Article Text
PDF
Abstract
Objective Contemporary data on the attainment of optimal diabetes treatment goals and the burden of diabetes complications in adult populations with type 2 diabetes in Africa are lacking. We aimed to document the current status of attainment of three key indicators of optimal diabetes care and the prevalence of five diabetes complications in adult African populations with type 2 diabetes.
Methods We systematically searched Embase, PubMed and the Cochrane library for published studies from January 2000 to December 2020. Included studies reported any information on the proportion of attainment of optimal glycated haemoglobin (HbA1c), blood pressure (BP) and low-density lipoprotein cholesterol (LDLC) goals and/or prevalence of five diabetes complications (diabetic peripheral neuropathy, retinopathy, nephropathy, foot ulcers and peripheral arterial disease). Random effect model meta-analysis was performed to determine the pooled proportion of attainment of the three treatment goals and the prevalence of five diabetes complications.
Results In total, 109 studies with a total of 63 890 participants (53.3% being females) were included in the meta-analysis. Most of the studies were conducted in Eastern African countries (n=44, 40.4%). The pooled proportion of attainment of an optimal HbA1c, BP and LDLC goal was 27% (95% CI 24 to 30, I2=94.7%), 38% (95% CI 30 to 46, I2=98.7%) and 42% (95% CI 32 to 52, I2=97.4%), respectively. The pooled prevalence of diabetic peripheral neuropathy, retinopathy, diabetic nephropathy, peripheral arterial disease and foot ulcers was 38% (95% CI 31 to 45, I2=98.2%), 32% (95% CI 28 to 36, I2=98%), 31% (95% CI 22 to 41, I2=99.3%), 19% (95% CI 12 to 25, I2=98.1%) and 11% (95% CI 9 to 14, I2=97.4%), respectively.
Conclusion Attainment of optimal diabetes treatment goals, especially HbA1c, in adult patients with type 2 diabetes in Africa remains a challenge. Diabetes complications, especially diabetic peripheral neuropathy and retinopathy, are highly prevalent in adult populations with type 2 diabetes in Africa.
- epidemiology
- epidemiology
- quality in health care
- general diabetes
Data availability statement
Data are available on reasonable request.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
Strengths and limitations of this study
To our knowledge, it is the first systematic review and meta-analysis to simultaneously investigate the status of attainment of the three key diabetes treatment goals and the burden of five common diabetes complications in an adult indigenous African population with type 2 diabetes.
The systematic review and meta-analysis included a large number of studies that assessed the extent of attainment of diabetes treatment goals and the prevalence of diabetes complications based on recommendations or definitions by internationally recognised associations.
There was high heterogeneity among the studies included in the meta-analysis.
A relative number of studies included in the meta-analysis had low to moderate quality on assessment.
Introduction
Globally, the burden of diabetes mellitus (DM) continues to exponentially rise to epidemic proportions, disproportionately affecting low-income and middle-income countries. The recent 2021 International Diabetes Federation (IDF) estimates show that about 24 million adults (1 in 22 adults) live with DM in Africa. The IDF also predicts that the greatest future increase in the prevalence of DM will occur in Africa because of the predicted ageing of Africa’s currently very young populations, as well as increasing urbanisation and associated lifestyle changes.1 This will ultimately lead to an immense strain on weak healthcare systems that are poorly structured and inadequately financed to manage non-communicable diseases (NCDs) like DM.2
In addition, the rates of undiagnosed DM continue to increase in Africa. Among the IDF regions, Africa has the highest proportion of undiagnosed diabetes: about 54% of all cases.1 The majority of patients are diagnosed late with coexisting debilitating complications, and suboptimal diabetes care remains common in most clinical settings in Africa.3 This could be explained by low awareness about DM, healthcare systems that are structured mainly to manage communicable diseases as opposed to NCD, low screening rates of DM to ensure early diagnosis, low availability of affordable essential diagnostic tests and medicines for DM and knowledge–practice gaps among healthcare practitioners.2 4–6
Published diabetes treatment guidelines by most international organisations like the IDF and American Diabetes Association (ADA) recommend targets of glycated haemoglobin (HbA1c) level of <7% (53 mmol/mol), blood pressure (BP) <140/90 mm Hg and low-density lipoprotein cholesterol (LDLC) <2.6 mmol/L (100 mg/dL) as key indicators of optimal diabetes care.7–9 Attainment of these treatment goals in diabetes care ultimately translates to reduced risk of onset and progression of diabetes complications and mortality.
