Clinical relevance of screening for ECG abnormalities in 67-year-old Danes: a population-based cohort study from the Viborg Screening Programme (VISP) Authors: David Van Der Giessen, Annette Høgh, Dorte Svenstrup, Jes S Lindholt, Marie Dahl Published: 2025 ABSTRACT AbstractObjectivesTo describe the point prevalence of major ECG abnormalities, their coexistence with screen-detected cardiovascular disease (CVD) and the proportion requiring referral for cardiac work-up and interventions.DesignA population-based cohort study.SettingPrimary and secondary care settings in Denmark.ParticipantsSince 2014, all 67-year-old Danish men and women from Viborg municipality were invited to the ‘Viborg Screening Programme’ (VISP).InterventionsVISP includes screening for lower extremity artery disease (LEAD), carotid plaque (CP), abdominal aortic aneurysm (AAA), hypertension, diabetes mellitus and cardiac conditions.Primary outcome measuresA single resting 12-lead ECG was recorded and coded using the Minnesota criteria. Major ECG abnormalities were divided into rhythm and rate disorders, signs of myocardial damage and conduction disorders.ResultsOver the first 5 years, 4612 (83.8% of those invited) were screened, with 4437 (96.4%) undergoing an ECG. We found major ECG abnormalities in 152 (3.4%), including 92 (2.1%) rhythm and rate disorders, 28 (0.6%) with signs of myocardial damage and 32 (0.7%) with conduction disorders. Fifty-nine (1.3%) had newly screen-detected ECG abnormalities, and 34 (0.8%) required intervention after cardiac consultation (32 medication adjustments and/or 11 advanced treatments). Participants with major ECG abnormalities had a higher frequency of coexisting CVD, including CP (46.4% vs 38.1%; p=0.040), LEAD (9.5% vs 5.3%; p=0.026) and AAA (2.7% vs 0.9%; p=0.032) and were also more common in men than in women (4.7% vs 2.2%; p<0.001). Adjusted OR showed significantly higher risk for screen-detected major ECG abnormalities in current smokers (aOR 2.07, 95% CI 1.05 to 4.09), but a lower risk in participants using lipid-lowering medication (aOR 0.39, 95% CI 0.19 to 0.80).ConclusionThis study provides a detailed overview of major ECG abnormalities in a population-based cardiovascular screening context. Although the yield of ECG screening was low, ECG may still add value as a simple and low-cost tool within multimodal programmes, enabling timely detection and treatment.Trial registration numberNCT03395509.Keywords: Cardiovascular Disease, Coronary heart disease, Myocardial infarction, Adult cardiology, Preventive Health Services STRENGTHS AND LIMITATIONS OF THIS STUDY This study presents a uniquely comprehensive overview of the point prevalence and cardiac follow-up of major ECG abnormalities. This study presents a uniquely comprehensive overview of the point prevalence and cardiac follow-up of major ECG abnormalities. This study includes detailed cardiovascular profiling of 67-year-old Viborg Screening Programme participants, including both men and women, addressing the persistent underrepresentation of women in cardiovascular research. This study includes detailed cardiovascular profiling of 67-year-old Viborg Screening Programme participants, including both men and women, addressing the persistent underrepresentation of women in cardiovascular research. The high participation rate (83.8%) strengthens the generalisability of findings to the target population. Yet, non-attenders, who may be at higher cardiovascular risk, may be underrepresented in this study. The high participation rate (83.8%) strengthens the generalisability of findings to the target population. Yet, non-attenders, who may be at higher cardiovascular risk, may be underrepresented in this study. The small sample sizes in certain ECG subgroups (conduction disorders and