If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
1 The authors CD and ED contributed equally to this work.
Carola Deischinger
Footnotes
1 The authors CD and ED contributed equally to this work.
Affiliations
Department of Internal Medicine III, Clinical Division of Endocrinology and Metabolism, Gender Medicine Unit, Medical University of Vienna, Waehringer Guertel 18–20, 1090 Vienna, Austria
1 The authors CD and ED contributed equally to this work.
Elma Dervic
Footnotes
1 The authors CD and ED contributed equally to this work.
Affiliations
Section for Science of Complex Systems, CeMSIIS, Medical University of Vienna, Spitalgasse 23, Vienna, AustriaComplexity Science Hub Vienna, Josefstädter Straße 39, 1080 Vienna, Austria
Department of Internal Medicine I, Clinical Division of Hematology and Hemostaseology, Medical University of Vienna, Waehringer Guertel 18–20, 1090 Vienna, Austria
Section for Science of Complex Systems, CeMSIIS, Medical University of Vienna, Spitalgasse 23, Vienna, AustriaComplexity Science Hub Vienna, Josefstädter Straße 39, 1080 Vienna, Austria
Section for Science of Complex Systems, CeMSIIS, Medical University of Vienna, Spitalgasse 23, Vienna, AustriaComplexity Science Hub Vienna, Josefstädter Straße 39, 1080 Vienna, Austria
Department of Internal Medicine III, Clinical Division of Endocrinology and Metabolism, Gender Medicine Unit, Medical University of Vienna, Waehringer Guertel 18–20, 1090 Vienna, AustriaGender Institute, Gars am Kamp, Austria
Aim: Exploring sex-specific impact of diabetes mellitus (DM) on venous thromboembolism (VTE) risk in population-level study.
•
In DM patients, females are associated with higher relative risk increase in VTE compared to males.
•
The relative risk increase for VTE peaks between 50 and 59 years with an OR of 1.65.
Abstract
Aims
The risk for developing venous thromboembolism (VTE) is about equal in both sexes. Research suggests diabetes mellitus (DM) is a risk factor for pulmonary embolism and deep vein thrombosis, both forms of VTE. We aimed at investigating the sex-specific impact of DM on VTE risk.
Materials and methods
Medical claims data were analyzed in a retrospective, population-level cohort study in Austria between 1997 and 2014. 180,034 patients with DM were extracted and compared to 540,102 sex and age-matched controls without DM in terms of VTE risk and whether specific DM medications might modulate VTE risk.
Results
The risk to develop VTE was 1.4 times higher amongst patients with DM than controls (95% CI 1.36–1.43, p < 0.001). The association of DM with newly diagnosed VTE was significantly greater in females (OR = 1.52, 95% CI 1.46–1.58, p < 0.001) resulting in a relative risk increase of 1.17 (95% CI 1.11–1.23) across all age groups with a peak of 1.65 (95% CI 1.43–1.89) between 50 and 59 years. Dipeptidyl peptidase 4 inhibitors were associated with a higher risk for VTE amongst female DM patients (OR = 2.3, 95% CI 1.3–4.3, p = 0.0096).
Conclusion
Amongst DM patients, females appear to be associated with a higher relative risk increase in VTE than males, especially during perimenopause.
Type 2 diabetes mellitus is a chronic disease of hyperglycemia which affects over 8% of the world’s population and is associated with multiple micro- and macrovascular complications such as an increased risk to develop cardiovascular diseases (CVD) [
] but also the risk to develop venous thromboembolism (VTE), the disease entity covering pulmonary embolism (PE) and deep vein thrombosis (DVT), is increased in patients with diabetes mellitus (DM) [
Diabetes as Risk Factor for Incident Coronary Heart Disease in Women Compared with Men: A Systematic Review and Meta-Analysis of 64 Cohorts Including 858,507 Individuals and 28,203 Coronary Events.