Despite the increasing burden of DM and its related complications, late diagnosis of diabetes and prevalent suboptimal diabetes care in clinical settings in Africa, there is an information gap regarding the current status of attainment of the recommended diabetes treatment goals and the prevalence of common diabetes complications to inform targeted strategies or interventions to reduce diabetes-related morbidity and mortality. This systematic review and meta-analysis aimed to document the proportion of attainment of optimal HbA1c, BP and LDLC goals and the prevalence of five diabetes complications (diabetic peripheral neuropathy, nephropathy, retinopathy, foot ulcers and peripheral arterial disease) in adult native populations with type 2 diabetes in Africa.
Methods
This systematic review and meta-analysis was conducted according to the criteria outlined in the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement.10 The PRISMA checklist is available as an online supplemental table 1. The study protocol was registered in the PROSPERO International Prospective Register of systematic reviews (CRD42020215576).
Supplemental material
Search strategy
We searched Embase, PubMed and the Cochrane library for published studies from January 2000 to December 2020. The following search terms were used after discussion with a medical librarian: “Quality of diabetes care” OR “Indicators of diabetes care” OR “status of diabetes care” OR “diabetes care” OR “glycaemic control” OR “blood pressure control” OR “lipid profile control” OR “screening of diabetes complications” OR “diabetes complications” OR “screening for diabetic retinopathy” OR “screening for diabetic peripheral nephropathy” OR screening for diabetic neuropathy” OR screening for diabetic foot ulcers OR “screening for peripheral arterial disease” OR “prevalence of diabetic retinopathy” OR “prevalence of diabetic peripheral nephropathy” OR “prevalence of diabetic peripheral neuropathy” OR “prevalence of diabetic foot ulcers” OR “prevalence of peripheral arterial disease”, AND “type 2 diabetes mellitus” OR “type 2 diabetes” AND Algeria OR Angola OR Benin OR Botswana OR ‘‘Burkina Faso’’ OR Burundi OR Cameroon OR ‘‘Cape Verde’’ OR ‘‘Central African Republic’’ OR Chad OR Comoros OR ‘‘Democratic Republic of Congo’’ OR Djibouti OR Egypt OR ‘‘Equatorial Guinea’’ OR Eritrea OR Ethiopia OR Gabon OR Gambia OR Ghana OR Guinea OR ‘‘Guinea Bissau’’ OR ‘‘Ivory Coast’’ OR ‘‘Cote d’Ivoire’’ OR Kenya OR Lesotho OR Liberia OR Libya OR Libya OR Madagascar OR Malawi OR Mali OR Mauritania OR Mauritius OR Morocco OR Mozambique OR Namibia OR Niger OR Nigeria OR Rwanda OR ‘‘Sao Tome’’ OR Senegal OR Seychelles OR ‘‘Sierra Leone’’ OR Somalia OR ‘‘South Africa’’ OR “South Sudan” OR Sudan OR Swaziland OR Tanzania OR Togo OR Tunisia OR Uganda OR Zaire OR Zambia OR Zimbabwe OR ‘‘Central Africa’’ OR ‘‘West Africa’’ OR ‘‘Western Africa’’ OR ‘‘East Africa’’ OR ‘‘Eastern Africa’’ OR ‘‘North Africa’’ OR ‘‘Northern Africa’’ OR ‘‘Southern Africa’’ OR ‘‘sub Saharan Africa’’ OR ‘‘sub-Saharan Africa’’ OR Africa.
In addition, references of included articles were hand-searched for any other original articles. The search and selection were restricted to studies written only in the English language.
Study selection criteria
The preliminary screening of titles and abstracts to identify potentially eligible articles was done by two independent reviewers (NC and DK). This was followed by removing all duplicates. After the initial screening, full texts of the potentially eligible studies were retrieved and closely reviewed for eligibility.
The inclusion criteria of studies were: cross-sectional, cohort or randomised controlled trials published between January 2000 and December 2020 in English language, studies reporting any data on proportion of adult patients with type 2 diabetes who attained the recommended optimal HbA1c, BP or LDLC targets and residing in African countries and studies reporting data on any of prevalence of diabetic nephropathy, peripheral neuropathy, retinopathy, foot ulcers or peripheral arterial disease in adult patients with type 2 diabetes in African countries.
