Are diabetes self-management programmes for the general diabetes population effective for people with severe mental illness?: a systematic review.

Background: Diabetes self-management education programmes are effective in improving health outcomes in the general population with diabetes. However, it is not known if these programmes include people who also have a severe mental illness (SMI) and, if so, what their outcomes are. The aim of this review was to examine if evaluations of diabetes self-management education programmes included people with SMI, and if so, whether the interventions were beneficial for this population. Methods: The inclusion criteria for this systematic review, defined by PICOS criteria, were: Population - Adults with type 2 diabetes; Intervention - self-management education programme; Comparator – another active intervention or usual care; Outcomes of interest – inclusion of people with SMI and the clinical, behavioural and psychosocial outcomes in this population; Study design - randomised controlled trials. The following bibliographic databases were searched from January 2004 to April 2018: Cochrane Library, Medline, Embase, PsychINFO, Allied and Complimentary Medicine Database, Health Technology Assessment, NHS Economic Evaluations Database and CINAHL. Data were extracted on study characteristics, inclusion and exclusion criteria, participant and intervention characteristics, number of participants with SMI, and outcomes for people with SMI, if reported. Authors were contacted by email for missing data. Results: A total of 410 trials were included. At least 42% of trials did not recruit any participants with SMI. Only nine confirmed inclusion of participants with SMI, of which six provided data on the number recruited. These six trials recruited a total of 1009 participants, of whom 31 (3.1%) had SMI. It was not possible to assess intervention effectiveness for people with SMI as none of the trials reported outcomes for these participants. Conclusions: This systematic review confirms that people with SMI are often excluded from trials of diabetes self-management education, resulting in a lack of an evidence base on which to base treatment paths for this vulnerable population. It cannot be assumed that programmes developed for the general diabetes population meet the needs of people with SMI. Future research needs to examine if and how these programmes could be adapted for people with SMI or if new programmes are required