The generally accepted Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) was followed as a guide for conducting the study.
Literature Search
A rigorous literature search was carried out in four well-known bibliographic databases. These were Web of Science, Scopus, Medline (PubMed) and CINAHL. Hand searches were also done in Google Scholar and grey literature (thesis) for relevant supplementary materials. Searches were conducted for articles published from January 1, 2011 to December 31, 2020 on type 2 diabetes mellitus in the Ashanti region of Ghana. This timeframe was chosen because much researches were found to have been conducted on the subject matter within the period, thereby making the issue to shoot into the limelight in the country. Reference lists of some downloaded articles were consulted to identify other relevant articles. Key words and medical subject headings (MeSH) terms were used for the searches in the above-listed bibliographic databases. Boolean operators (“AND” and “OR”) were used to separate MeSH terms and key words. Based on the objective of the study, the following MeSH terms were used: “Diabetes mellitus”, “Type 2 diabetes”, “Diabetes mellitus, type 2”, “Risk factors” and “Prevalence”. The key words included “Ghana”, “Kumasi”, “type 2 diabetes”, “type II diabetes”, “T2D”, “T2DM”, “DM2”, “diabetes mellitus”, “prevalence”, “risk factors” and “Ashanti region”. The searches focused on articles published on adult diabetes patients (18 years and above). This was because in Ghana, one is considered an adult when he/she is above 18 years [29]. Searches were also narrowed down to only articles published in English for two key reasons. Firstly, because English is the official language in Ghana and second, financial constraints coupled with difficulty of accessing the services of language translators. Authors and experts who have over the years contributed significantly to the literature on diabetes were contacted to inquire about any additional published and unpublished materials. The searches started on May 15, 2021 and were completed on June 30, 2021.
Study selection and quality assessment
After downloading all necessary articles, the study employed the United States Preventive Services Task Force Procedure manual [USPSTF] to determine whether the articles were relevant for meeting the overriding objective of the study [30]. In this regard, a multi-stage screening procedure was adopted to minimise errors, to enhance transparency of outcomes and above all, to make the study producible. Two reviewers reviewed all the articles at all stages. First and foremost, the reviewers read through the titles and abstracts of all retrieved articles for relevance by broadly applying the inclusion criteria. In the event where there were uncertainties about whether or not an article should be included, the reviewers strictly erred on the side of the inclusion criteria. Decision to either include or exclude such articles were made after agreements were reached between the two independent reviewers.
The final screening process commenced when all included articles were established. The two reviewers independently screened the full-text articles for further relevance based on studies’ population, exposure/risk and outcomes (PEO) [31]. Articles were further excluded with reasons, where the main text failed to provide relevant information for achieving the objective of the study. Potential articles to be included in the study were assigned a mutually agreed-upon unique code, to facilitate further assessment.
Data extraction
A standardised form was generated for the study using some generic items and relevant information which were unique to the study. Details on the form included: Author(s) name(s), year of publication, characteristics of respondents, sampling technique, study area, and findings/results of the studies. The two independent reviewers sifted this information concurrently. Differences in opinions regarding some risk factors were resolved by agreement. Finally, to ensure validity and reliability of the instrument, the form was pre-tested on three different studies before adopting it for the data extraction.
(SEE APPENDIX 1) Table 1: Summary of findings of studies included in the review for examining the risk factors of type 2 diabetes mellitus among adult Ghanaians in the Ashanti region of Ghana
Setting
The Ashanti Region is the third largest of the 16 regions in Ghana. With a total land surface area of 24,389 km2, the region occupies about 10 per cent of the total land surface area of the country [32]. Even though the region has the largest population of the other regions, Kumasi, which is its capital accounts for nearly one-third of the region’s population. The region has a population density of 148.1 persons per square kilometres [32]. In terms of vegetation, it is located in the moist semi-deciduous forest belt. However, due to rapid rate of urbanisation, climate change and bush burning, the north-eastern part of the region has reduced to guinea savanna [32]. Due to the region’s abundant mineral and agricultural resources, coupled with its strategic geographical location (between longitude 0.15W and 2.25W, and latitude 5.50N and 7.46N) and spatial interaction with other regions, it is largely referred to as the business hub of the country [33].
Despite its population density, Katey et al. [34] and Ashiagbor et al. [35] found that access to health facilities is one of the major problems facing the people. However, access to primary health care facilities have been adjudged as very efficient. Information on the levels of educational attainment and literacy show that between 40.0 and 50.0 per cent of the population in the region, particularly females, either have no formal education or have only pre-school education [35]. The proportion of the population with basic education vary from 67.7 per cent in the Kumasi metropolis to 86.9 per cent in the Amansie West District [32].