Any disagreements that arose were resolved by consensus. We excluded retrospective studies, case series and reports, studies published in languages other than English and studies whose full texts could not be retrieved.
Data extraction
After identifying the eligible original studies, they were collated and sent to additional reviewers to extract the relevant study information using a Microsoft Excel 2016 form. The information of interest that was extracted from the eligible studies included: the last name of the first author and year of publication, country(ies) and region(s) of Africa where the study was conducted, type of study design, number of study participants, the mean age of study participants, the proportion of female participants, the proportion of participants with a current or history of smoking, the proportion of participants on oral hypoglycaemic agents, insulin, lipid-lowering agents (statins) and antihypertensive agents, mean body mass index (BMI) and HbA1c of study participants, the proportions of participants with optimal HbA1c, BP and LDLC targets and the prevalence of diabetic nephropathy, peripheral neuropathy, retinopathy, foot ulcers and peripheral arterial disease.
Operational definitions
All included studies defined optimal targets of HbA1c, BP and LDLC as <7% (53 mmol/mol), <140/90 mm Hg and <2.6 mmol/L or 100 mg/dL, respectively, as recommended by the IDF and ADA diabetes treatment guidelines.9 11
The definitions and measurements of diabetes complications greatly varied between studies. The following definitions were used for each diabetes complication by the various studies: micro/macroalbuminuria and/or an estimated glomerular filtration rate <60 mL/min/1.73 m2 for the presence of diabetic nephropathy, signs and symptoms suggestive of peripheral neuropathy, use of neuropathy screening scores like neuropathy disability score, Michigan Neuropathy Screening Instrument, neuropathy symptom score and 10 g monofilament testing for the presence of diabetic peripheral neuropathy, presence of lesions like soft or hard exudates, cotton wool spots, microaneurysms, neovascularisation and retinal haemorrhages on funduscopy for diabetic retinopathy, presence of foot ulcers on clinical inspection for diabetic foot ulcers and the presence of measured ankle brachial index <0.9 using Doppler studies for peripheral arterial disease.
Assessment of quality of studies
The quality of all eligible studies included in the systematic review and meta-analysis was assessed using the Newcastle-Ottawa Scale (NOS).12 This was done by two independent authors (NC and SNL). The total score of the adapted scale is eight stars. Studies with more than six stars were considered high quality, while those with 5 and 6 stars, and <5 stars were considered of moderate and low quality.
Study outcomes
The study outcomes were the pooled proportions of attainment of the recommended optimal HbA1c, BP and LDLC goals and the pooled prevalence of diabetic nephropathy, peripheral neuropathy, retinopathy, foot ulcers and peripheral arterial disease in adult patients with type 2 diabetes in Africa.
Data analysis
All analyses were performed using STATA V.16.0 statistical software (Stata Corp, USA). The descriptive data of all eligible studies included in the systematic review and meta-analysis like age, gender, the proportion of participants on specific glucose-lowering agents, BMI and HbA1c were summarised using frequencies and 95% CIs and mean±SD.
For the continuous variables, the average estimated value was obtained from each of the studies, and this was used in the final analysis, while for the categorical variables, the proportions were estimated for each of the studies and used in the final analysis.
The pooled proportions of achievement of optimal HbA1c, BP and LDLC goals and the prevalence of diabetic nephropathy, peripheral neuropathy, retinopathy, foot ulcers and peripheral arterial disease were determined using a random effect model meta-analysis and presented in forest plots. The DerSimonian and Laird method was used for pooling random effects estimates.13
The heterogeneity of studies was assessed using the I2 value and corresponding 95% CIs. Based on the Cochrane collaboration guide, the I2 values of 0%–40%, 30%–60%, 50%–90% and 75%–100% were considered not important, moderate, substantial and considerable levels of heterogeneity, respectively.14 To further explore heterogeneity effects across studies, we conducted a meta-regression analysis to assess whether the heterogeneity could be explained by the study level characteristics, that is, age, sex of participants and region, in which the study was conducted. The age, BMI and sex of the participants was defined as the estimated mean age and BMI of participants and the proportion of females from each of the study, respectively. The region of the study was defined as the area (Northern, Southern, Eastern, Western, and Central Africa) where the study was conducted. One effect measure per study was considered in the meta-regression. All the variables were included in the model together to assess for variability.
We assessed the presence of publication bias using the Egger test of bias with p<0.05 indicating significant publication bias.15 A narrative review was also used to present the study results. Information about all included studies was also summarised in tables.
