In people with poorly controlled type 1 diabetes, cognitive behaviour therapy combined with motivational enhancement therapy reduces HbA1c after 12 months
- Correspondence to Leah Brennan
Level 6, The Alfred Centre, 99 Commercial Road, Melbourne, VIC 3004, Australia;
Implications for practice
▪ Motivational enhancement therapy (MET) +cognitive–behavioural therapy (CBT) can improve 12-month glycaemic control in adults with poorly controlled type 1 diabetes and is more effective than MET alone or usual care (UC).
▪ MET+CBT and MET do not result in better quality of life, depression, body mass index or diabetes self-care activities.
▪ With extensive training and supervision, nurses can be trained to deliver diabetes-specific MET and CBT.
▪ Cost-effectiveness was not demonstrated.
Implications for research
Further research is required to determine:
▪ active components of combined MET+CBT and patient's perceptions;
▪ maintenance of MET+CBT effects over time;
▪ use of MET+CBT in a variety of formats (eg, group, online).
Type 1 diabetes is associated with a range of biopsychosocial consequences.1 Treatment targets self-care (eg, diet, exercise, medication) aimed at optimising glycaemic control thus minimising the risk of medical/health complications. However, between quarter and half of all adults with type 1 diabetes have suboptimal glycaemic control.1
Maintaining behaviour changes in the absence of immediate consequences is a particular challenge for people with type 1 diabetes. Psychological approaches aimed at enhancing motivation and initiating and maintaining behaviour change may help improve glycaemic control.1
CBT aims to assist the patient develop helpful thoughts and behaviours to facilitate optimal health behaviours.2 Motivational interviewing aims to explore and resolve ambivalence, thus enhancing the patient's readiness to change.3
A three-arm parallel, unblinded, randomised controlled trial (RCT) was used to determine the relative effectiveness of MET+CBT, MET and UC in improving glycaemic control in adults with type 1 diabetes. It was hypothesised that (1) MET+CBT would be more effective than UC; (2) MET would be more effective than UC; (3) MET+CBT would be more effective than MET in improving glycaemic control at 12-month follow-up. Secondary outcomes included assessment of cost-effectiveness, psychosocial outcomes and moderators.
A total of 344 participants met inclusion criteria (eg, 18–65 years, type 1 diabetes for minimum of 2 years, current haemoglobin A1c (HbA1c) 8.2–15%) and were randomised to receive UC (n=121), MET+UC (n=177) or MET+CBT (n=106). UC was based on the UK guidelines for type 1 diabetes care.1 MET comprised four 50 min individual sessions over a 2-month period. CBT comprised eight individual sessions over 4 months.
Six nurses received comprehensive training and supervision to deliver the intervention. Clinician competence in MET and CBT was established before intervention commencement; manuals and workbooks were used to maximise treatment integrity. Treatment integrity was assessed throughout the trial; some nurses demonstrated low competency in some techniques, and there was some overlap of techniques between conditions.
The primary outcome, HbA1c, was measured quarterly. Cost and quality of life were measured 6 monthly. Other secondary outcomes including diabetes specific treatment beliefs, quality of life, psychiatric comorbidity, diabetes self-care behaviours, diabetes adherence and adverse effects were measured annually.
An analysis of covariance, adjusting for baseline values, was used to assess intervention group difference at 12 months and potential treatment moderators. Logistic regression was used to calculate the odds of severe hypoglycaemic episodes.
For 305 participants (88%), 12-month HbA1c was available. After adjusting for baseline HbA1c levels, 12-month HbA1c levels were significantly lower in MET+CBT (n=95) compared with UC (n=105) (adjusted mean difference 0.45%, 95% CI 0.12% to 0.79%, p=0.008), non-significantly lower in MET (n=105) compared with UC (n=105) (adjusted mean difference 0.16%, 95% CI −0.20% to 0.51%, p=0.38) and non-significantly lower in MET+CBT compared with MET (adjusted mean difference 0.30%, 95% CI −0.07% to 0.66%, p=0.11).
There were no group differences in hypoglycaemic episodes at 12 months. Treatment condition was not associated with secondary outcomes. Younger participants, and those with higher baseline HbA1c, demonstrated greater improvements in 12-month HbA1c in the MET+CBT condition. There were no other significant treatment moderators. 82.9% of the MET group and 56% of the MET+CBT group attended all sessions. Patients with higher baseline depression scores were more likely to drop out of the study.
Both intervention groups had significantly higher health and social costs than the UC group. There were no significant differences in quality-adjusted life years (QALYs) between groups. Incremental cost-effectiveness ratios were lower for MET+CBT compared with UC than for MET alone. MET+CBT had a good probability of cost-effectiveness compared with MET using HbA1c; however, it was dominated by MET and had low probabilities of cost-effectiveness using QALYs.
This study provides evidence that (1) MET+CBT, but not MET alone, can improve glycaemic control in poorly controlled type 1 diabetes and (2) nurses with comprehensive training and supervision can deliver MET and CBT.
Strengths of this study include use of a RCT design and valid comparison condition, inclusion of ‘difficult-to-treat’ patients, attempts to ensure treatment fidelity and assessment of a range of biopsychosocial outcomes and moderators. Limitations include failure to standardise treatment length, the absence of a CBT-only comparison group, poor attendance rate for the MET+CBT condition, nurses' failure to achieve competence in all techniques and inability to consistently differentiate MET and CBT techniques.