Introduction:

3.6. Associated risk factors

The summary of adjusted Odds Ratios (aOR) from analytical studies (case-control and cohort) confirms several major determinants of type 2 diabetes in the African population :

• Age ≥45 years : aOR=3.12 (IC95%:2.45-3.98, p<0.001).
• Overweight / Obésity (BMI≥25 kg/m2) : aOR=2.84 (IC95%:2.21-3.65, p<0.001).
• Associed high blood pressure : aOR=2.42 (IC95%:1.95-3.01, p<0.001).
• Family history of diabetes : aOR=2.15 (IC95%:1.72-2.69, p<0.001).
• Physical inactivity / sedentarity lifestyle : aOR=1.78 (IC95%:1.41-2.25, p<0.001).

3.7. Chronic complications and undiagnosed diabetes

    3.7.1.  Vascular complications

The pooled analysis limited to studies describing the clinical profile of diabetic patients (n=38) reports a high frequency of complications at the time of assessment :

• Peripheral neuropathy: 38.2% (95% CI: 32.5-44.2%) ;
• Diabetic retinopathy: 31.4% (95% CI: 26.2-37.0%) ;
• Nephropathy / Microalbuminuria: 27.5% (95% CI: 22.1-33.5%) ;
• Cardiovascular diseases: 14.2% (95% CI: 11.0-18.0%) ;
• Diabetic foot ulcer: 11.8% (95% CI: 9.2-14.8%) ;
• Strokes (CVA): 7.6% (95% CI: 5.8-9.8%).

3.7.2. Proportion of undiagnosed diabetes

Among the general population studies that included systematic biological screening (n=24), the estimated proportion of people with diabetes who were unaware of their glycemic status was 53.8% (95% CI: 46.2-61.2%).

3.8. Meta-regression, sensitivity analysis, and publication bias

3.8.1. Meta-regression

The multivariable meta-regression explains 42.6% of the variance between studies (R^2=42.6%). The place of residence (urban vs rural, p<0.001), geographic region (p=0.002), and year of publication (p=0.014) are the three main sources of heterogeneity identified.

3.8.2. Sensitivity analysis

The sensitivity analysis by sequentially leaving out one study at a time (leave-one-out) showed minimal variations in the overall combined prevalence (ranging between 8.2% and 8.6%), confirming the stability of the results.

3.8.3. Assessment of publication bias

Egger's test highlighted a slight asymmetry in the funnel plot (t=2.41, p=0.018). Using Duval and Tweedie's Trim-and-Fill method with imputation of 11 potential studies adjusted the combined prevalence to 7.6% (95% CI: 6.6–8.7%), indicating a moderate impact of publication bias on the overall effect measure.

4. Discussion

4.1. Key results

This systematic review and meta-analysis primarily aimed to estimate the combined prevalence of type 2 diabetes (T2D) among adults living in Africa based on studies published between 2015 and 2025. In total, 84 studies, representing a combined population of 142,850 adult participants, were included in the systematic review and meta-analysis. The combined prevalence of T2D was estimated at 8.4%, with very high heterogeneity between the studies.

This estimate confirms that type 2 diabetes is now a major public health problem in Africa. The increasing burden of diabetes is part of an epidemiological transition marked by rapid urbanization, changes in eating habits, less physical activity, rising overweight and obesity, and a gradually aging population [1–5].

According to recent estimates from the International Diabetes Federation (IDF), the number of people living with diabetes keeps rising worldwide, with a particularly worrying increase in low- and middle-income countries [1].

This development is a major challenge for African health systems, which have to deal at the same time with infectious diseases, non-communicable diseases, and health emergencies.

The estimated proportion of 53.8% of undiagnosed diabetic people in studies that carried out systematic biological screening is also a major finding. It highlights the big gap between the actual burden of diabetes and the proportion of cases detected by health systems. Diabetes underdiagnosis in Africa has already been widely documented and is a significant barrier to preventing complications [6,7].

