Classification into five subgroups alters treatment strategies for type 2 diabetes, Swedish study finds

Researchers at Universidade of Lund, at Suécia, conducted a prospective cohort study of patients diagnosed with type 2 diabetes.
The work classified individuals into five subgroups using clinical variables and machine learning.
The results showed significant variations in the risk of complications and mortality over up to 14 years of follow-up.
This approach reinforces the importance of personalized strategies for preventing and managing the disease.
- Identified subgroups
- Heavy autoimmune diabetes (SAID), with autoimmune features.
- Heavy insulin-deficient diabetes (SIDD), marked by severe insulin deficiency.
- Heavy insulin-resistant diabetes (SIRD), with strong insulin resistance.
- Mild obesity-related diabetes (MOD), associated with moderate obesity.
- Mild age-related diabetes (MARD), linked to aging.
Origin of classification
The division into subgroups emerged from an analysis carried out in 2018 by a team led by Swedish researchers.
They used data such as the presence of anti-GAD antibodies, beta-cell function measured by HOMA2-β, age at diagnosis, body mass index and HbA1c levels.
The clustering method made it possible to group patients with similar disease progression profiles.
This initial classification already indicated differences in glycemic control and the risk of complications.
The current study extended observation to long-term outcomes.

Differences in prognosis
Longer follow-up revealed distinct patterns between groups.
Patients in the SIRD subgroup were at increased risk of chronic kidney disease and other complications associated with insulin resistance.
The SIDD group showed an increased risk of diabetic retinopathy and other issues linked to severe insulin deficiency.
The SAID subgroup maintained characteristics similar to type 1 diabetes in some aspects.
These findings highlight that not all patients with type 2 diabetes progress in the same way.
Risks of comorbidities
Each subgroup exhibited a specific profile of comorbidities.
For example, individuals with severe insulin resistance faced a higher incidence of cardiovascular problems.
Other groups had a predominance of microvascular complications.
Mortality also varied, with higher rates in certain subgroups depending on dominant risk factors.
These variations reinforce the heterogeneity of the disease.
Implications for treatment
The results indicate the need for different management approaches.
Patients in subgroups with high renal risk may benefit most from kidney-protective medications.
Those with severe insulin deficiency require early intensive insulin therapy.
The classification helps predict progression and choose preventive interventions.
This contributes to advances in precision medicine applied to diabetes.
Study Details
The research involved a large cohort of adult patients in Suécia.
Average follow-up reached several years, allowing robust analysis of long-term events.
The authors emphasized the clinical relevance of stratification.
Data included national registries to track comorbidities and deaths.
The publication took place in a magazine specializing in endocrinology and diabetes.
Variations in outcomes
One of the mild groups, related to obesity, showed a more benign evolution compared to the severe ones.
Another, associated with age, presented moderate risks.
These distinctions help identify patients who require more intensive monitoring.
Early stratification can optimize resources and improve clinical outcomes.
The researchers plan to validate the findings in diverse populations.
Benefits of Stratification
The division into subgroups makes it possible to better predict the course of the disease.
It guides decisions about medications, such as SGLT2 inhibitors or GLP-1 agonists, depending on the profile.
Patients at higher risk of specific complications receive targeted interventions.
This reduces the occurrence of long-term adverse events.
The approach represents an important step towards personalizing diabetes care.






