1. Key messages

In 2017, 6.1% of the Belgian population was diagnosed with diabetes. However, more than one in three people with diabetes do not know they have the disease, which brings the estimated true prevalence of diabetes to 10%.
Diabetes prevalence is increasing over time as a result of both the ageing of the population and a true increase in the risk of developing diabetes.
The risk of developing diabetes is higher in Wallonia and Brussels than in Flanders, and is higher for individuals with a lower socio-economic status. Socioeconomic differences in the prevalence of diabetes are especially found for the prevalence of unknown or insufficiently controlled diabetes.

2. Background

Diabetes is a condition that is characterized by recurrent high blood sugar levels. If left untreated, diabetes may cause long-term complications such as foot ulcers, damage to the eyes, chronic kidney disease, and cardiovascular disease. A diabetic diet and exercise are important parts of diabetes management, but in a large number of cases, drug therapy will also be required.

A distinction is made between three main types of diabetes:

  • Type 1 diabetes, resulting from an autoimmune destruction of the insulin-producing beta cells in the pancreas;
  • Type 2 diabetes, due to the body's inefficient use of insulin, mainly caused by a combination of excessive body weight and insufficient exercise; and
  • Gestational diabetes, occurring in pregnant women without a previous history of diabetes.

In absence of an exhaustive diabetes register, information on the use of antidiabetic treatment or on diabetes referring nomenclature is considered a good proxy of the prevalence of the diagnosed diabetes. In Belgium, this information is available through the InterMutualistic Agency (IMA-AIM), a platform where data gathered from the seven Belgian health insurance funds are collected and analyzed. IMA-AIM data are, amongst others, made available through the IMA-AIM Atlas [1].

In the IMA-AIM database, diabetes prevalence is estimated based on the number of insured people with delivery of antidiabetics (ATC code A10) or with diabetes referring nomenclature (diabetes convention, diabetes pass, diabetes care trajectory). Women who gave birth during the year under review are excluded in order to exclude gestational diabetes.

Socio-economic characteristics are scarce in the IMA-AIM database. The status of “insured people having right to an increased compensation” is the only available proxy indicator, and it has only two values, yes or no. Persons with a low income or specific medical characteristics such as invalidity are entitled to an increased compensation status.

Since diabetes is a disease with insidious onset, many cases remain undiagnosed. Information on non-diagnosed diabetes prevalence is available in the first edition of the Belgian Health Examination Survey 2018 (BELHES) [2], which aims to collect objectives indicators on the health of the population in representative sample of people aged 18+. In the BELHES, diabetes prevalence is estimated based on measurement of fasting blood glucose and glycated hemoglobin linked with the self-reported data on diabetes from the Health Interview Survey [3].

3. Diabetes prevalence


In 2017, 6.1% of the Belgian population was diagnosed with diabetes according to IMA-AIM database. The prevalence of diabetes increases with age, and is higher among men, especially in the older age groups. However, the results of the BELHES showed that more than one in three people with diabetes do not know they have the disease. When taking into account these undiagnosed cases, the diabetes prevalence increases to 10%.

The BELHES further showed that 18% of the patients on diabetes medication are not well controlled. In other words, 5% of the population is suffering from diabetes either without being aware of it, or without proper diabetes control.

Prevalence of diabetes by age and sex, Belgium, 2017
Source: IMA-AIM Atlas [1]

Trends and regional differences

The prevalence of diabetes is highest in Wallonia and lowest in Flanders, despite the relatively higher age of the Flemish population. The relatively low diabetes prevalence in Brussels is the result of the young age structure: when corrected for age, the diabetes prevalence in Brussels becomes higher than the Belgian average. In Wallonia more people are unaware that they have the disease than in Brussels and Flanders.

From 2007 to 2017, the prevalence of diabetes has increased in all three regions, as a result of both the ageing of the populations and a true increase in the risk of developing diabetes.

  • Crude
  • Age-standardized

Crude prevalence of diabetes in Belgium and its regions, 2007-2017
Source: IMA-AIM Atlas [1]

Age-standardized prevalence of diabetes in Belgium and its regions, 2007-2017
Source: Own calculations based on IMA-AIM Atlas [1]

Socio-economic differences

The risk of developing diabetes is nearly twice as high among individuals with an increased compensation status than among individuals with a normal compensation status. The prevalence of diabetes has been increasing evenly in both groups. The BELHES has shown that people with a lower education are significantly more likely to suffer from ignored or poorly controlled diabetes compared to people with a higher education (RR=3.6, adjusted for age and gender). This could indicate that there are socioeconomics inequalities in the screening and follow up of diabetes.

Age-standardized prevalence of diabetes by compensation status, Belgium, 2007-2017
Source: Own calculations based on IMA-AIM Atlas [1]

4. Read more

View the metadata for this indicator


Age-standardized prevalence
The age-standardized prevalence is a weighted average of the individual age-specific prevalences using an external standard population. It is the prevalence that would be observed if the population had the age structure of the standard population. Since age has an important influence on the risk of diabetes, this standardization is necessary when comparing several populations that differ with respect to their age structure. Here, the Belgian 2017 mid-year population is used as standard population. 
Increased compensation status
Persons with a low income are entitled to an increased compensation status. They pay less for healthcare and have other financial benefits. Currently, this is the only variable available in the IMA-AIM Atlas that allows assessing socio-economic differences.


  1. InterMutualistic Agency Atlas. http://atlas.aim-ima.be/base-de-donnees
  2. Belgian Health Examination Survey (BELHES), Sciensano, 2018. https://his.wiv-isp.be
  3. Belgian Health Interview Survey (BHIS), Sciensano, 1997-2018. https://his.wiv-isp.be