European Journal of Cardiovascular Medicine

Cardiovascular disease (CVD) is the leading cause of illness and death worldwide, with a particularly fast-growing burden in low- and middle-income countries like India [1].

With increasing exposure to lifestyle-related risk factors, Indian adults are becoming more prone to cardiovascular diseases. Indian healthcare professionals are particularly vulnerable due to their strenuous job profile, but have been less studied compared to other vulnerable populations. QRISK3 is a validated tool to estimate 10-year cardiovascular risk [2], it is a widely used algorithm that estimates the 10-year risk of cardiovascular events by incorporating demographic, clinical, and lifestyle factors [3]. Understanding the cardiovascular risk profile of medical professionals is essential, not only for their own health but also for the credibility and sustainability of health care in general.

This study aims to describe the distribution of QRISK3 scores and identify independent predictors of high cardiovascular risk in a large cohort of Indian healthcare practitioners using contemporary data. The reason for selecting QRISK3 as the primary tool is because it incorporates comorbidities (e.g., CKD, rheumatoid arthritis, migraine, severe mental illness/psychotropics) and an ethnicity term relevant to South Asian populations and has recent external validation. However, to understand how QRISK3 estimates relate to other widely used tools, we additionally performed head-to-head agreement and reclassification analyses comparing QRISK3 with Atherosclerotic Cardiovascular Disease (ASCVD) risk estimator [4] , Framingham (General CVD) [5], and SCORE2 [6].

Study Design and Population

We conducted a cross-sectional analysis of data collected from 9,804 Indian healthcare practitioners between January and March 2025. Participants were physicians and specialists from various regions across India. The dataset included demographic characteristics, clinical history, physical measurements, and laboratory values relevant for cardiovascular risk estimation.

Data were collected directly from participating physicians after obtaining their verbal agreement. No laboratory investigations were initiated. Laboratory values included in the analysis were based on existing clinical records of the participating physicians and were available at the time of data collection.

The study protocol (Protocol No. ERIS/QRS/24/004) was reviewed and approved by the Medilink Ethics Committee, located at Medilink Hospital & Research Centre, Ahmedabad, India.

QRISK3 Score and Risk Stratification

The 10-year cardiovascular risk score was calculated using the QRISK3 algorithm [3]. Based on standard thresholds, participants were categorized into three groups: low risk (<10%), intermediate risk (10–20%), and high risk (>20%). Self-reported categorical variables such as smoking status, diabetes, and medication use were included, alongside measured continuous parameters like systolic blood pressure, BMI, lipid profiles, and HbA1c. We also calculated comparative 10-year risk scores using the available parameters ASCVD (PCE), Framingham General CVD and SCORE2 (diabetes only) to better interpret QRISK3.

Statistical analysis

Baseline characteristics were compared across QRISK3 categories and by sex using chi-square tests for categorical variables and one-way ANOVA (unadjusted) for continuous variables.

A multivariable logistic regression model was developed to identify independent predictors of a high QRISK3 score (>20%). To maintain model simplicity and prevent overfitting, stepwise selection based on the Akaike Information Criterion (AIC) was used. This method retains variables that meaningfully contribute to the outcome and excludes those with limited predictive value. Binary variables with fewer than 50 positive cases were excluded to improve model stability and avoid unreliable estimates.

Agreement between QRISK3 and other cardiovascular risk scores was assessed using Bland–Altman analysis (bias and 95% limits of agreement) and weighted κ statistics for categorical agreement across risk bands (<10%, 10–20%, and >20%). Reclassification was evaluated by cross-tabulating QRISK3 risk bands against each comparator’s bands and summarizing the number and percentage of participants who moved to a higher, lower, or unchanged risk category, emphasizing transitions across the 10% and 20% thresholds.

Age- and sex-adjusted mean 10-year cardiovascular risks were also estimated across four geographic zones (North, South, East, and West) for each scoring system (ASCVD PCE, Framingham CVD, QRISK3, and SCORE2). Adjusted (least-squares) means were derived from linear models controlling for age and sex. Pairwise comparisons between zones were performed using Tukey’s Honest Significant Difference (HSD) test to identify significant differences between regions.