The analysis of the associated factors found several determinants that are regularly reported, including older age, overweight or obesity, high blood pressure, a family history of diabetes, and physical inactivity. These factors are consistent with the currently recognized pathophysiological mechanisms and behavioral determinants of type 2 diabetes [8–10].

Finally, studies on complications reported a significant frequency of neurological, retinal, kidney, and cardiovascular issues, highlighting the potential consequences of a late diagnosis and poor control of cardiometabolic risk factors.

4.2. A concerning rise in type 2 diabetes in Africa

The combined prevalence of 8.4% observed in this meta-analysis confirms the growing importance of type 2 diabetes in Africa. However, this estimate should be interpreted with caution given the very high heterogeneity observed between studies.

The rise of diabetes in Africa is happening in a context of rapidly changing lifestyles. Urbanization, changes in diet, less physical activity, and rising obesity are gradually increasing the risk of heart and metabolic problems [3–5].

This transition is happening while African health systems are still heavily burdened by infectious diseases, maternal and child health issues, and health crises. Diabetes, therefore, falls into a situation of double disease burden, which is typical of many countries on the continent [3,4].

This trend is especially worrying because type 2 diabetes can stay without symptoms for several years. A significant number of people with it can go undiagnosed until complications in the heart, kidneys, nerves, or eyes show up [1,6,7].

So, the prevalence observed in this meta-analysis might only represent part of the actual burden of type 2 diabetes in Africa, especially in populations with limited access to screening.

4.3. A strong heterogeneity between African studies

One of the main methodological findings of this meta-analysis is the presence of a very high heterogeneity between the included studies.

This heterogeneity is understandable given the extreme diversity of the African continent. The studies differed particularly in the country where they were conducted, the recruitment setting, the age structure of the populations, the sampling methods, the diagnostic criteria, and the characteristics of the health systems.

Differences in diagnostic methods can also affect the estimates. Some studies relied on fasting blood sugar, others on HbA1c, oral glucose tolerance, a previous medical diagnosis, or even taking antidiabetic medication.

The statistical heterogeneity observed shouldn't just be seen as a weakness of the meta-analysis. It can also reflect real epidemiological differences between African populations.

In this context, using a random-effects model and doing subgroup analyses are particularly relevant [11,12].

The goal isn't just to get a single continental estimate, but also to explore the factors that might explain the differences seen between the studies.

4.4. Regional differences and urbanization

The studies included were mostly represented in West Africa and East Africa, while Central Africa and some parts of Southern Africa were less represented.

This distribution doesn't necessarily mean that type 2 diabetes is more common in regions with the most studies. It could reflect differences in research capacity, funding, university infrastructure, and availability of epidemiological data.

Urbanization is still an important factor in the epidemiological transition in Africa. People living in cities are more exposed to environments that encourage a sedentary lifestyle, eating high-calorie foods, and having less daily physical activity [3–5,13].

However, type 2 diabetes should no longer be seen as exclusively urban. Changes in diet, rising overweight rates, and lifestyle shifts are also affecting rural populations.

Prevention strategies should therefore be adapted to both urban and rural settings and take into account the social, economic, and cultural factors specific to each community.

4.5. Aging, overweight, and obesity

Older age was among the factors most commonly associated with type 2 diabetes in the studies included. Aging is particularly linked to reduced insulin sensitivity, changes in body composition, and a gradual decline in pancreatic β-cell function [8,9].

This question is particularly important in the context of Africa's demographic transition. The gradual increase in life expectancy should lead to a rise in the number of elderly people exposed to chronic diseases.

Being overweight and obese are also major factors in type 2 diabetes. The buildup of fat tissue, especially around the organs, promotes insulin resistance and metabolic problems [8–10].

Preventing type 2 diabetes can't just be about screening for high blood sugar. It also needs to include promoting a balanced diet, regular physical activity, and maintaining a healthy body weight.

