This systematic review is reported following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (37) (Additional File 1). The review was prospectively registered at the International Prospective Register of Systematic Reviews (PROSPERO) on 28th November 2018 (https://www.crd.york.ac.uk/PROSPERO/, registration number: CRD42018116373). A detailed summary of the review methods is set out in protocol, which has been published elsewhere (38).
Eligibility criteria
The review included studies if they; i) were conducted in adults aged 18 years or over and diagnosed with type 2 diabetes, ii) were any observational studies (cross-sectional, case-control or cohort), iii) examined the association between subthreshold depression and any diabetes self-care behaviours and iv) were published in English. Studies were excluded if they were qualitative studies, review articles, and non-peer reviewed (grey) literature.
The self-care behaviours that were considered in the study were healthy eating, being physically active, monitoring of blood glucose, taking medication, not smoking, and foot care (27). Currently, there is no agreed definition of subthreshold depression. Different terms – e.g. minor, subthreshold – are used to describe people with subthreshold depression (19). For the purposes of our review we defined subthreshold depression as minor, subthreshold, subclinical, subsyndromal depression or a mood disorder that does not meet the diagnostic criteria for major depressive disorder (39).
Data source and search strategy
The search strategy was developed in consultation with a medical librarian experienced in systematic review database searching. We conducted a systematic search of the literature in the following databases: MEDLINE, EMBASE, PsycINFO, Emcare, and CINAHL. Databases were searched using a combination of medical subject headings (MeSH) and keywords relating to “type 2 diabetes”, “depression” and “self-care." Boolean operators ("AND," and "OR"), proximity operators ("ADJ," and "N") and truncation were incorporated into the search strategy as required to cater for the different use of terms. Search results were limited to English. Additional file 2 shows the Medline (ovid) search strategy used for this systematic review.
Study selection
All the references identified were imported into Endnote X9 (40), a reference manager software program. Following the removal of duplicates in Endnote, citations were imported into Covidence systematic review software (41). All studies were then assessed for eligibility in Covidence on a two-step process, 1. title and abstract screening, and 2. screening of full texts. Two authors MS and FA, applied the predetermined eligibility criteria to all the articles by screening titles and abstracts independently. All disagreements concerning inclusion or exclusion of papers were judged by a third author (AG). Full-text reading of the articles was again done independently by two authors (MS and FA). The final decision on the inclusion of the full- text articles was made after a discussion with the review team members.
Data extraction
Two authors (MS and RG) independently extracted data from each of the included studies in the data extraction tool. A third author compared the authors’ data and resolved inconsistencies by referring to the full-text article and thorough discussion. The following data were extracted: citation, country of study, aim of the study, population characteristics (age, gender), study design (cross-sectional, case-control or cohort) and setting (community or hospital), sample size, sample size calculation, sampling technique, data source (survey or secondary data), definition of subthreshold depression, measure used to examine subthreshold depression and self-care behaviours, analysis, confounder variables adjusted and key observation of the study. There was no need to contact the authors of any studies for the study information.
Quality Appraisal
All the included studies were assessed for methodological rigour using the Joanna Briggs Institute (JBI) Critical Appraisal Checklist tool (42). The JBI has separate checklists for cross-sectional (8 criteria), case-control (10 criteria), and cohort studies (11 criteria). Each component of the checklist was rated as yes, no, unclear, or not applicable. Two authors (MS and RG) independently evaluated the quality of each study, and disagreement between them were resolved by discussion within the review team.
Data analysis
Due to a smaller number of studies and variability in the outcome measures, results could not be combined by meta-analysis. Instead, a narrative synthesis of the study was conducted. Tables and narrative summaries are used to present the study and participant characteristics and findings of the studies.