Kautzky-Willer A, Harreiter J. Sex and Gender Differences in Therapy of Type 2 Diabetes. Diabetes Res Clin Pract. 2017;131:230–41. https://doi.org/10.1016/j.diabres.2017.07.012.
] but whether the risk of VTE concerning sex differs in patients with DM is currently unknown.
The exact reasoning behind the connection between DM and VTE has not yet been fully understood and might be caused by a state of hypercoagulability in DM [
]. Studies have demonstrated decreased levels of protein C and antithrombin in hyperglycemic conditions and higher levels of factor II, VII, VIII, fibrinogen, and soluble tissue factor in combination with shortened activated partial thromboplastin time (aPTT) and prothrombin time (PT) as well as the inhibition of endogenous fibrinolysis in patients with DM [
]. However, the coexistence of VTE and DM might be at least partly explained by shared risk factors including age, obesity, metabolic syndrome, cardiovascular diseases, and other comorbid conditions requiring hospitalization [
]. How sex might interact with the risk to develop VTE in patients with DM is unclear. Therefore, large-scale evidence on potential sex differences in the risk of VTE in DM is needed.
The primary aim of this nationwide study was to evaluate if females with DM are more likely to be diagnosed with VTE in comparison to their male peers. In addition, we provide epidemiological data on the risk of VTE in persons with DM compared to persons without DM.
2. Materials and methods
2.1 Study population
This retrospective cohort study is conducted based on population-wide medical claims data. The analyzed dataset covers approximately 45,000,000 hospital stays of about 9,000,000 individuals in Austria from 1997 until 2014, with primary and secondary diagnoses. All diagnoses are recorded in the form of level-3 ICD-10 codes (International Classification of Diseases). The primary diagnosis gives the reason for the hospital admission. Conditions that coexist at the time of admission are listed as secondary diagnoses. In this study, we treated primary and secondary diagnoses as equally relevant. The dataset contains the patient's ID, age at the time of the hospital stay, date of admission, date of discharge, primary diagnosis, secondary diagnoses, and the type of discharge (e.g. discharge from the hospital, discharge against medical advice, death, transfer to another hospital) for each in-hospital stay. Similar to previously published studies on the same dataset [
], we excluded patients with a hospital stay during the first 5 years of the observation period to assure the comparability of the health status of our study population.
The DM cohort consists of all patients with at least one diagnosis from E10 to E14 in 2003–2014 (primary or secondary diagnosis), for further explanation see Table 1. The age-matched controls are extracted in the ratio 1:3 resulting in 180 034 patients with DM (70 739 females, 109 295 males) and 540 102 controls (212 217 females, 327 885 males). Patients with VTE are determined as patients with at least one of three diagnoses I26, I80 or I82 reported as a primary or secondary diagnosis. Surgery is an important and frequent risk factor for VTE. Hence, we identified patients who had surgery before diagnosing with VTE (based on hospital departments where patients were encountered) and excluded them from the cohort of DM patients and the matched control group. Fig. 1 shows a flowchart illustrating the patient selection process.
Table 1Overview of ICD 10 codes and their description used in the study. ICD 10 = International Classification of Diseases, 10th edition. C00-D48: Neoplasms; E00-E90: Endocrine, nutritional and metabolic diseases; I00-I99: Diseases of the circulatory system; O00-O99: Pregnancy, childbirth and the puerperium; S00-T98: Injury, poisoning and certain other consequences of external causes.
In the age group of 20–79, we calculated odds ratios for being diagnosed with VTE in DM patients and the control group. The gender gap was calculated as a ratio between these odds ratios for females and males. The age group of over 80 years was analyzed separately as we observed decreasing diagnose frequencies in these patients, rendering the data less reliable.
To estimate the robustness and potential confounders of the results, we conducted a sensitivity analysis. In this analysis, we separately excluded all patients with diagnoses potentially provoking or preventing VTE (i.e., encounter for delivery (O80-O82), overweight and obesity (E66), neoplasms (C00-D48), heart failure (I50), and fractures (T02, T08, T10, T12, S02, S12, S22, S32, S42, S52, S62, S72, S82, S92, M80, M84)) to account for the major risk factors of VTE[
] and mortality as a competing risk. All used ICD codes are listed in Table 1.
