The global estimation of diabetes among adults aged 20-79 years is projected to increase from 536.6 million people in 2021 to 783.2 million people in 2045 with a predicted expenditure of USD 1,054 billion, which represents an increase of 9.1% compared to that of 2021 (1). Additionally, diabetes-related mortality in 2021 represented 12.2% of global deaths from all causes in people aged 20-79 years. Diabetes-associated deaths among women are reported to be much higher than in men, especially after the age of 60-70 years (2).

Prospective studies and meta-analysis have shown that women with diabetes have a higher risk of cardiovascular disease (CVD) than men in comparison with their non-diabetic counterparts (3–7), and have greater difficulty in achieving the therapeutic targets of cardiovascular risk factor (CVRF) control, especially lipid control (8–11). An Italian study attempted to establish the precise time at which excess risk begins in women. Their findings revealed that excess risk of acute myocardial infarction and major cardiovascular events started earlier (46 years), and lasted over the age of 85 years, while ‘risk-windows’ started later and had a shorter duration for congestive heart failure (56-65 years) and ischemic stroke (66-75 years) (12).

Most studies have been conducted with prevalent cases of diabetes and have not described the sex differences at the onset of type 2 diabetes mellitus (T2DM). Available data is scarce regarding sex differences in prediabetes, the prelude to diabetes, and in people with newly-diagnosed T2DM. Regarding CVRFs, conversion from prediabetes to diabetes has been shown to be associated with an increased body mass index (BMI), fasting insulin, triglycerides (TGs) and blood pressure (BP), and lower high-density lipoprotein cholesterol (HDL-c), differences that were greater in women than men (13, 14). The Coronary Artery Risk Development in Young Adults (CARDIA) study, a longitudinal observational cohort study done in the US metropolitan communities, found that CVRF worsened more rapidly after the development of T2DM in women than in men, but they did not find differences between women and men before diabetes (15). However, it should be noted that the proportion of those in the cohort who developed diabetes was small. Overall, to our knowledge, no studies have evaluated the differences between men and women in a large cohort at the onset of T2DM. For this reason, this study sought to describe the clinical characteristics, the degree of glycaemic control and cardiovascular risk factor control at the onset on T2DM, together with any changes 1 year post-diagnosis, in a population-based cohort of newly-diagnosed subjects with T2DM in Catalonia (Spain).

This study describes the clinical characteristics, degree of control of CVRFs and their change 1 year after diagnosis according to sex in a large population-based cohort of 13,629 subjects with newly-diagnosed T2DM in Catalonia (Spain). To our knowledge, this is the only large population-based study addressing this topic in southern Europe. Only a few studies have included newly diagnosed T2DM as the primary study population (24–26), and far fewer have done so differentiating by sex (27–29) particularly with the main objective of describing men and women at onset of T2DM (15, 30).

Of the total number of subjects with newly-diagnosed T2DM, the proportion of men was higher than that of women. Women were older than men, evenly distributed in the age ranges from 50 years onwards. The distribution among men was higher in the age range 50-65 years. In agreement with our results, the incidence of T2DM has been reported to be higher in men than in women and in the oldest age groups according to studies in Europe (31, 32) and Spain (33). In secondary prevention subjects, the age at the onset of diabetes was higher, which is evident because CVD develops at older ages (34).

Men were more frequently smokers and former smokers than women, especially in secondary prevention. The role of smoking in CV morbidity and mortality is widely known (35–37) but it has also been associated with an increased risk of T2DM. The pooled relative risk (RR) of T2DM has been reported to be 1.37 for current smoking and 1.14 for former smoking (38).

Several sex differences in baseline characteristics were observed. We found a higher BMI in women at time of diagnosis. It has been estimated that women have a BMI 1.8 kg/m² higher than men at T2DM diagnosis despite similar levels of HbA1c (39). This variance has been primarily linked to sex-specific physiological differences in fat distribution. Notably, women exhibit a distinctive fat distribution characterized by a higher proportion of subcutaneous fat mass and comparatively lower levels of liver and visceral fat content. This favorable pattern changes after post-menopause, when the fat distribution in women transition from a gynoid pattern to an android pattern accompanied by an increase in cardiometabolic risk (40). In parallel, women also tend to display heightened glucose sensitivity in comparison to men (41). A possible consequence of these sex-specific metabolic nuances is that women require a greater weight gain and adiposity accumulation to meet the diagnostic criteria for T2DM. This phenomenon contributes to an extended duration of the prediabetes state in women, where an elevated presence of CV risk factors is evident (42).