We also performed a sensitivity analysis based on the NOS scores of the studies (excluding moderate and low-quality studies) and compared the analysis with all the eligible studies and with only high-quality studies to identify any differences in the pooled estimates of the rates of attainment of optimal diabetes treatment goals and the prevalence of the five diabetes complications.
Patient and public involvement
The main research question and outcomes of interest of the systematic review and meta-analysis were informed by the need to understand the burden of diabetes complications in patients with type 2 diabetes in Africa and the extent of attainment of optimal diabetes care to inform strategies aimed to improve optimal management of diabetes in the region. Because it was a systematic review and meta-analysis, we did not involve patients in its design, recruitment and conduct.
Results
Figure 1 summarises the article selection in a PRISMA flow diagram.
PRISMA flow diagram of selection of eligible studies. PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-Analyses.
The literature search returned a total of 835 articles. From these, 222 duplicates were removed. Titles and abstracts of the remaining 613 articles were reviewed, and 235 articles were identified for full-text retrieval. Of the 235 articles, 126 were excluded, and the remaining 109 articles were included in this systematic review and meta-analysis. A total of 48 and 89 eligible studies contained information on optimal diabetes treatment goals and diabetes complications, respectively, while 28 studies reported information on both.
The 126 excluded articles included five studies published in French language, 21 retrospective studies, six studies with general populations (not entirely patients with type 2 DM), 18 studies whose full texts were unable to be retrieved and 76 studies that did not report outcomes of interest.
Characteristics of included studies
The majority of studies were performed in Eastern African countries (44, 40.4%).3 16–58 The proportion of studies conducted in Western, Northern, Southern and Central Africa was 22% (n=24 studies),3 59–80 16.5% (n=18 studies),81–99 15.6% (n=17 studies)100–116 and 8.3% (n=9 studies),3 59 117–123 respectively. Three studies were conducted in more than one region of Africa (Western, Central and Eastern).3 58 59 Most of the studies were cross-sectional in design (100, 91.7%).
Considerable heterogeneity was noted across the studies with the I2 value ranging from 97.4% to 99.3% for studies reporting the burden of diabetes complications and 94.7%–98.7% for studies reporting the extent of attainment of optimal diabetes treatment goals. However, on meta-regression after adjusting for age and sex of study participants, and region where each study was conducted, the heterogeneity based on I2 of studies on the prevalence of diabetes complications decreased, ranging from 1.4% for studies on diabetic foot ulcers to 95.6% for studies on diabetic nephropathy. For studies on the proportion of attainment of optimal treatment goals, the heterogeneity also decreased, to 56.3%, 92.1% and 95.4%, for studies reporting optimal HbA1c, LDLC and BP goals.
Characteristics of study participants
Table 1 summarises the characteristics of all participants in the studies included in the systematic review and meta-analysis.
General characteristics of all participants (n=63 890) included in the systematic review and meta-analysis
The studies had a total of 63 890 participants (ranging from 40 to 11 866) with 53.3% being female. The mean±SD age, BMI and HbA1c of the participants was 54.9±4.7 years (ranging from 40.5 to 63.9 years), 27.9±0.5 kg/m2 (ranging from 20.6 to 42.9 kg/m2) and 9.0±1.5% (ranging from 6.5% to 13.9%), respectively. Among the studies that reported data on the type of glucose-lowering therapies used by participants, treatment with oral hypoglycaemic agents, insulin, statins and antihypertensives was reported in about 65% (95% CI 34 to 96.6), 31.3% (95% CI 26.3 to 36.2), 25.7% (95% CI 0.5 to 86.7) and 73.3% (95% CI 64.1 to 82.5) of participants, respectively.
Assessment of study quality and publication bias
The assessment of the quality of studies and funnel plots assessing publication bias are summarised in online supplemental table 2 and online supplemental figures 1–8, respectively.
Supplemental material
Supplemental material
Supplemental material
Supplemental material
Supplemental material
Supplemental material
Supplemental material
Supplemental material
Supplemental material
Based on the NOS, 84 (77.1%) of the included studies were of high quality, with 17 (15.6%) studies and 8 (7.3%) studies being of moderate and low quality, respectively.