To ensure the validity of our results, we analyzed the time order of the investigated diseases to identify which disease is diagnosed first. First, we calculated the time difference between the first DM diagnosis and the first VTE diagnosis for each patient over the period 2003–2014. Based on calculated time differences between DM and VTE, we grouped patients into four groups: (i) both diseases are diagnosed during the same hospital stay (time difference smaller than seven days) or the time difference is (ii) less than three months, (iii) more than three months but less than one year or, finally, (iv) more than one year. We proceeded to calculate the time order ratio (TOR) TOR(DM → VTE) = N(DM → VTE)/N(VTE → DM) and tested the null hypothesis that N(DM → VTE) = N(VTE → DM) to evaluate whether TOR(DM → VTE) is significantly different from 1, assuming that both counts stem from a binomial distribution with equal success probability. For more details see our previous publication using a similar method [
For a comparison of patients on different diabetes medications, we used an additional dataset on patients’ prescribed medication derived from medical claims data on the population with DM in Austria (N = 904,032). The dataset covers a six-year period (2012–2017) and contains detailed information on the prescribed medication (date, amount, etc. of specific ATC codes), data on hospital stays (including primary and secondary diagnoses in form of ICD-10 codes) as well as general information (patients’ date of birth, sex, place of residence, etc.). In a supplementary cross-sectional analysis, we analyzed patients’ prescribed drugs (metformin, insulin) and their effect on the risk of being diagnosed with VTE for the available comparable years 2012 to 2014. We excluded patients who were prescribed multiple different antidiabetic drugs to ensure homogeneous groups. Case and control cohorts were defined through patients’ primary and secondary hospital diagnoses (ICD-10 codes E10-E14 for DM and codes I26, I80, and I82 for VTE). Analogous to the main analysis we performed an age- and sex-specific 1:3 case-control matching. The case-cohort contains patients with DM and VTE (N = 844) while the control cohort consists of patients with DM without VTE (N = 2,385). We then calculated the odds ratios for being diagnosed with VTE in DM patients using the cohorts described above. We used Chi-squared statistics to test for sex-specific significance (p-values) between case and control cohorts.
To investigate the risk of developing VTE in relation to prior DM, we performed a Cox regression analysis with DM as the exposure variable and VTE as the outcome, accounting for potential confounders [
]. DM patients were matched 1:1 with controls on age and date of diagnosis of DM (cases) or any other diagnosis from range A00-N99, except E10-E14 (controls). We excluded all patients who were diagnosed with VTE before these dates. Confounders were sex, age, and diagnoses for encounter for delivery, overweight and obesity, neoplasms, heart failure, and fractures respectively. We computed Kaplan-Meier curves for male and female cases and controls and reported Hazard Ratios (HRs) with 95% confidence intervals (CIs) on DM and VTE risk.
3. Results
3.1 Study population
180,034 patients with DM (70739 females, 109,295 males) and 540,102 controls (212217 females, 327,885 males) were included in our study. Detailed baseline characteristics are visible in Table 2. Of all patients with DM, 2.09% were diagnosed with PE, 2.17% with thrombosis, phlebitis, and thrombophlebitis, 0.78% with other venous embolism and thrombosis; in the control group, 1.53% received the diagnosis PE, 1.57% thrombosis, phlebitis and thrombophlebitis and 0.54% other venous embolism and thrombosis. The DM group had on average more hospital days, stays, and more diagnoses than their healthy controls. Regarding VTE risk factors, patients with DM were more likely to be suffering from cancer, fractures, obesity and overweight, and heart failure. In comparison, female patients with DM had a higher average age but a fairly similar number of diagnoses as well as hospital stays than males.
