Statistics from Altmetric.com
Cochrane Central Register of Controlled Trials (issue 4, 2002), Medline (1966 to January 2003), EMBASE/Excerpta Medica (1980 to January 2003), and PsycINFO (1974 to January 2003); proceedings of conferences on diabetes (1997–2002); reference lists of identified studies and reviews; and leading authors and experts in the field.
Study selection and assessment:
published or unpublished randomised controlled trials (RCTs) in any language comparing psychological interventions (supportive or counselling therapy, cognitive behaviour therapy, brief psychodynamic therapy, or interpersonal therapy) with usual care, education, waiting list, or attention control in patients ⩾18 years of age with type 2 diabetes. Studies with techniques not clearly described were excluded. Individual study quality was assessed using the criteria of Schulz and Jadad (randomisation procedure, allocation concealment, withdrawals, dropouts, intention to treat analysis, and masking of outcome assessors).
changes in glycated haemoglobin (HbA1c) and blood glucose concentrations. Secondary outcomes were changes in body weight and psychological distress.
25 studies met the selection criteria. Pooling of 12 RCTs showed lower HbA1c concentrations in patients who received a psychological intervention than in control group patients (table⇓). Groups did not differ for changes in blood glucose concentration or body weight (table⇓). 5 of 5 RCTs showed reductions in psychological distress with a psychological intervention (table⇓).
In patients with type 2 diabetes mellitus, psychological interventions improve glycaemic control and reduce psychological distress.
Improving self care behaviours, glycaemic control, body weight, and psychological distress are important diabetes treatment goals. The review by Ismail et al supports the findings of other meta-analyses showing that behavioural or psychological interventions in conjunction with diabetes education are valuable in achieving improved glycaemic control.1,2 However, separating the effect of psychological counselling from the usual behavioural component of education is difficult. The authors differentiate between psychological and educational interventions with a broad interpretation of psychological interventions but a narrow definition of health education that excludes behaviour modification. Thus, in addition to established psychotherapy, studies were included if interventions used goal setting, contracts, problem solving, activity scheduling, or stress management. Several of these are common techniques in diabetes education and have been included in the US National Standards for Diabetes Self-Management Education.3 Furthermore, only 7 of 25 studies had interventions done by a psychologist or psychiatrist. In the other 18 studies, interventions were mainly done by nurses, dietitians, or multidisciplinary teams and involved intensive education rather than psychotherapy.
People with diabetes often have psychological problems such as depression, disordered eating, and stress. Surprisingly only 4 studies addressed these issues, and only 2 could be included in the analyses that found a reduction in distress.
The meta-analysis found no difference in weight loss between groups, although some participants did lose weight. These studies did not examine weight loss or gain in relation to glycaemic control but results emphasise the need for more innovative behavioural weight loss interventions for people with type 2 diabetes.
For correspondence: Dr K Ismail, Academic Department of Psychological Medicine, Institute of Psychiatry, London, UK.
Source of funding: in part, Psychiatry Research Trust, London, UK.