Regarding the assessment of publication bias, there was observed publication bias, especially in studies about the prevalence of diabetic nephropathy, peripheral neuropathy and attainment of optimal BP control. The proportion of studies investigating the prevalence of diabetic nephropathy, peripheral neuropathy, peripheral arterial disease, retinopathy and foot ulcers located within the funnel plot was 30% (n=12), 46.1% (n=13), 55.6% (n=10), 57% (29) and 90% (n=26), respectively. About 46%, 65% and 73% of studies that reported the proportion of attainment of optimal BP, HbA1c and LDLC treatment goal were located within the funnel plot, respectively.
Extent of attainment of optimal HbA1c, BP and LDLC goals
Data on the reported proportions achieving the three diabetes treatment goals are summarised in tables 2–4 and as forest plots in figures 2–4.
Forest plot summarising studies on the proportion of attainment of an optimal low-density lipoprotein cholesterol goal in percentage. ES, effect size.
Forest plot summarising studies on the proportion of attainment of an optimal blood pressure goal in percentage. ES, effect size.
Forest plot summarising studies on the proportion of attainment of an optimal glycated haemoglobin goal in percentage. ES, effect size.
Indicators of optimal glycated haemoglobin goal
Indicators of optimal blood pressure (BP) goal
Indicators of optimal LDLC goal
Data on attainment of optimal HbA1c, BP and LDLC goals were reported in 34 studies,3 20 21 23 35–37 44–47 59–61 63 64 67 84 87 90 92 93 97–99 104 105 111 116 117 120 124 26 studies,3 18 20 21 24 36 40 41 45 47 61 64 67 70 77 87 90 91 96 97 105 107 111 113 120 121 and 11 studies,21 37 39 47 61 87 90 94 97 111 116 respectively. The pooled proportion of attainment of an optimal HbA1c, BP and LDLC goal in the respective studies was 27% (95% CI 24 to 30, I2=94.7%), 38% (95% CI 30 to 46, I2=98.7%) and 42% (95% CI 32 to 52, I2=97.4%), respectively.
The lowest proportion of attainment of optimal HbA1c was reported in a study performed in Egypt (4.4%)97 and the highest in a study performed in Nigeria (52.5%).64 Among studies reporting the extent of attainment of an optimal BP goal, the proportion ranged from 1.5% in a study performed in Uganda47 to 85.9% in a study performed in Ethiopia.24 Among the studies reporting information on the optimal LDLC goal, attainment of optimal targets ranged from 20.4% in a study performed in Botswana111 to 84.8% in a study performed in Sudan.94
Regarding the attainment of the diabetes treatment goals in each region of Africa surveyed, the lowest and highest proportion of attainment of an optimal HbA1c goal was noted in the Central (20%, 95% CI 16 to 23) and Western regions (37%, 95% CI 29 to 46), respectively. For the attainment of an optimal BP control, the Western region had the least proportion (31%, 95% CI 20 to 43), while the Northern region had the highest (42%, 95% CI 24 to 61). An optimal LDLC target was least achieved in the Southern region (27%, 95% CI 24 to 30) and most achieved in the Northern region (53%, 95% CI 32 to 74).
Prevalence of diabetic retinopathy, peripheral neuropathy, nephropathy, foot ulcers and peripheral arterial disease
Information on the pooled and specific prevalence of diabetes complications as reported by the different studies is summarised in tables 5–9 and as forest plots in figures 5–9.
Forest plot summarising studies on the prevalence of diabetic retinopathy. ES, effect size.
Forest plot summarising studies on the prevalence of diabetic foot ulcers. ES, effect size.
Forest plot summarising studies on the prevalence of diabetic nephropathy. ES, effect size.
Forest plot summarising studies on the prevalence of diabetic neuropathy. ES, effect size.
Forest plot summarising studies on the prevalence of peripheral arterial disease. ES, effect size.
Prevalence of diabetic nephropathy
Prevalence of diabetic peripheral neuropathy
Prevalence of diabetic retinopathy
Prevalence of diabetic foot ulcers
Prevalence of peripheral arterial disease
The prevalence of diabetic retinopathy, nephropathy, peripheral neuropathy, foot ulcers and peripheral arterial disease was reported in 51 studies,3 19 24 26 28 30 38 41 48 51 53 54 56–58 65–67 70 72 74 76 77 81 82 86 88 89 91 95–97 102–107 109 112–116 118 120–123 125 40 studies,3 19 21 27 28 30–32 38 46 48 53 57 60 62 64 66 67 69 70 76 81 82 86 88 89 91 96 97 100 105 108–110 113 114 117–119 125 36 studies,3 19 25 27 28 30 33 34 37 38 43 48 51–53 55 57 58 65 67 68 73 76 79 81 85–88 96 97 105 109 118 125 29 studies3 16–19 21 22 25 27 29 38 42 43 48 49 51 53 54 57 58 67 80 85 87 95 97 113 114 125 and 18 studies,3 20 25 30 43 47 50 52 61 67 70 75 78 85 86 91 97 105 respectively.