Table 2Baseline characteristics table of all patients with DM, their respective controls and the prevalence (in %) of venous thromboembolism (VTE) as well as possible confounding factors for VTE amongst all patients between the age of 20 and 79 years in Austria from 2003 to 2014. Age was recorded at halfway point of the observed period: 31.12.2008.
Diabetes mellitus
Matched controls
Parameters and diagnoses
All
Female
Male
All
Female
Male
Number of patients
180,034
70,739
109,295
540,102
212,217
327,885
Age (in years, mean +/- SE)
63+/-12
64+/-13
62+/-12
62+/-12
64+/-13
62+/-12
Number of hospital stays (mean +/- SE)
6+/-8
6+/-8
6+/-8
4+/-6
4+/-6
4+/-6
Number of hospital days (in days, mean +/- SE)
47+/-61
48+/-62
46+/-60
23+/-41
23+/-40
23+/-41
Number of hospital diagnoses (mean +/- SE)
11+/-8
11+/-8
11+/-8
6+/-6
6+/-6
5+/-5
I26 – Pulmonary embolism (in %)
2.09
2.47
1.84
1.53
1.56
1.5
I80 – Thrombosis, phlebitis and thrombophlebitis (in %)
3.2 Sex differences in the impact of DM on VTE risk
As visible in Fig. 2 and Table 2, VTE risk amongst healthy controls is rather similar between sexes; female patients with DM, however, display a consistently higher relative VTE risk than their male counterparts. Simply put, whereas healthy females and males have a similar probability to develop VTE, there is a significant sex difference among patients with DM (see Fig. 2). Across all age groups (Fig. 2), female DM patients are at an increased risk to be diagnosed with VTE compared to females without DM (OR = 1.52, 95% CI 1.46–1.58, p < 0.001) whereas males with DM only have a 1.30 times higher risk when compared to males without this diagnosis (95% CI 1.26–1.35, p < 0.001), which results in a gender gap with a 1.17-fold (95% CI 1.11–1.23) increased risk for the diagnosis of VTE when compared to male patients with DM between the age of 20 and 79 years. As visible in table S1, the relative risk increase is mitigated by the fact that females in their 20 s and 30 s display a rather similar probability to be diagnosed with VTE than their male counterparts. A relative risk increase for VTE amongst female DM patients appears to emerge at age 40 and onwards with a peak at ages 50 to 59 years (gender gap (GG) = 1.65, 95% CI 1.43–1.89). In the age group 80 + years, the gender gap diminishes to 1.01 (95% CI 0.93–1.10).
Fig. 2Percentage of DM patients (solid lines) diagnosed with VTE compared to patients without DM (dotted lines) with VTE (in %): the gender gap of female (red lines) and male patients (blue lines) suffering from VTE is larger in the DM cohort (female patients with DM: OR = 1.52, 95% CI 1.46–1.58, p < 0.001; male patients with DM: OR = 1.30, 95% CI 1.26–1.35, p < 0.001) between the age of 20 and 80 years. The effect diminishes in the age group 80 + years (1.01 (95% CI 0.93–1.10). In the population without DM, males are equally likely to be diagnosed with VTE than females. DM = diabetes mellitus. VTE = venous thromboembolism. OR = odds ratio. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
VTE is an umbrella term for the ICD 10 codes “Pulmonary embolism”, “Thrombosis, phlebitis and thrombophlebitis” and “Other venous embolism and thrombosis”. When separately investigated, pulmonary embolism displayed the largest gender gap with female DM patients having a higher relative risk from age 50 onwards with a peak at age 50 to 59 years with a gender gap of 2.2 (95% CI 1.79–2.71) compared to males. For the disease group “Thrombosis, phlebitis and thrombophlebitis”, the gender gap is visible from age 40 until 69, whereas DM patients aged 70 + see a similar risk increase in both sexes. For the subgroup “Other venous embolism and thrombosis”, DM appears to be associated with a risk increase for females at age 50 to 59 years.