myocardial damage) limit statistical power for between-group comparisons. The small sample sizes in certain ECG subgroups (conduction disorders and myocardial damage) limit statistical power for between-group comparisons. Introduction Major ECG abnormalities are linked to cardiovascular disease (CVD),1–8 which is the leading cause of preventable morbidity and mortality worldwide.9 The prevalence of major ECG abnormalities in the general population is estimated at 1.7%–11.3%, increasing with rising age.2–4 6 10–12 ECG abnormalities may be asymptomatic, and the first serious event, like sudden cardiac death, myocardial infarction (MI), heart failure or stroke, may occur without prior diagnosis. Screening for CVD with ECGs is recommended only in symptomatic or high-risk individuals,8 13 14 based on traditional risk prediction tools like the Systematic COronary Risk Evaluation.13 However, early detection and treatment of both asymptomatic and symptomatic CVD can cost-effectively prevent escalation to CVD events and early death.15–17 Jørgensen et al4 showed that the addition of ECG recordings can improve CVD risk prediction in individuals ≥65 years. Similarly, results from a recent meta-analysis indicate that population-based screening for atrial fibrillation (AF) from the age of 65 years can be a cost-effective strategy.12 Moreover, the European Society of Cardiology (ESC) recommends opportunistic screening for AF in individuals ≥65 years of age.13 In a multiple CVD screening trial, the Danish Cardiovascular Screening Trial (DANCAVAS), including an ECG-gated CT scan, found an all-cause mortality reduction of 11% in 65- to 69-year-old men.17 The majority of existing studies focus on different ECG abnormalities in combination with CVD risk without considering the benefit of screening for the broad spectrum of ECG abnormalities requiring treatment.8 14 Moreover, cardiac follow-up after screen-detected ECG abnormalities is unreported in the existing literature. This paper is a substudy of the ‘Viborg Screening Programme’ (VISP), which invited 67-year-old men and women to participate in a population-based screening for multiple cardiovascular conditions, including a single resting ECG.18 The overall screen-detected prevalence of cardiac arrhythmia in the VISP cohort has been reported previously.18 In this substudy, we investigated the point prevalence of major ECG abnormalities, classified according to the Minnesota criteria19 and sex. We also report the coexistence of ECG abnormalities and screen-detected CVD, along with the proportion requiring referral for cardiac work-up and interventions. Methods From August 2014, all 67-year-old residents in the municipality of Viborg, Denmark received an invitation for tax-paid cardiovascular screening called the VISP20 with ongoing enrolment. ECGs were first included in VISP from October 2014, as it was time-consuming to arrange the necessary infrastructure including referral for cardiac work-up after screen-detected major ECG abnormalities. This substudy includes those invited during the first 5 years of enrolment. In brief, data from the VISP cohort include self-reported information and screening-derived cardiovascular findings. Self-reported data included comorbidities, current medication use, together with alcohol intake (high intake defined as a weekly alcoholic unit intake >7 for women and >14 for men) and smoking habits (never, previous and current smoker). Congestive heart failure, hypertension, age, diabetes, stroke, vascular disease, age, sex category (CHA2DS2-VASc)21 score was calculated based on self-reported data. During screening, waist circumference was measured. A high waist circumference was defined as >94 cm for men and >80 cm for women. Body mass index (BMI, kg/m²) was then calculated based