4.6. High blood pressure and cardiometabolic risk

The link between high blood pressure and type 2 diabetes found in the included studies highlights the importance of combining cardiometabolic risk factors.

Diabetes and high blood pressure share several risk factors, including age, obesity, being inactive, and poor eating habits. Having both really ups the risk of heart and kidney problems [14–17].

In African contexts, this combination is particularly worrying because both diseases are sometimes diagnosed late. Bringing blood pressure checks, diabetes screening, body measurements, and heart risk assessments into primary care could help catch high-risk people earlier.

Managing type 2 diabetes shouldn't just focus on blood sugar control anymore. Current recommendations favor a more holistic approach that includes blood pressure, cholesterol, weight, and whenever it's available and appropriate, treatments that have shown cardiovascular and kidney benefits [15–17].

4.7. Family history and physical inactivity

A family history of diabetes was also among the factors linked to type 2 diabetes. This link probably reflects the interaction between genetic susceptibility and shared environmental and behavioral factors within families [8,9].

Physical inactivity was also a commonly reported factor. The decrease in physical activity is one of the major consequences of lifestyle changes linked to urbanization and changes in work activities and modes of transportation.

Regular physical activity improves insulin sensitivity, helps control weight, and lowers cardiometabolic risk [18].

In Africa, efforts to promote physical activity should be adapted to local realities. They can include not only structured exercise, but also walking, active transport, work activities, and community programs.

4.8. Major importance of undiagnosed diabetes

One of the most worrying findings of this review is the estimated 53.8% of people with diabetes who go undiagnosed among populations that underwent systematic biological screening.

In other words, in these populations, more than one in two people with diabetes didn't know their blood sugar status.

This result is a major public health issue. Late diagnosis reduces the chances of secondary prevention and exposes patients to a prolonged period of uncontrolled high blood sugar, which can lead to complications [6,7].

Several factors can contribute to this underdiagnosis in Africa: limited access to healthcare, the cost of lab tests, the distance to medical facilities, inadequate screening programs, and low risk perception among people without symptoms [3,4,6].

Strengthening targeted screening in primary healthcare could therefore be a particularly relevant strategy, especially for people with high blood pressure, obesity, a family history of diabetes, or other cardiometabolic risk factors.

4.9. Chronic complications

The included studies reported a significant burden of chronic complications, including peripheral neuropathy, retinopathy, kidney damage, and cardiovascular complications.

These complications are a major cause of disability, early death, and health care costs for people living with diabetes [19–21].

The high frequency of certain complications can especially be explained by late diagnosis, prolonged high blood sugar, and difficulties in accessing comprehensive care.

In many African contexts, access to screening for retinopathy, nephropathy, and neuropathy is still limited. Managing type 2 diabetes should go beyond just prescribing a blood sugar-lowering treatment.

An integrated approach including blood sugar control, blood pressure, lipids, weight, foot exams, retinopathy screening, and kidney function assessment seems necessary.

Heart diseases hold a special place because they contribute a lot to sickness and death among people with diabetes [15–17].

4.10. Implications for African health systems

The results of this meta-analysis have several important implications for health policies.

First, DT2 should be more integrated into primary healthcare strategies. Targeted screening of people with risk factors could allow for earlier identification of cases.

Secondly, prevention efforts should start before diabetes appears. Promoting a balanced diet, physical activity, and maintaining a healthy weight should be strengthened in communities, schools, workplaces, and healthcare facilities [4,18].

Thirdly, access to essential medicines and diagnostic technologies needs to be improved. Effective care notably requires the regular availability of antidiabetic drugs, blood sugar measuring devices, and tests to detect complications.

Access to newer antidiabetic treatments, like SGLT2 inhibitors and GLP-1 receptor agonists, is also becoming an emerging issue given their proven heart and kidney benefits. However, their cost and availability still significantly limit their use in many African countries [15–17].

Finally, developing national registries and diabetes surveillance systems would help improve data availability and more effectively guide resource allocation.