By conducting a sensitivity analysis, we aimed at correcting for potential confounders for the increased relative risk for VTE amongst females with DM (Fig. 3).[
] None of the investigated risk factors, encounter for delivery (O80-O82), overweight and obesity (E66), neoplasms (C00-D48), heart failure (I50), fractures (T02, T08, T10, T12, S02, S12, S22, S32, S42, S52, S62, S72, S82, S92, M80, M84) or mortality significantly affected our results.
Fig. 3Sensitivity test investigating confounding factors for the increased relative risk for venous thromboembolism (VTE) amongst female patients with DM: Encounter for delivery (O80-O82), overweight and obesity (E66), neoplasms (C00-D48), heart failure (I50), fractures (T02, T08, T10, T12, S02, S12, S22, S32, S42, S52, S62, S72, S82, S92, M80, M84) and mortality did not affect the results. The pink line represents the results unadjusted for confounding factors. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
To ensure that DM affected the probability of being diagnosed with VTE and not vice versa, we conducted a time directionality analysis. DM is typically diagnosed before VTE (TOR > 1) for all patient groups that were not diagnosed with VTE and DM during the same hospital stay (see Fig. 4). To investigate the risk of developing VTE in relation to prior DM, we performed a Cox regression analysis with DM as the exposure variable and VTE as the outcome, accounting for potential confounders. Patients with DM had a hazard ratio of 1.13 (95% CI 1.09–1.18, p < 0.001) for being diagnosed with VTE compared with their controls without DM, for details see the cox regression in figure S1 in the supplements. Fig. 5 shows Kaplan-Meier curves for patients with diabetes and matched controls and their respective probability of being diagnosed with VTE.
Fig. 4Time directionality for patients with DM – diabetes mellitus (E10-E14) and VTE – venous thromboembolism (I26, I80, I82), respectively: DM is typically diagnosed before VTE. The greater the time difference between the analyzed diagnoses, the stronger the dominance of patients first diagnosed with DM (TOR > 1). The size of the dots corresponds with the proportion of each group. The numbers in the small boxes represent the proportion of patients diagnosed at that time point. Significance levels of the TOR are indicated by asterisks (* p < 0.05, ** p < 0.01, *** p < 0.0001). TOR = time order ratio.
The analysis concerning diabetes medication resulted in no difference in the insulin or metformin prescription rate between the patients who developed VTE and those who did not (Insulin: males; OR = 0.87, p = 0.54; females: OR = 0.97, p = 0.95; Metformin: males; OR = 1.1, p = 0.49; females: OR = 0.86, p = 0.18; for details see Table 3). However, the proportion of patients taking DPP-4 inhibitors was higher in the group of female DM patients who developed VTE than those who did not (males: OR = 1.6, p = 0.28; females: OR = 2.3, p = 0.0096; for details see Table 3).
Table 3Overview of prescribed medication in both cases and matched control group for both sexes. The proportion of patients taking DPP-4 inhibitors was higher in the group of female DM patients who developed VTE. The proportion (%) of patients with a specific prescription is given for each sex in both groups. P-values (p-val) from Chi-squared statistics indicate significant differences between case and control groups. Odds ratios (OR) describe the ratio of the odds of being in the case group if exposed to some given medication to the odds of being in the control group if exposed to the same medication. SD = standard deviation.
The aim of this study was to analyze whether there is a potential gender gap in the higher relative VTE risk of patients with DM. According to our results, DM appears to be associated with a higher risk to develop VTE in females than in males, especially in the age group 40 years and older. The relative risk to develop VTE is increased 1.17-fold in females with DM compared to males with DM across all age groups, with a peak at the age of 50 to 59 (GG = 1.65). The reason for females with DM being more at risk to develop VTE than their male counterparts is yet unclear. Potential explanations include females being more likely to be diagnosed with overweight and obesity than males, which might impact VTE risk via the common pathway inflammation [
Colling ME, Tourdot BE, Kanthi Y. Inflammation, Infection and Venous Thromboembolism. Circ Res. 2021;2017–36. https://doi.org/10.1161/CIRCRESAHA.121.318225.