on self-reported height and weight. VISP included assessment for the following CVDs20: major ECG abnormalities (a single resting 12-lead ECG), HT (blood pressure ≥160/100 mm Hg), carotid plaque (CP) (a focal structure stretching into the arterial lumen for ≥0.5 mm or ≥50% of the surrounding vessel), abdominal aortic aneurysm (AAA) (infrarenal aortic diameter ≥30 mm on ultrasound), lower extremity artery disease (LEAD) (ankle-brachial index <0.9 or ≥1.4) and diabetes mellitus (DM) (HbA1c of ≥48 mmol/mol). In this study, sex was defined based on the Danish social security number (the last number of the Danish social security number indicates biological sex, with even numbers for women and odd numbers for men). Supplementary cardiology data and follow-up Trained screening nurses performed single resting 12-lead ECGs with participants in supine position, using a Welch Allyn electrocardiograph set at a 10 mm/mV calibration and a speed of 25 mm/s. Subsequently, once a week, the ECGs were delivered to our collaborating Department of Cardiology, Viborg Regional Hospital, Viborg, Denmark, where specialised cardiac nurses assessed all ECGs. If any abnormalities were identified, a cardiologist stratified those with major ECG abnormalities according to the Minnesota classification.19 Major ECG abnormalities were grouped and defined as: Rhythm and rate disorders including AF or atrial flutter and supraventricular or ventricular arrhythmia/tachycardia (Minnesota codes 8-2, 8-3, 8-4-1, 8-4-2). Rhythm and rate disorders including AF or atrial flutter and supraventricular or ventricular arrhythmia/tachycardia (Minnesota codes 8-2, 8-3, 8-4-1, 8-4-2). Signs of myocardial damage including major Q/QS-wave abnormalities, ST deviation and T-wave inversion (Minnesota codes 1-1, 1-2, 1-3, 4-1, 4-2, 4-3, 5-1, 5-2, 9-2). Signs of myocardial damage including major Q/QS-wave abnormalities, ST deviation and T-wave inversion (Minnesota codes 1-1, 1-2, 1-3, 4-1, 4-2, 4-3, 5-1, 5-2, 9-2). Conduction disorders including third-degree atrioventricular block (AV) block, second-degree AV block (Mobitz type II), Wolff-Parkinson-White pattern (WPW pattern), left bundle branch block (LBBB), bifascicular block and prolonged QTc (Minnesota codes 6-1, 6-2, 6-4, 7-1, 7-4). Conduction disorders including third-degree atrioventricular block (AV) block, second-degree AV block (Mobitz type II), Wolff-Parkinson-White pattern (WPW pattern), left bundle branch block (LBBB), bifascicular block and prolonged QTc (Minnesota codes 6-1, 6-2, 6-4, 7-1, 7-4). In case of multiple coexisting major ECG abnormalities, the cardiologist assigned each participant to the Minnesota code with the most potential for harm. All participants with major ECG abnormalities without optimal medical treatment according to ESC guidelines or with newly screen-detected major ECG abnormalities were invited to the cardiology outpatient clinic, Viborg Regional Hospital, to receive further follow-up, including echocardiography and treatment in accordance with international guidelines.13 The electronic medical records of all VISP participants with major ECG abnormalities were manually reviewed in September 2022 to assess cardiological follow-up after screening. These data were stratified by major ECG abnormality group and included data on changes in medication and any further examinations consisting of: echocardiography, 24-hour blood pressure monitoring, 24- or 72-hour ECG monitoring, cardiac CT, cardiac MRI, coronary angiography and advanced treatments, including percutaneous coronary intervention (PCI), coronary artery bypass grafting (CABG), implantation of pacemaker and cardioversion. Statistical methods Descriptive statistics were used to present our results. Categorical variables were shown as absolute numbers (n) and