Prevalence of diabetic peripheral neuropathy and retinopathy
Diabetic peripheral neuropathy and retinopathy were the most prevalent diabetes complications in the included studies with a pooled prevalence of 38% (95% CI 31 to 45, I2=98.2%) and 32% (95% CI 28 to 36, I2=98%), respectively. A wide variation was noted in the prevalence of diabetic peripheral neuropathy across the studies, with prevalence ranging from 4% in a study conducted in Eritrea51 to 83.3% in a study conducted in Nigeria.68 A study by Makwero and colleagues109 conducted in Lesotho reported the lowest prevalence of diabetic retinopathy of 4.7%, while the study by Megalla and colleagues97 conducted in Egypt reported the highest (90%).
According to the regions of Africa surveyed, the lowest and highest prevalence of diabetic peripheral neuropathy was noted in the Central (22%, 95% CI 18 to 27) and Western regions (61%, 95% CI 45 to 75), respectively. Studies conducted in the Eastern region reported the lowest prevalence of diabetic retinopathy (23%, 95% CI 19 to 28) while studies conducted in the Northern region reported the highest prevalence (51%, 95% CI 37 to 65).
Prevalence of diabetic nephropathy, peripheral arterial disease and foot ulcers
The pooled prevalence of diabetic nephropathy, peripheral arterial disease and foot ulcers in the included studies was 31% (95% CI 22 to 41, I2=99.3%), 19% (95% CI 12 to 25, I2=98.1%) and 11% (95% CI 9 to 14, I2=97.4%), respectively.
The prevalence of diabetic nephropathy and peripheral arterial disease ranged from 2.2% in Ethiopia19 to 90% in Nigeria64 and 2.7% in a study performed in Morocco91 to 52.5% in a study performed in Nigeria,78 respectively. Regarding the burden of diabetic foot ulcers, there was also an observed heterogeneity, with prevalence ranging from 0.4% in Ethiopia53 to 86.7% in Egypt.97
Studies conducted in the Central, Eastern and Southern regions reported a comparable prevalence of diabetic nephropathy (22%, 25% and 28%, respectively) with the highest prevalence reported in studies conducted in the Western region (47%). Regarding the prevalence of PAD, studies conducted in the Southern (8%, 95% CI 6 to 10) and Western (29%, 95% CI 13 to 48) regions reported the lowest and highest prevalence, respectively. A comparable prevalence of diabetic foot ulcers was noted in studies conducted in the Southern, Western and Eastern regions (7%, 8% and 10%, respectively), with the highest prevalence noted in studies conducted in the Northern region (21%).
On sensitivity analysis considering only high-quality studies, the pooled prevalence of the five diabetic complications and the proportion of attainment of the three optimal diabetes treatment goals did not differ from those obtained in the preliminary analysis with all eligible studies included. The pooled prevalence of diabetic foot ulcers, peripheral arterial disease, diabetic nephropathy, diabetic retinopathy and diabetic peripheral neuropathy after sensitivity analysis was 9% (95% CI 7 to 12, I2=92.9%), 20% (95% CI 13 to 28, I2=98.4%), 31% (95% CI 21 to 42, I2=99.4%), 33% (95% CI 28 to 37, I2=98.2%) and 40% (95% CI 32 to 48, I2=99%), respectively. The pooled proportion of attainment of optimal HbA1c, BP and LDLC treatment goal was 27% (95% CI 23 to 30, I2=94.5%), 37% (95% CI 29 to 46, I2=99.0%) and 43% (95% CI 31 to 55, I2=97.9%), respectively.
Discussion
To our knowledge, this is the first systematic review and meta-analysis to simultaneously document the proportion of attainment of the three key indicators of optimal diabetes care (HbA1c, BP, and LDLC goals) and the burden of five diabetes complications in an indigenous adult population with type 2 diabetes in Africa. In this study of a total of 63 890 study participants, we report that, generally, a small proportion of adult patients with type 2 diabetes in Africa attain optimal diabetes treatment targets, especially HbA1c and BP goals (less than 40%). In addition, diabetes complications are relatively common with diabetic neuropathy being the most prevalent (38%) followed by diabetic retinopathy (32%), nephropathy (31%), peripheral arterial disease (19%) and foot ulcers (11%).