4.11. Study strengths

This systematic review and meta-analysis has several strong points.

First, it covers a recent period, from 2015 to 2025, allowing for a summary of contemporary data on type 2 diabetes in Africa.

Secondly, several international and regional databases were searched, including PubMed/MEDLINE, Scopus, Web of Science, and African Journals Online, supplemented by Google Scholar and a manual search of the references.

Third, the selection of studies and data extraction were done independently by two reviewers, which helped reduce the risk of selection and extraction errors.

Fourth, using a random-effects model, subgroup analyses, a leave-one-out sensitivity analysis, and a check for publication bias allows for a thorough look at how robust the results are [11,12,22–25].

Finally, this study didn't just focus on the prevalence of type 2 diabetes. It also looked at risk factors, chronic complications, and undiagnosed diabetes, giving a broader view of the epidemiological burden of type 2 diabetes in Africa.

4.12. Study limits

Several limits still need to be taken into consideration.

The first one is about the very high heterogeneity observed between the studies. This limits the direct comparability of the estimates and calls for a cautious interpretation of the combined prevalence of 8.4%. This value shouldn't be seen as a uniform prevalence applicable to all African countries.

Secondly, the diagnostic methods weren't consistent. Some studies relied on biological measures, while others used a previous medical diagnosis, treatment history, or self-reported data.

Third, most of the studies were cross-sectional. This type of design is good for estimating prevalence, but it doesn't allow for establishing a temporal or causal relationship between risk factors and the development of type 2 diabetes.

Fourth, some African regions were less represented in the available literature, which can limit how generally the results can be applied.

Fifth, limiting it to publications in French and English may have introduced a language bias.

Finally, the analyses regarding risk factors, complications, and undiagnosed diabetes were based on subsets of the 84 included studies. So, these secondary estimates should be interpreted taking into account the number and characteristics of the studies available for each analysis.

4.13. Research perspectives

Future research should focus on representative population studies using standardized diagnostic criteria and harmonized protocols.

A particular effort should be made for countries and regions that are currently underrepresented in the scientific literature. Carrying out standardized national surveys would help improve the comparability of estimates between countries and track changes in diabetes over time.

Longitudinal studies are also needed to better figure out the timing of the relationship between obesity, high blood pressure, physical inactivity, socio-economic factors, and the onset of type 2 diabetes.

Interventional research should also evaluate the effectiveness and feasibility of community screening strategies, lifestyle interventions, and integrated care models for chronic diseases that are adapted to African health systems.

Finally, future meta-analyses will need to clearly distinguish between the number of studies and the number of estimates. When the same study reports prevalence separately for men and women, it counts as a single study, even if it contributes to both subgroups. This distinction is crucial to avoid artificially inflating the number of studies included in sex-stratified analyses.

5. Conclusion

This systematic review and meta-analysis highlights a significant burden of type 2 diabetes among adults in Africa, with an estimated combined prevalence of 8.4% across the included studies. The very high heterogeneity observed reflects the diversity of epidemiological, demographic, and socio-economic contexts on the continent.

The high proportion of undiagnosed diabetics, estimated at 53.8% in studies with systematic biological screening, is particularly worrying. It highlights the existence of a large pool of undetected diabetes and the need to strengthen early screening.

Old age, being overweight and obesity, high blood pressure, family history, and physical inactivity appear to be important factors linked to type 2 diabetes. At the same time, neurological, kidney, eye, and heart complications make up a big part of the disease's clinical burden.

Faced with this trend, African countries need to strengthen strategies for primary prevention, early screening, and integrating diabetes care into primary health services. Improving access to essential medicines, biological monitoring, and screening for complications is also a priority.

Beyond individual care, fighting type 2 diabetes in Africa requires a multi-sector approach that combines nutrition policies, promotion of physical activity, improvement of the urban environment, strengthening of monitoring systems, and the development of health strategies suited to local realities.