frequencies (%), using the χ2 test to compare frequencies between participants with and without major ECG changes, and frequencies between different major ECG abnormality groups. The CHA2DS2-VASc score, the only continuous variable, was non-normally distributed and reported as median with range. Comparisons were made using the non-parametric Wilcoxon rank test for two groups or the Kruskal-Wallis test for three groups. Unadjusted and adjusted logistic regression analyses were used to estimate ORs with 95% CIs for factors associated with newly screen-detected major ECG abnormalities. Multivariable logistic regression models were adjusted for variables that were significant in the unadjusted analyses. A p-value of ≤0.05 was considered significant. All statistics were analysed using Stata (V.17.0). Observations with only one or two observations were noted as ‘<3’ to comply with the General Data Protection Regulation. Results Over the first 5 years of enrolment, 5505 citizens were invited to participate in the VISP. Of these, 4612 (83.8%) accepted the invitation. A total of 175 participants were excluded (3.8%), including 10 (0.2%) due to missing screening data and 165 (3.6%) who had no ECG taken. This left 4437 (80.6%) participants for further analysis, among whom no sex differences were seen (figure 1). Participant characteristics with self-described risk factors, comorbidities, medication use along with screening outcomes are summarised per subgroup in table 1. Flowchart of the invited Viborg Screening Programme (VISP) population. Table 1Participant characteristics ± major ECG abnormalities and subdivided into three categories of major ECG abnormalitiesTotaln (%)Without ECG abnormalitiesn (%)With ECG abnormalitiesn (%)Rhythm and rate disordersn (%)Signs of myocardial damagen (%)Conduction disordersn (%)Total44374285 (96.6)152 (3.4)92 (2.1)28 (0.6)32 (0.7) Women2201 (49.6)2153 (97.8)48 (2.2)**20 (0.9)**14 (0.6)14 (0.6) Men2236 (50.4)2132 (95.4)104 (4.7)**72 (3.2)**14 (0.6)18 (0.8) BMI (kg/m2), missing n=75 (1.9%) <251664 (37.5)1623 (37.9)*41 (27.0)17 (18.5)**12 (42.9)12 (37.5) 25–301772 (38.8)1661 (38.8)61 (40.1)37 (40.2)14 (50.0)10 (31.3) >30976 (22.0)931 (21.7)*45 (29.6)34 (37.0)**<3†9 (28.1) High waist circumference, missing n=50 (1.1%)3022 (68.9)2906 (68.6)*116 (78.4)73 (82.0)*17 (60.7)26 (83.9) Smoking, missing n=38 (0.9%) Never1818 (41.0)1769 (41.3)*49 (32.2)30 (32.6)9 (32.1)10 (31.3) Previous1826 (42.0)1793 (41.8)69 (45.4)45 (48.9)12 (42.9)12 (27.5) Current719 (16.2)689 (16.1)30 (19.7)14 (15.2)7 (25.0)9 (28.1) High alcohol intake, missing n=195 (4.4%)862 (20.3)835 (20.4)27 (19.2)17 (20.0)5 (19.2)5 (16.7)Self-reported comorbidities, missing n=51 (1.2%) Hypertension1978 (45.1)1890 (44.6)**88 (59.5)58 (65.2)**17 (60.7)13 (41.9) Previous myocardial infarction199 (4.5)184 (4.3)**15 (10.1)9 (10.1)*4 (14.3)*<3† Angina pectoris181 (4.1)170 (4.0)*11 (7.4)9 (10.1)*0<3† Previous stroke231 (5.3)222 (5.2)9 (6.1)7 (7.9)<3†<3† Diabetes mellitus439 (10.0)418 (9.9)21 (14.2)14 (15.7)5 (17.9)<3† Chronic obstructive pulmonary disease325 (7.4)319 (7.5)6 (4.1)4 (4.5)0<3† CHA2DS2-VASc score, median (range)2 (1–7)2 (1–7)2 (1–6)2 (1–6)2 (1–4)2 (1–4)Self-reported medication use, missing n=36 (0.8%) Antiplatelet833 (18.9)811 (19.1)22 (14.8)9 (10.0)*6 (21.4)7 (22.6) Non-vitamin K antagonist oral anticoagulants128 (2.9)90 (2.1)**38 (25.0)36 (39.1)**0<3† Vitamin K antagonist oral anticoagulants103 (2.3)71 (1.7)**32 (21.1)31 (33.7)**<3†0 Antihypertensives1859 (42.4)1766 (41.5)**93 (61.4)63 (70.0)**16 (57.1)14 (45.2) Of which beta blocker540 (12.3)494 (11.6)**46 (30.9)35 (38.9)**7 (25.0)*4 (12.9) Antiarrhythmics14 (0.3)8 (0.2)**6 (4.0)6 (6.7)**00 Lipid-lowering1373 (31.2)1320 (31.0)53 (35.6)40 (44.4)*7 (25.0)6 (19.4) Antidiabetics336 (7.6)319 (7.5)17 (11.4)11 (12.2)4 (14.3)<3†Screening results Carotid plaque, missing n=5 (0.1%)1700 (38.4)1630 (38.1)*70 (46.4)47 (51.1)*11 (39.3)12 (38.7) Lower extremity arterial disease, missing n=33 (0.7%)238 (5.4)224 (5.3)*14 (9.5)9 (10.1)*3 (10.7)<3† Hypertension, missing n=30 (0.7%)655 (14.9)629 (14.7)26 (17.6)13 (14.6)7 (25.0)6 (19.4) Diabetes mellitus, missing n=8 (0.2%)398 (9.0)379 (8.9)19 (12.6)13 (14.1)<3†4 (12.9) Abdominal aorta aneurysm, missing n=10 (0.2%)43 (1.0)39 (0.9)*4 (2.7)<3†<3†<3†* p<0.05, ** p<0.001.