Review: case management and team changes are particularly effective strategies for reducing HbA1c concentrations in type 2 diabetes
Q Which quality improvement (QI) strategies help to improve glycaemic control in patients with type 2 diabetes mellitus?
Medline (1966 to April 2006) and Cochrane Effective Practice and Organisation of Care Group database (which includes the results of periodic searches of Medline, EMBASE/Excerpta Medica, and CINAHL), and bibliographies of key articles.
Study selection and assessment:
randomised controlled trials (RCTs), quasi-randomised trials, and controlled before-after studies assessing interventions that met the definition for ⩾1 of 11 types of QI strategies (audit and feedback, case management, team changes, electronic patient registries, clinician education, clinical reminders, facilitated relay of clinical information to clinicians, patient education, promotion of self management, patient reminder systems, or continuous QI) and reported mean pre-intervention and post-intervention glycated haemoglobin (HbA1c) concentrations in adult outpatients with type 2 diabetes. 66 trials (reported in 58 articles) met the inclusion criteria (50 RCTs, 3 quasi-randomised trials, and 13 controlled before-after trials.
Meta-analysis showed that overall, QI interventions reduced HbA1c concentrations by 0.42% (95% CI 0.29 to 0.54) over a median follow up of 13 months. Trials with mean baseline HbA1c concentrations ⩾8.0% (n = 45) showed greater reductions than those with baseline concentrations <8.0% (n = 21) (0.54% v 0.20%, p = 0.005). As well, studies with <180 participants showed greater reductions than those with ⩾180 participants (0.61% v 0.27%, p = 0.004).
Pooled effect sizes (controlling for mean baseline HbA1c concentrations and study sample sizes) were statistically significant for all but 2 strategies: clinician reminders and continuous QI. Team change strategies (26 trials) reduced mean HbA1c concentrations by 0.67% (CI 0.43% to 0.91%), and case management strategies (26 trials) reduced concentrations by 0.52% (CI 0.31 to 0.73). Patient education (38 trials), patient reminders (14 trials), electronic patient registries (8 trials), and clinician education (20 trials) were each associated with mean reductions of 0.40–0.49%.
Interventions that included team changes were associated with additional reductions of 0.33% (CI 0.12 to 0.54) compared with interventions that did not include team changes. As well, interventions that included case management were associated with additional reductions of 0.22% (0.00 to 0.44).
Among case management trials (n = 26), those that allowed medication changes without a physician’s prior approval had greater additional reductions than those that did not (0.49%, CI 0.19 to 0.78). Among trials assessing team changes (n = 26), those with multidisciplinary teams had greater additional reductions than those without multidisciplinary teams (0.37%, CI 16 to 58).
Several quality improvement strategies reduce HbA1c concentrations in patients with type 2 diabetes. 2 strategies, case management and team changes, are associated with the largest reductions in HbA1c concentrations and with additional reductions in interventions that included these strategies compared with those that did not.
- Sandra L Upchurch, RN, PhD, CDE
The importance of good glycaemic control is critical to the prevention of serious complications in patients with diabetes. Diabetes care includes not only self care by the individual but also medical care management. Recent reviews have summarised the results of interventions to improve glycaemic control, either by diabetes self management education or by medical management approaches.1,2 A strength of the meta-analysis by Shojania et al is the use of an inclusive taxonomy of quality improvement strategies to classify studies that include aspects of both self care and medical management. 11 categories were delineated for analysis. The concept of the taxonomy is complex and also a limitation in the sense that the definitions of categories are not always mutually exclusive, and some have at least one other strategy included. Only 5 studies evaluated a single strategy. Although most categories of strategies showed at least some improvement in glycaemic control, 2 (case management and team changes) showed the largest improvements. From the earliest3 to more recent trials,4 case management has shown promise in improving diabetes outcomes. Importantly, however, Shojania et al showed that programmes that allowed nurse or pharmacist case managers to make medication changes without waiting for physician approval produced reductions in HbA1c approximately twice as large as other case management interventions. In fact, case management programmes that did not allow case managers to make independent medication adjustments worked no better than the other strategies. Adjustment of medications by case managers is influenced by scope of practice, which will vary by governing regulations. Clearly, as the number of people with diabetes increases in both the developed and developing world, health worker roles may need to change to include case management that incorporates more independence in medication adjustment.
For correspondence: Dr K G Shojania, Ottawa Health Research Institute and Department of Medicine, University of Ottawa, Ottawa, Ontario, Canada.
Sources of funding: Agency for Healthcare Research Quality and Department of Veterans Affairs.