All analyses were performed using R software (version 4.3.1). A two-sided p value < 0.05 was considered statistically significant.

Among 9,804 Indian healthcare practitioners, the mean 10-year QRISK3 score was 11.0 ± 10.9%, indicating moderate average cardiovascular risk (Table 1). The cohort was predominantly male (91.6%), with a mean age of 52.3 ± 10.5 years and BMI of 26.7 ± 3.3 kg/m². Type 2 diabetes was present in 17.8% of participants.

Women were underrepresented in the present population (n = 826, 8.4%) and demonstrated significantly lower cardiovascular risk scores than men (4.0% vs. 11.6%; p < 0.001). This may be primarily attributed to their younger age as compared to men (46.4 ± 8.6 years vs. 52.8 ± 10.5 years, p< 0.001) and healthier lifestyle behaviors. For instance, the proportion of non-smokers was significantly higher among women (99.3%) than among men (84.4%, p< 0.001). Current or past smoking was reported by 14.4% of participants. Antihypertensive medications and statins were used by 18.6% and 18.1%, respectively.

Table 1 further shows the distribution of participants across QRISK3 categories: low risk (<10%), intermediate risk (10–20%), and high risk (>20%). The proportion of males increased from 88% in the low-risk group to 99% in the high-risk group, and mean age rose from 46.4 to 66.8 years (p < 0.001). Type 2 diabetes, chronic kidney disease, atrial fibrillation, smoking, elevated blood pressure, and adverse lipid profiles were all more common in higher-risk groups. Use of antihypertensive medication and statins also increased with risk level, reaching 49.8% and 49.1%, respectively, in the high-risk group.

In multivariable logistic regression (Figure 1), chronic kidney disease was the strongest independent predictor of high cardiovascular risk (OR: 90.4, 95% CI: 38.1–223.6), followed by atrial fibrillation (OR: 18.9, 95% CI: 7.6–48.3), rheumatoid arthritis (OR: 7.2, 95% CI: 4.2–12.4), and migraine (OR: 2.2, 95% CI: 1.4–3.3). Smoking was also a strong and graded predictor, with risk strongly correlated to the number of cigarettes smoked: heavy smokers had the highest odds (OR: 27.9, 95% CI: 5.2–129.5), followed by moderate smokers (OR: 22.6, 95% CI: 8.9–57.0), reflecting a clear dose–response relationship.

Agreement between QRISK3 and other risk scores

Bland–Altman analyses (Figure 2) demonstrated that QRISK3 produced higher 10-year cardiovascular risk estimates than ASCVD (PCE), with a mean bias of +4.2% (limits of agreement −13.6% to +21.8%). In contrast, agreement with Framingham CVD was close, showing a mean bias of −0.7% (limits −10.2% to +9.8%). As shown in Figure 3, QRISK3 was strongly correlated with both ASCVD (R² = 0.86) and Framingham (R² = 0.81).

When participants were categorized into risk bands (<10%, 10–20%, >20%), QRISK3 up-classified 23% of individuals compared with ASCVD and 5% compared with Framingham, while down-classification occurred in fewer than 1% and 7% of cases, respectively (Table 2). Weighted kappa coefficients indicated substantial agreement between QRISK3 and ASCVD (κ = 0.77) and almost perfect agreement with Framingham (κ = 0.89). Agreement with SCORE2 was lower (κ = 0.38), suggesting more frequent reclassification across categories (Table 3).

Overall, these results indicate that QRISK3 provides higher estimated cardiovascular risk than ASCVD and SCORE2, while showing close concordance with the Framingham model. This suggests that QRISK3 may identify a greater proportion of individuals at elevated risk, potentially improving preventive risk stratification.

Zone differences in 10-year cardiovascular risk

Table 4 shows that age- and sex-adjusted 10-year cardiovascular risk varied across geographic zones among 9,804 participants, comprising 651 from the East, 2,557 from the North, 3,045 from the South, and 3,551 from the West. For QRISK3, the highest adjusted mean risk was observed in the East (10.8%, 95% CI 10.3–11.3), followed by the South (10.2%, 9.8–10.6) and North (10.0%, 9.5–10.4), with the West showing the lowest risk (9.4%, 9.0–9.8). A similar East-highest and West-lowest pattern was seen for the ASCVD (PCE), Framingham CVD, and SCORE2 models.