†Insufficient observations to report data.BMI, body mass index; CHA2DS2-VASc score, congestive heart failure, hypertension, age, diabetes, stroke, vascular disease, age, sex category score. Participant characteristics ± major ECG abnormalities and subdivided into three categories of major ECG abnormalities * p<0.05, ** p<0.001. †Insufficient observations to report data. BMI, body mass index; CHA2DS2-VASc score, congestive heart failure, hypertension, age, diabetes, stroke, vascular disease, age, sex category score. Major ECG abnormalities A total of 152 (3.4%) VISP participants had major ECG abnormalities, with a higher prevalence in men than in women (4.7% vs 2.2%; p<0.001). Participants with major ECG changes were significantly more likely to have comorbidities, including HT (59.5% vs 44.6%; p<0.001), previous MI (10.1% vs 4.3%; p<0.001) and angina pectoris (7.4% vs 4.0%; p=0.040). No differences were seen for DM or previous stroke. The group with major ECG abnormalities also reported to use more medication related to CVD, like (non)-vitamin K antagonist oral anticoagulants, HT medication and antiarrhythmics. However, no differences were observed regarding cholesterol-lowering, antiplatelet or diabetes medication. Table 1 also displays a significantly higher coexistence of screen-detected CP (46.4% vs 38.1%; p=0.040), LEAD (9.5% vs 5.3%; p=0.026) and AAA (2.7% vs 0.9%; p=0.032) among the group with major ECG abnormalities. During VISP assessment, 63 participants with major ECG abnormalities were referred for cardiac outpatient follow-up. Of these, 34 (0.8%) received an intervention, most commonly a medication change (n=32, 0.7%), while another nine were monitored because treatment was considered potentially necessary at a later stage. In total, 11 participants (0.2%) underwent advanced treatment (PCI, CABG, pacemaker implantation or cardioversion). The false-positive rate was 0.5% (20 participants were called in but did not require interventions or follow-up out of 4394 participants; 4374 true negatives and 20 false positives). Overall, 59 participants (38.8% of those with major ECG abnormalities) were identified for the first time, corresponding to 1.3% of the total study population. Overall, no procedure-related complications were observed among participants who underwent cardiac interventions based on screen-detected ECG abnormalities. Rhythm and rate disorders Rhythm and rate disorders were seen among 92 (2.1%) participants, with an overrepresentation of AF followed by atrial flutter and supraventricular tachycardia (table 2). Compared with those without rhythm and rate disorders, participants with these disorders were significantly more likely to be men (3.2% vs 0.9%; p<0.001), have a BMI >30 (37.0% vs 21.7%; p<0.001), a high waist circumference (82.0% vs 68.6%, p=0.007), HT (65.2% vs 44.7%; p<0.001), a history of MI (10.1% vs 4.4%; p=0.011), and angina pectoris (10.1% vs 4.0%; p=0.004) (table 1). For the CHA2DS2-VASc score, no difference was observed between participants with and without rhythm and rate disorders (table 1). Before screening, participants with rhythm and rate disorders were less likely to be on antiplatelet therapy but were more likely to receive other cardiovascular medications, including anticoagulants, antihypertensives, antiarrhythmics