]. According to our analysis, the relative risk increase in VTE amongst females with DM compared to their male counterparts emerges only at the age of 40 years onwards resulting in a VTE gender gap in a disease group otherwise rather equally distributed amongst the sexes [
]. An explanation could be that females in their 20 s and 30 s have a slightly higher risk to develop VTE than males (see Fig. 2) due to reproductive factors such as childbirth and/or oral contraceptive intake [
]. At that age, DM might be less of a contributing factor compared to hormonal changes and exogenous hormone intake. Our lack of data on contraceptive pill intake is a limitation of this study and should be included in further studies conducted on this topic. Another potential explanatory model for a VTE gender gap only at age 40 onwards is that sex differences in DM disease severity and duration might directly impact the probability to develop VTE, similar to the relation of DM disease severity/duration to the risk to develop macrovascular complications [
Mannucci E, Monami M, Lamanna C, Gori F, Marchionni N. Prevention of Cardiovascular Disease through Glycemic Control in Type 2 Diabetes: A Meta-Analysis of Randomized Clinical Trials. Nutr Metab Cardiovasc Dis. 2009;19(9):604–12. https://doi.org/10.1016/j.numecd.2009.03.021.
]. However, in our analysis, male and female diabetes patients had a similar number of hospital diagnoses, number of hospital days and hospital stays indicating similar disease severity. To further investigate the potential effects of DM disease severity on VTE risk, we separated the patients into insulin-treated and those solely on oral antidiabetics (OAD). Initiation of insulin treatment in type 2 diabetes mellitus can be a sign of longer disease duration and increased severity [
]. Similar to what was expected, the insulin prescription rate was not different between the group who developed VTE and those who did not, thus, rendering the hypothesis of DM disease severity being a confounding factor improbable.
An analysis concerning oral antidiabetic medication intake was conducted to evaluate potential differences in VTE risk (see Table 3), for instance, if metformin intake might mitigate the risk of patients with DM developing VTE. A previously published study showed a potential protective effect of metformin on the risk of developing deep vein thrombosis [
]. We were not able to corroborate these results as both cohorts developed VTE at similar rates. However, dipeptidyl peptidase 4 inhibitors were associated with a higher risk for VTE amongst female DM patients in our cohort, which is in line with previously published literature. An analysis of a WHO database concluded that DPP-4 inhibitors display an increased risk for VTE [
]. We observed no differences in VTE rates in patients taking sulfonylureas or glitazones.
To investigate other potential confounders of the VTE gender gap amongst DM patients, a sensitivity test on suspected covariates for VTE was conducted. Neither the diagnosis encounter for delivery (O80-O82) nor overweight and obesity (E66), neoplasms (C00-D48), heart failure (I50), fractures (T02, T08, T10, T12, S02, S12, S22, S32, S42, S52, S62, S72, S82, S92, M80, M84) nor mortality as a competing risk had a significant impact on our results.
A clear strength of our investigation is that we establish consistent and robust time ordering across different analytic approaches based on different technical assumptions emphasizing the robustness of our findings. However, this study also has certain limitations solely because of the type of investigation that was conducted, namely a hospital-record-based analysis of the entire Austrian population. Therefore, we had no access to primary care facilities outside of Austrian hospitals as patients had to have been registered in a hospital at least once to be included in the analysis. As VTE is often diagnosed and treated in an outpatient setting, further investigations on the topic should most certainly include data from a primary care setting. Furthermore, we were not able to correct for some risk factors of VTE such as length of hospital stay as most patients had a stay of over 3 days, or other factors such as oral contraceptive intake, smoking, pregnancy, and bed rest of over 5 days outside of a hospital setting as they were not recorded in this dataset. Furthermore, we estimate that the diagnosis “Obesity and overweight” was underdiagnosed by doctors in our database as the estimated percentage of overweight persons in Austria is higher. Moreover, the diagnosis “diabetes mellitus” included both type 1 and type 2 diabetes mellitus as a reliable distinction was not possible in this database. As we did not have access to laboratory results, the degree of hyperglycemia (e.g. HbA1c) could only be taken into account by analyzing the use of insulins. The age group of over 80 years was analyzed separately as we observe decreasing diagnose frequencies in these patients. This might be due to a survival bias as people who age healthily are more likely to reach such a high age. The gender gap in this age group diminished to 1.01 which can potentially be explained by gender differences in the survival bias. Furthermore, it is more likely that we observe the actual onset of VTE or DM at younger ages than 80 years and therefore observe a stronger gender gap at younger age. We also could not analyze the potential effects of GLP-1 (glucagon-like-peptide-1) agonists or SGLT-2 (sodium-glucose-linked transporter-2) inhibitors as they have been introduced after the year 2016 and our hospital dataset only includes data from 2003 to 2014.