Proportions of attainment of the optimal diabetes treatment goals
A wide heterogeneity in the attainment of the optimal diabetes treatment goals was noted across all five regions of Africa. This could probably be explained by the marked differences in the populations studied, healthcare systems and knowledge-practice gaps among healthcare practitioners.
Similar to our study findings, achievement of optimal HbA1c, BP and LDLC treatment goals has also been widely reported to be a significant clinical challenge in several studies performed in Caucasian and Asian populations with type 2 diabetes in high-income and middle-income countries.126–131 In one large registry-based study of >100 000 adults with a self-reported diagnosis of diabetes carried out between 1999 and 2010 in USA, 33.4%–48.7% of adult patients with diabetes did not achieve the recommended HbA1c, BP and LDLC treatment targets. Less than 15% met all the three treatment targets in addition to smoking cessation.126
Similarly, a low proportion of achievement of an optimal HbA1c target was also reported by a large international, multicentre observational study of 2704 multiracial adult populations with diabetes from 10 countries (two from Africa, five from the Middle East and three from South Asia). About 46% of the participants were Caucasian. An optimal HbA1c goal of <7% (53 mmol/mol) was reported in only 25.8% of the participants.128
In the Japan Epidemiology Collaboration on Occupational Health study, which enrolled 3070 adult employees of large manufacturing companies, optimal HbA1c, BP and LDLC goals as recommended by the ADA were noted in 44.9%, 76.6% and 27.1% of participants, respectively. Only 11.2% of participants attained all three treatment goals.129
The burden of diabetes complications in Africa
Regarding studies on the burden of diabetes complications in Africa, there were few that investigated the prevalence of diabetic foot ulcers and peripheral arterial disease with diabetic retinopathy, peripheral nephropathy and neuropathy being the most studied. Diabetic peripheral neuropathy and retinopathy remain the most prevalent diabetes complication and diabetic foot ulcers the least prevalent.
With regards to the prevalence of diabetic foot ulcers, an earlier published systematic review and meta-analysis on the characteristics, prevalence and outcomes of diabetic foot ulcers in Africa by Rigato et al132 reported a pooled prevalence of diabetic foot ulcers of 13%, a finding close to what we observed (11%). In another systematic review and meta-analysis on the prevalence of diabetic peripheral neuropathy in African populations with DM, Shiferaw et al133 reported a slightly higher overall prevalence of 46% compared with what we found in our study (38%) while including fewer studies (n=23).
Similar to our study, considerable heterogeneity was also reported in the documented prevalence of the varied diabetes complications in Africa in most previously published systematic reviews. This may be due to variations in clinical definitions of diabetes complications in the studies. Burgess et al134 and Achigbu et al135 reported a wide disparity in the prevalence of diabetic retinopathy in the included studies of 7%–62.4%, and 13%–82.6%, respectively. Noubiap et al136 in a systematic review on the burden of diabetic nephropathy in 2015 reported an overall prevalence of chronic kidney disease in patients with diabetes ranging between 11% and 83.7%. Johnston et al in a systematic review that aimed to assess the epidemiological and clinical reports regarding Peripheral arterial disease (PAD) in Sub-saharan Africa (SSA) documented the prevalence of PAD in patients with diabetes as reported by three studies to range from 39% to 52%.137
Compared with Caucasian and Asian adult populations with type 2 diabetes, our study has demonstrated that adult African patients are disproportionately affected by complications of DM. The Joint Asia Diabetes Evaluation programme that undertook comprehensive risk assessments of 3687 adult patients with type 2 DM recruited from seven Asian countries reported a prevalence of peripheral arterial disease, diabetic neuropathy, macroalbuminuria and microalbuminuria and diabetic retinopathy of 3.1%, 15%, 18.8% and 20.4%, respectively.138
The National Health and Nutrition Examination Survey conducted from 1988 to 1994 and 1999–2018 in USA in 1486 non-pregnant adults (aged ≥20 years) with newly diagnosed diabetes (diagnosed within the past 2 years) also documented a low burden of most diabetes complications. Diabetic foot ulcers, peripheral arterial disease, diabetic retinopathy, neuropathy and nephropathy (albuminuria) were prevalent in 6.3%, 9.2%, 12.1%, 14.5% and 18.7%, respectively.139
The documented low proportions of attainment of optimal diabetes treatment goals (optimal HbA1c, BP and LDLC targets) in Africa is associated with an increased risk of onset and progression of diabetes complications, hence increasing morbidity and mortality in addition to causing a significant economic strain on the meagre health resources. This generally observed low proportion of attainment of key diabetes treatment goals and high prevalence of diabetes complications, notably diabetic neuropathy, retinopathy and nephropathy in Africa, exists broadly due to challenges related to screening, diagnosis and management of DM.