and/or lipid-lowering drugs (table 1). CP (51.1% vs 38.1%; p=0.011) and LEAD (10.1% vs 5.3%; p=0.047) were significantly more common in participants with rhythm and rate disorders (table 1). Table 2Major ECG abnormalities described by Minnesota code, divided by sexECG groupECG abnormality (n=152)Minnesota code20TotalMenn (%)Womenn (%)Rhythm and rate disorders9272 (78.3%)20 (21.7%)Atrial fibrillation/flutter and supraventricular or ventricular arrhythmia/tachycardia8-2, 8-3, 8-4-1, 8-4-2927220Signs of myocardial damage2814 (50.0%)14 (50.0%)Major Q/QS-wave abnormalities1-1, 1-2, 1-33<3*<3*ST-deviation4-1, 4-2, 4-31468T-wave inversion5-1, 5-21055ST-elevation9-2<3<3*0Conduction disorders3218 (56.3%)14 (43.8%)Second- and third-degree AV block6-1, 6-2330Wolff-Parkinson-White pattern6-4<3*0<3*Left bundle branch block and bifascicular block7-1, 7-4281513Prolonged QTc–000*Insufficient observations to report data.AV block, atrioventricular block. Major ECG abnormalities described by Minnesota code, divided by sex *Insufficient observations to report data. AV block, atrioventricular block. Of the 92 cases with rhythm and rate disorders, 18 (0.4% of all participants) participants were newly diagnosed through VISP screening. Of these, 17 (0.4%) underwent medication adjustments during cardiac follow-up, and 5 (0.1%) required advanced treatments (eg, PCI stenting, cardioversion or pacemaker implantation) (table 3). Table 3Follow-up divided by ECG abnormality groupTotaln (%)Rhythm and rate disorders n (%)Signs of myocardial damage n (%)Conduction disorders n (%)Total152922832Screen-detected59 (38.8)18 (19.6)20 (71.4)21 (65.6)Referred for out-clinic cardiac work-up63 (41.4)20 (21.7)23 (82.1)20 (62.5)Further examination*20 (13.2)<38 (28.6)10 (31.3)Change in medication32 (21.1)17 (18.5)12 (42.9)3 (9.4)Advanced treatment†11 (7.2)5 (5.4)3 (10.7)3 (9.4)Intervention total‡34 (22.4)17 (18.5)12 (42.9)5 (15.6)No intervention, referred for later control9 (5.9)<33 (10.7)4 (12.5)*24-hour blood pressure monitoring, 24- or 72-hour ECG monitoring, cardiac CT, cardiac MRI, coronary angiography.†Percutaneous coronary intervention (PCI), coronary artery bypass grafting (CABG), implantation of pacemaker and cardioversion.‡Changes in medication and/or advanced treatment. Follow-up divided by ECG abnormality group *24-hour blood pressure monitoring, 24- or 72-hour ECG monitoring, cardiac CT, cardiac MRI, coronary angiography. †Percutaneous coronary intervention (PCI), coronary artery bypass grafting (CABG), implantation of pacemaker and cardioversion. ‡Changes in medication and/or advanced treatment. Signs of myocardial damage Signs of previous or current myocardial damage were seen in 28 (0.6%) participants, including major Q/QS-wave abnormalities, ST deviation and T-wave inversion (table 2). Participants with signs of myocardial damage were more likely to have a history of MI (14.3% vs 4.5%; p=0.013) (table 1). However, no significant association was found with VISP screen-detected CVD. Among the 28 participants with signs of myocardial damage, 20 (0.5% of all VISP participants) were newly identified through screening. In total, 12 (0.3%) required medication adjustments and 3 (0.1%) underwent PCI or CABG (table 3). Conduction disorders Overall, conduction disorders were seen in 32 (0.7%) of the participants, including second-degree AV block (Mobitz type II), WPW pattern, LBBB and bifascicular block with no significant sex difference (table 1). No cases of prolonged QTc were found. Additionally, no self-reported comorbidities, medication use or screening results were significantly more prevalent in this group. Conduction disorders were newly identified through screening in 21 (0.5% of the total number of VISP participants). Among them, 3 (0.1%) required medication