In the compact-letter display, zones sharing the same letter do not differ significantly (p ≥ 0.05), whereas those with different letters differ significantly (p < 0.05). For example, in the QRISK3 model, the East (letter 3) had a significantly higher adjusted mean than the West (letter 1), while North and South (both 2) did not differ significantly.

Figure 1. Adjusted odds ratios and 95% confidence intervals for high QRISK3 scores (>20%) among Indian healthcare professionals.

CKD indicates chronic kidney disease; Total/HDL Ratio, total cholesterol to high-density lipoprotein cholesterol ratio; LDL, low-density lipoprotein cholesterol; BMI, body mass index; SBP, systolic blood pressure.

Figure 2. Bland–Altman agreement between QRISK3 and other 10-year risk scores.

Panels show QRISK3 versus ASCVD (PCE), Framingham CVD, and SCORE2. The y-axis is the difference (QRISK3 comparator, %); the x-axis is the mean of the two scores (%). Dotted lines indicate mean bias and 95% limits of agreement (±1.96 SD). Positive values indicate higher risk estimated by QRISK3. Points represent individual participants; fitted lines are not shown to avoid implying causality.

Figure 3. Linear regression agreement between QRISK3 and comparator scores.

Scatterplots show QRISK3 (y-axis) versus ASCVD (PCE), Framingham CVD, SCORE2, and UKPDS (10-year CHD risk; diabetics only) on the x-axis. Each panel includes the least-squares regression line with 95% CI and the line of identity (y=x). Panel labels report the fitted equation and R². Axes are in percent risk.

Table 1. Distribution of cardiovascular risk categories (QRISK3) among Indian healthcare professionals.

The table shows participant characteristics across low (<10%), intermediate (10–20%), and high (>20%) QRISK3 risk categories. CKD indicates chronic kidney disease; AF, atrial fibrillation; RA, rheumatoid arthritis; T2DM, type 2 diabetes mellitus; Statin, use of statin medication; BP Meds, use of antihypertensive drugs; LDL, low-density lipoprotein cholesterol; HDL, high-density lipoprotein cholesterol; BMI, body mass index; SBP, systolic blood pressure; and HbA1c, glycated hemoglobin.

Table 2. Reclassification across 10-year risk bands (<10%, 10–20%, >20%) comparing QRISK3 with ASCVD (PCE), Framingham CVD, and SCORE2.

Table 3. Weighted κ (quadratic) across risk bands (<10%, 10–20%, >20%) comparing each calculator with QRISK3.

Table 4. Age- and sex-adjusted mean 10-year cardiovascular risk by geographic zone across four risk scores (ASCVD [PCE], Framingham CVD, QRISK3, SCORE2).

For each score, zone contrasts are presented as mean differences with 95% CI and Tukey-adjusted p-values. Values are adjusted (least-squares) means with 95% CIs. Compact-letter display (Tukey) indicates pairwise comparisons: zones sharing a letter are not significantly different (p≥0.05). Zone sample sizes: East (n=651), North (n=2,557), South (n=3,045), West (n=3,551). Positive differences indicate higher adjusted mean risk in the first-named zone.

This cross-sectional analysis of 9,804 Indian healthcare professionals revealed a substantial cardiovascular risk burden in this medically trained population. The mean 10-year QRISK3 score was 11.0%, and 15.7% of participants were classified in the high-risk category (>20%). Men had significantly higher mean risk than women (11.6% vs. 4.0%), consistent with sex differences observed in other populations.

Chronic kidney disease (CKD), atrial fibrillation, and rheumatoid arthritis were the strongest independent predictors of high QRISK3 scores: CKD (OR = 90.4; prevalence 1.2%), atrial fibrillation (OR = 18.9; 1.1%), and rheumatoid arthritis (OR = 7.2; 2.3%). These findings highlight the role of systemic and inflammatory conditions, particularly CKD, in elevating cardiovascular risk beyond traditional factors [7] and underscore the need for targeted prevention in this group [8].

