Type I diabetes: motivational enhancement therapy delivered with CBT by nurse therapists to people with type I diabetes leads to lowering of HbA1C values
Does motivational enhancement therapy (MET) with or without cognitive behavioural lead to improved glycaemic control in people with type I diabetes?
344 adults (18–65 years) with type 1 diabetes for a minimum duration of 2 years, defined as onset at younger than 35 years of age, a current glycated haemoglobin (HbA1C) value between 8.2% and 15.0% and onset of insulin therapy within 6 months of diagnosis or ketones in the urine. The patients had diabetes for a median of 18 years.
London and Greater Manchester; September 2003 and August 2005.
MET or a combination of MET and cognitive–behaviour therapy (CBT), both groups also had usual diabetes care. These groups were compared to usual care alone. The interventions were delivered by nurse therapists and were developed by the researchers.
Primary outcome was HbAlc at 12 months from randomisation. Secondary outcomes included 1 year costs assessed by the Client Service Receipt Inventory at baseline, 6 months and 12 months and quality-adjusted life years (QALYs) at baseline and 12 months.
Randomised controlled trial.
Single blind; technicians assessing HbA1C were blind to allocation, blinding not possible with psychological assessment.
The median HbA1C value for all 344 participants was 9.4% (IQR 8.8–10.2). Treatment differences were adjusted for baseline HbA1C and also depression, fear of hypoglycaemia, diabetes self care activities and quality of life. MET plus CBT significantly reduced adjusted mean 12-month HbA1C compared with usual care (difference: −0.45%, 95% CI 0.16% to 0.79%, p=0.008), although this was not significant when compared with MET alone (difference: −0.30%, 95% CI −0.07% to 0.66%, p=0.11). Compared to usual care, there was no effect of MET alone on HbA1C (difference: −0.16%, 95% CI 0.20% to 0.51%, p=0.38). Reduction in HbA1C was associated with age and HbA1C at baseline (younger participants had a greater reduction in HbA1C, as did those with higher baseline HbA1C). There was no effect of MET or MET plus CBT on secondary outcomes such as depression or quality of life. Both interventions were associated with higher total health and social care costs compared to usual care alone, but there were no differences in societal costs. Cost effectiveness depended on the relative importance placed on HbA1C and QALY outcomes. If value was placed on HbA1C outcomes then MET plus CBT versus usual had a greater probability of being cost effective (at a £20 000 threshold) than did MET alone versus usual care. Conversely, if value was placed on quality-of-life outcomes MET versus usual care had a greater probability of cost effectiveness than the combination. However, for all determinations, probabilities were low.
A combination of MET and cognitive behavioural therapy leads to a reduction in HbA1C values in people with type I diabetes compared to usual care. The probability of cost effectiveness of MET alone or in combination with CBT depends on which outcome is used.
This 3-arm randomised controlled trial of 344 patients with type 1 diabetes compared usual care, usual care + motivational enhancement therapy (MET) and usual care + MET + cognitive behavioural therapy (CBT). The outcome measures included haemoglobin A1c (HbA1c) at 12 months and cost effectiveness. Results indicated that the MET + CBT arm had significantly lower HbA1c than those receiving usual care, while the MET arm was not better than usual care and the two treatment conditions (MET and MET + CBT) were not significantly different from one another. Intervention effect did not vary by demographic variables and the interventions had no effect on depression, body mass index, adherence to diet, quality of life or hypoglycaemic fear.
The strengths of the study include its randomised design, large sample, blinded interviewers, manualised treatment and ongoing supervision and fidelity assessment. In addition, the economic evaluation was thorough. Weaknesses include weak potency of the interventions. This may have been due to the short period of training given to the nurses and/or the limited diabetes education in the treatments, which is known to be an important component of diabetes care.1 There was also a lack of control for attention (time with nurses) in that usual care had fewer nurse meetings than the MET, which had fewer meetings than MET + CBT making it difficult to determine whether attention contributed to the effects of the intervention in the MET + CBT arm. Another major weakness is that a CBT-only arm was not included, so it is difficult to determine whether CBT alone is comparable to, or superior to, MET + CBT. Finally, the study did not include type 2 patients who comprise 90% to 95% of the population of individuals with diabetes, which limits generalisability.
Overall, this study provides evidence that MET + CBT is an effective intervention for individuals with type 1 diabetes. However, future studies need to replicate the findings and address the weaknesses identified above.