Awareness of diabetes in the general African population and healthcare practitioners remains very poor, resulting in delayed diagnosis of diabetes. The challenge of ready access to affordable essential diabetes medicines like insulin and statins and diagnostic tests or equipment like glucometers for home self-monitoring of glucose, HbA1c and lipid profile tests remains highly prevalent in most African countries.140–144
Effective management of diabetes and its related cardiovascular risk factors like hypertension and dyslipidaemia in most healthcare settings in Africa also remains a significant clinical challenge.3 Most healthcare facilities especially the lower tier ones lack local or institution-specific comprehensive diabetes treatment guidelines to guide healthcare practitioners on how to optimally manage diabetes, in addition to the evident knowledge–practice gaps among healthcare practitioners.2
Healthcare systems in most African countries remain poorly structured to optimally manage most NCDs like diabetes along with an inadequately funded health sector. Most African countries have not yet fulfilled the 2001 Abuja Declaration of allocating 15% of their national annual budget to the health sector.2 145
This systematic review and meta-analysis had its strengths and limitations. To our knowledge, it is the first to simultaneously investigate the status of attainment of the three key diabetes treatment goals and the burden of five common diabetes complications in an adult indigenous African population with type 2 diabetes. The systematic review and meta-analysis included a large number of studies that assessed the extent of attainment of diabetes treatment goals and the prevalence of diabetes complications based on recommendations or definitions by internationally recognised associations.
It also had its limitations. There was considerable heterogeneity in the included studies. This could be explained by the differences in study sites (tertiary vs low-tier hospitals or private vs public hospitals), patient characteristics (age, duration of diabetes, coexisting medical conditions), regions where the studies were conducted and diagnostic modalities used to identify diabetes complications. The systematic review also excluded studies published in French, which is the official language of some African countries. However, these were very few. There was evidence of publication bias in some of the included studies especially studies investigating the prevalence of diabetic nephropathy and peripheral neuropathy and the proportion of attainment of an optimal BP goal. About 23% of the included studies were moderate and low quality on assessment using the NOS for cross-sectional studies.
Conclusion
Achievement of optimal diabetes treatment goals, especially HbA1c and BP, in adult African patients with type 2 diabetes remains low in Africa. Diabetes complications especially diabetic peripheral neuropathy and retinopathy also remain highly prevalent. Implementation of universal diabetes screening and education initiatives coupled with improving knowledge about diabetes management among healthcare practitioners and ready access to affordable essential diabetes diagnostic tests and medicines in Africa are integral in improving overall optimal diabetes care and reducing the burden of diabetes complications.
Considering the projected future increase in the prevalence of diabetes globally, especially in the African region, there is an urgent need to address glaring gaps in diabetes care and to develop simple and pragmatic interventions to improve treatment outcomes and reduce the burden of diabetes complications.
Data availability statement
Data are available on reasonable request.
Ethics statements
Patient consent for publication
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Twitter @PWSsekamatte
Contributors DK and NC: conceived the research idea, performed the preliminary screening of titles and abstracts to identify potentially eligible articles and wrote the initial draft of the manuscript; DK, NC, IA-B, SNL, ISe, APK and SN: retrieved full texts and identified the eligible articles; KK, SNL, AP-K, SN, PS, FB, LEM, WO, TDM, NEN, ISa: extracted data from the identified eligible articles; DK and ISe performed the data analysis and interpretation; NC, KK and SNL: performed the assessment of the quality of studies; KS, PH, LtB, JV, RvC and JAC: offered additional data interpretation and supervised this work. All the authors reviewed the different versions of the manuscript and read and approved the final draft of the manuscript. DK is the overall guarantor and accepts full responsibility for the work and/or the conduct of the study, had access to the data, and controlled the decision to publish.
Competing interests None declared.
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.