In accordance, we observed a poorer lipid profile in women than in men, especially in total cholesterol and LDL-c but not in HDL and TGs. Several studies report similar results in baseline characteristics of prediabetes or newly-diagnosed T2DM subjects (28, 43, 44), and emphasize the more adverse changes in cardiometabolic risk factors in women as a continuous process in the transition from normoglycemia to diabetes (45, 46). Women could potentially face prolonged exposure to hyperglycemia or an inadequate state of glucose levels, leading to heightened vascular damage and increases in CVRFs (47–49). Otherwise, women had lower mean glucose and HbA1c levels than men which may suggest a better insulin sensitivity pattern in women especially before developing TDM2 (50, 51). This is consistent with the baseline characteristics in different studies with prediabetes or newly-diagnosed T2DM subjects (27, 28, 44, 52, 53). In line with this, the DECODE Study group found that impaired fasting glucose was more common in men whereas impaired glucose tolerance was higher in women. As a consequence, diabetes may go undiagnosed, especially in females, as the fasting glucose determination alone is the standard method for diagnosis (54).

Regarding the control of CVRF and its management, differences were observed between sexes, most notably in primary prevention subjects. The achievement of HbA1c < 7% was higher in women than in men, both at baseline and after 1 year, but the differences between the sexes were higher at baseline than thereafter. A higher proportion of women did not receive antidiabetic treatment at baseline, although after 1 year, the prescription of the antidiabetic drugs was similar between groups. Similarly, BP < 140/90 mmHg was better achieved in women than in men and there were no major differences in anti-hypertensive treatment with the exception of diuretics, which were mostly prescribed in women. The use of thiazide diuretics has been associated with an increased risk of developing T2DM as opposed to RAAS blocker use that seems to reduce the risk of T2DM (55).

By contrast, the attainment of LDL-c < 100 mg/dL was worse in women at baseline and even worse 1 year after diagnosis. Statins were more frequently prescribed in women, but the frequency of statin prescription in men 1 year after diagnosis was considerably higher. It should be noted that statins have been reported to increase the risk of T2DM. Factors associated with this effect of statins are the type of statin, the dose and the potency (56).

It is noteworthy that when the CVRF achievement is better in women than men at baseline (e.g. as seen for HbA1c and BP targets), the difference between the sexes narrowed considerably one year after diagnosis, accompanied by a greater intensification of treatment in men. In contrast, the achievement of LDL-c target levels was better for men at baseline and even more so 1 year post-diagnosis i.e. the difference did not narrow over time for women. This fact can be seen in more detail in the percentage change in CVRF targets and also in the percentage change of clinical characteristics where the improvement over time was higher in men than in women ( Supplementary Table 3 and 4 ). The reasons why women are not seeing the same improvements as men are not known, however this issue suggests that more aggressive treatment would be useful in women especially in primary prevention (57, 58).

The present study has some limitations. Ethnicity was not known, and we could not differentiate the sample according to this variable. Some residual confounders were not available in our study, such as physical activity, nutritional status, sex hormones or the use of hormone replacement therapy, which appear to play a protective role in the onset of diabetes (59); all these factors together with the socioeconomic status may have yielded somewhat different results. Observational studies and cross-sectional design do not allow the establishment of causal relationships between the variables. The retrospective design of our study introduces the possibility of selection bias, as the study draws on pre-existing records rather than a prospectively designed protocol. Also, there were no data on the doses of the prescribed drugs, on contraindications, or on treatment adherence, which may influence differences in the disease management. Moreover, the SIDIAP database may have limitations related to the accuracy and completeness of the recorded data. Lastly, the dates of the drug prescription and the dates of the blood tests were unknown, and therefore it was not possible to know which came first; however, this issue affected both groups. For future research, a controlled study design, including prospective data collection, could overcome many of the above-mentioned limitations.

In conclusion, this study shows that there are differences between men and women in CV risk factors and their control, not only long after their diagnosis, but also at the onset of the disease, especially in primary prevention. These differences are especially evident in the lipid profile and the achievement of its targets. It is also important to note that improvements in the control of CV risk factors over time (1-year post-diagnosis) were more evident in men than in women, suggesting that women might benefit from a more aggressive treatment approach in the first year after the onset of the disease.

Restrictions apply to the availability of some or all data generated or analyzed during this study because they were used under license. Requests to access these datasets should be directed to DM, didacmauricio@gmail.com.

The studies involving humans were approved by the Ethics Committee of the Primary Healthcare University Research Institute (IDIAP) Jordi Gol (P22-207). The studies were conducted in accordance with the local legislation and institutional requirements. Written informed consent for participation was not required from the participants or the participants’ legal guardians/next of kin in accordance with the national legislation and institutional requirements.

AR-M: Data curation, Formal Analysis, Investigation, Methodology, Project administration, Validation, Visualization, Writing – original draft, Writing – review & editing, Conceptualization. JF-N: Conceptualization, Supervision, Validation, Visualization, Writing – review & editing. JR: Formal Analysis, Methodology, Writing – review & editing. QM-C: Formal Analysis, Writing – review & editing. MB: Writing – review & editing, Data curation, Validation, Resources. BV: Writing – review & editing, Data curation, Validation, Resources. DM: Conceptualization, Methodology, Supervision, Validation, Visualization, Writing – review & editing.