Smoking status also played a critical role in cardiovascular risk. Although the overall proportion of heavy (0.3%) and moderate smokers (0.8%) was low, these groups had markedly elevated risk, with ORs of 27.9 and 22.6, respectively. This strong and graded association indicates that cardiovascular risk increases sharply with the number of cigarettes smoked. Light and ex-smokers also showed elevated risk, though to a lesser extent.

In our study, 7.4% of healthcare professionals reported smoking. This seems lower than the 10.7% smoking rate in the general Indian population but still concerning [9]. Notably, smoking was rare among women, which may partially explain their lower cardiovascular risk compared to men.

One might expect healthcare workers to be more aware of the health risks. However, long working hours, high stress, and heavy workloads may drive some of them to smoke. These factors, worsened by lack of physical activity and irregular routines, exacerbate the overall cardiovascular risk scores [10].

A previous study among over 5,000 newly diagnosed type 2 diabetes patients [11] reported a higher average Qrisk (15.3% overall; 12.2% in women, 17.1% in men) than in our study. However, only 7.5% of those patients were using statins. In contrast, 18.1% of healthcare professionals in our study were on statins. This may reflect better awareness and more proactive risk management among healthcare providers.

Difference Qrisk3 and other risk scores

Comparison across risk scores showed that QRISK3 produced higher 10-year cardiovascular risk estimates than both ASCVD (Pooled Cohort Equations) and Framingham, leading to reclassification of a substantial proportion of participants into higher risk categories. This difference likely reflects the broader range of predictors included in QRISK3, such as chronic kidney disease, atrial fibrillation, rheumatoid arthritis, severe mental illness, steroid use, erectile dysfunction, and systolic blood pressure variability, along with its ethnicity adjustment for South Asian populations.

The differences between the scores may also be explained by how each model was developed and calibrated. QRISK3 was developed in the United Kingdom, ASCVD in the United States, and SCORE2 in European cohorts, each reflecting region-specific population characteristics and baseline event rates. As overall risk increases, the divergence between these models becomes more pronounced.

Regional differences

The observed geographic pattern, i.e. the highest cardiovascular risk in the East and lowest in the West, aligns with prior data from India showing regional variation in cardiovascular burden. The India State-Level Disease Burden Initiative showed substantial inter-state differences in CVD disability-adjusted life years and key risk factors such as high blood pressure, diabetes, obesity, and tobacco use, with several eastern and north-eastern states showing the highest burden [12]. National surveys similarly reveal higher tobacco consumption in these regions [13]. Studies also documented higher stroke and overall CVD mortality in eastern and southern India [14,12], and more recent analyses indicate marked state-level variation in hypertension prevalence (e.g., Sikkim among the highest, Rajasthan among the lowest) [15]. Together, these findings provide an explanation for the modest East > South ≈ North > West differences observed in our cohort.

Strengths and limitations

This study should be interpreted with its limitations in mind; the absence of a non-healthcare professional comparison group limits the ability to determine whether healthcare workers have higher or lower cardiovascular risk than the general population. Additionally, because of the cross-sectional design, we can identify associations but cannot establish causal relationships. Some predictors, such as chronic kidney disease and atrial fibrillation, had low prevalence, leading to wide confidence intervals despite large effect sizes. Finally, the cohort was predominantly male (over 90%), which may overstate overall average risk; inclusion of more women could yield lower mean scores.

This study is, to our knowledge, the largest nationwide evaluation of cardiovascular risk among Indian healthcare professionals. The sample size and broad regional coverage enhance its representativeness. Use of QRISK3, a model incorporating ethnicity and comorbidity, provides contextually relevant risk estimation for Indian populations. An additional strength is the inclusion of direct comparisons between QRISK3 and other established risk models (ASCVD, Framingham, and SCORE2), allowing for a better understanding of their relative performance and applicability in the Indian context.

This study demonstrates that a substantial proportion of Indian healthcare professionals, particularly older males, carry a high cardiovascular risk, with 15.7% classified in the high-risk QRISK3 category. High-risk individuals were characterized by a composition of traditional and emerging risk factors, with chronic kidney disease, atrial fibrillation, rheumatoid arthritis, and smoking standing out as the most impactful predictors.

