers’ Handbook 4.2.2 [updated December 2003]. 2003. Competing interests: None declared