5. Conclusion
This is the first analysis showing DM might be associated with VTE to a greater degree in females than males. To investigate this topic in detail, prospective analyses will be necessary, especially including investigations as to the pathogenesis of this relative risk increase. This would be an important step in the prevention of VTE in patients with DM, especially if they are female. At this point, one might conclude that females with DM should be more carefully monitored with respect to VTE development, with special attention to perimenopause.
Ethics approval and data sharing statement
We made secondary use of a research database of medical claims records that are safeguarded and maintained by the Austrian National Institute of Public Health (GÖG) and where it has been ensured that no individual is identifiable. This is a consolidated research database that is only available to selected partners under a strict data protection policy. The use of the data takes place in agreement and cooperation with GÖG. The data contains no names of individual patients or healthcare providers. All unique identifiers including postal codes and date of birth have been removed. All team members working with the data have signed a confidentiality agreement and declaration of commitment ensuring that research will be undertaken in accordance with the applicable data protection regulations. The Federal Law on Documentation in the Health Care System in Austria provides the legal basis for written informed consent not being required for this study: It allows the documentation of health-related data in the intra- and extramural outpatient and inpatient care sectors, as well as for the processing of patients' and service providers' data in pseudonymized form for certain purposes including (long-term) monitoring of epidemiological developments relevant to health policy as well as the implementation and further development of integrated health structure planning and health services research.
Funding
Funding was received from the WWTF – Vienna Science and Technology Fund (MA16-045).
Contributorship
C.D. and E.D. wrote the manuscript, researched data and compiled graphs and tables. S.N., M.K. and P.K. contributed to the design, methods and reviewed/edited the manuscript. A.K-W. contributed to the discussion and reviewed/edited the manuscript. A.K-W. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. All authors approved the final version.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Appendix A. Supplementary data
The following are the Supplementary data to this article:
Diabetes as Risk Factor for Incident Coronary Heart Disease in Women Compared with Men: A Systematic Review and Meta-Analysis of 64 Cohorts Including 858,507 Individuals and 28,203 Coronary Events.
Kautzky-Willer A, Harreiter J. Sex and Gender Differences in Therapy of Type 2 Diabetes. Diabetes Res Clin Pract. 2017;131:230–41. https://doi.org/10.1016/j.diabres.2017.07.012.
Colling ME, Tourdot BE, Kanthi Y. Inflammation, Infection and Venous Thromboembolism. Circ Res. 2021;2017–36. https://doi.org/10.1161/CIRCRESAHA.121.318225.
Mannucci E, Monami M, Lamanna C, Gori F, Marchionni N. Prevention of Cardiovascular Disease through Glycemic Control in Type 2 Diabetes: A Meta-Analysis of Randomized Clinical Trials. Nutr Metab Cardiovasc Dis. 2009;19(9):604–12. https://doi.org/10.1016/j.numecd.2009.03.021.
01004-X&id=gr1.jpg){kind=link}
01004-X&id=gr2.jpg){kind=link}
01004-X&id=gr3.jpg){kind=link}
01004-X&id=gr4.jpg){kind=link}
01004-X&id=gr5.jpg){kind=link}