It is a common but dangerous misconception that healthcare providers inherently manage their own health effectively. This study highlights the need to challenge that assumption. The medical community must recognise that healthcare professionals, like any other high-risk group, require regular risk assessment, structured health monitoring, and support for preventive actions. Institutions and healthcare systems should promote a culture where personal health is prioritized, not only for the benefit of the individual, but for the entire healthcare system.

Conflict of interest: None

Funding: None.

1. GBD 2019 Risk Factors Collaborators. Global burden of 87 risk factors in 204 countries and territories, 1990-2019: A systematic analysis for the Global Burden of Disease Study 2019. Lancet. 2020;396(10258):1223-1249. doi:1016/S0140-6736(20)30752-2
2. Hippisley-Cox J, Coupland C, Brindle P. Development and validation of QRISK3 risk prediction algorithms to estimate future risk of cardiovascular disease: Prospective cohort study. BMJ. 2017;357:j2099. doi:1136/bmj.j2099
3. Parsons RE, Liu X, Collister JA, Clifton DA, Cairns BJ, Clifton L. Independent external validation of the QRISK3 cardiovascular disease risk prediction model using UK Biobank. Heart. Published online July 2023:heartjnl-2022-321231. doi:1136/heartjnl-2022-321231
4. Goff DC, Lloyd-Jones DM, Bennett G, et al. 2013 ACC/AHA Guideline on the Assessment of Cardiovascular Risk. Circulation. 2014;129(25_suppl_2):S49-S73. doi:1161/01.cir.0000437741.48606.98
5. D’Agostino RB, Vasan RS, Pencina MJ, et al. General Cardiovascular Risk Profile for Use in Primary Care: The Framingham Heart Study. Circulation. 2008;117(6):743-753. doi:1161/CIRCULATIONAHA.107.699579
6. SCORE2 working group and ESC Cardiovascular risk collaboration. SCORE2 risk prediction algorithms: New models to estimate 10-year risk of cardiovascular disease in Europe. Eur Heart J. 2021;42(25):2439-2454. doi:1093/eurheartj/ehab309
7. Jankowski J, Floege J, Fliser D, Böhm M, Marx N. Cardiovascular Disease in Chronic Kidney Disease: Pathophysiological Insights and Therapeutic Options. Circulation. 2021;143(11):1157-1172. doi:1161/CIRCULATIONAHA.120.050686
8. Zannad F, McGuire DK, Ortiz A. Treatment strategies to reduce cardiovascular risk in persons with chronic kidney disease and Type 2 diabetes. J Intern Med. 2025;297(5):460-478. doi:1111/joim.20050
9. Health and Family Welfare Government of India M of. Global Adult Tobacco Survey: GATS-2 India 2016–17. Published online 2017. https://www.mohfw.gov.in/sites/default/files/GATS-2%20FactSheet.pdf
10. Naik MR, Jacob RM, Reddy SP. Prediction of Cardiovascular Risk among Healthcare Professionals using Atherosclerotic Cardiovascular Disease Risk Score in a Tertiary Care Hospital in Aurangabad, India. Annals of Clinical Cardiology. 2021;3(2):69. doi:4103/ACCJ.ACCJ_18_21
11. Unnikrishnan AG, Sahay RK, Phadke U, et al. Cardiovascular risk in newly diagnosed type 2 diabetes patients in India. PLOS ONE. 2022;17(3):e0263619. doi:1371/journal.pone.0263619
12. The changing patterns of cardiovascular diseases and their risk factors in the states of India: The Global Burden of Disease Study 1990–2016. Lancet Glob Health. 2018;6(12):e1339-e1351. doi:1016/S2214-109X(18)30407-8
13. Rai B, Bramhankar M. Tobacco use among Indian states: Key findings from the latest demographic health survey 2019–2020. Tob Prev Cessation. 2021;7(March):1-2. doi:18332/tpc/132466
14. Gupta R, Guptha S, Sharma KK, Gupta A, Deedwania P. Regional variations in cardiovascular risk factors in India: India heart watch. World J Cardiol. 2012;4(4):112-120. doi:4330/wjc.v4.i4.112
15. Mohammad R, Bansod DW. Hypertension in India: A gender-based study of prevalence and associated risk factors. BMC Public Health. 2024;24(1):2681. doi:1186/s12889-024-20097-5