adjustments, 10 (0.2%) needed further examination, 5 (0.1%) required intervention including 3 (0.1%) changes in medication and 3 (0.1%) underwent pacemaker implantation. Notably, all participants who received a pacemaker had second-degree AV block (Mobitz type II) detected during screening but third-degree block during 24-hour ECG monitoring (table 3). Factors associated with screen-detected ECG abnormalities In participants with newly screen-detected major ECG abnormalities (n=59) adjusted analyses (adjusted for current smoking, the use of lipid-lowering medication, screen-detected HT and screen-detected LEAD) showed a significantly higher risk for screen-detected major ECG abnormalities in current smokers (aOR 2.07, 95% CI 1.05 to 4.09), but a lower risk in participants using lipid-lowering medication (aOR 0.39, 95% CI 0.19 to 0.80) (table 4). Table 4Factors associated with newly screen-detected major ECG abnormalitiesOR (95% CI)aOR (95% CI)Men1.44 (0.86 to 2.43)BMI (kg/m2), missing n=75 (1.9%) <25Reference (1) 25–300.85 (0.48 to 1.51) >300.75 (0.37 to 1.53)High waist circumference, missing n=50 (1.1%)1.53 (0.82 to 2.86)Smoking, missing n=38 (0.9%) NeverReference (1) Previous0.98 (0.53 to 1.80) Current1.95 (1.01 to 3.75)*2.07 (1.05 to 4.09)*High alcohol intake, missing n=195 (4.4%)0.85 (0.43 to 1.69)Self-reported comorbidities, missing n=51 (1.2%) Hypertension0.92 (0.55 to 1.55) Previous myocardial infarctionn/a† Angina pectoris0.40 (0.06 to 2.94) Previous stroke0.31 (0.04 to 2.27) Diabetes mellitus0.85 (0.34 to 2.13) Chronic obstructive pulmonary disease0.68 (0.21 to 2.18)Self-reported medication use, missing n=36 (0.8%) Antiplatelet0.58 (0.26 to 1.29) Non-vitamin K antagonist oral anticoagulants1.0 (n/a)† Vitamin K antagonist oral anticoagulants0.72 (0.10 to 5.27) Antihypertensives0.83 (0.49 to 1.42) Of which beta blocker0.53 (0.19 to 1.46) Antiarrhythmicsn/a† Lipid-lowering0.40 (0.20 to 0.82)*0.39 (0.19 to 0.80)* Antidiabetics1.14 (0.45 to 2.88)Screening results Carotid plaque, missing n=5 (0.1%)0.91 (0.53 to 1.56) Lower extremity arterial disease, missing n=33 (0.7%)2.49 (1.12 to 5.56)*1.91 (0.73 to 3.29) Hypertension, missing n=30 (0.7%)2.02 (1.12 to 3.66)*1.78 (0.97 to 3.29) Diabetes mellitus, missing n=8 (0.2%)1.17 (0.50 to 2.74) Abdominal aorta aneurysm, missing n=10 (0.2%)1.81 (0.24 to 13.36)Adjusted ORs are adjusted for current smoking, self-reported lower extremity artery disease, self-reported lipid lowering medication use and screen-detected hypertension.*p<0.05, **p<0.001.†Not reported because of perfect separation.aOR, adjusted OR; BMI, body mass index; n, numbers. Factors associated with newly screen-detected major ECG abnormalities Adjusted ORs are adjusted for current smoking, self-reported lower extremity artery disease, self-reported lipid lowering medication use and screen-detected hypertension. *p<0.05, **p<0.001. †Not reported because of perfect separation. aOR, adjusted OR; BMI, body mass index; n, numbers. Discussion In a population-based screening for CVD among 67-year-old Danes, the point prevalence of major ECG abnormalities was 3.4% (n=152). Of these, 38.8% (n=59) were newly diagnosed through screening, corresponding to 1.3% of all screened participants. This point prevalence aligns with existing literature2–4 6 10–12 22 23; however, no consensus exists on which ECG abnormalities should be classified as major.1–3 6 11 12 22 This lack of standardisation contributes to substantial heterogeneity across studies, leading to wide variations in reported prevalence rates.2 3 6 11 12 22 Major ECG abnormalities are known predictors of CVD, but their relationship with subclinical CVD remains unknown. We observed a higher prevalence of screen-detected CVD (CP, LEAD and AAA) in participants with coexisting major ECG abnormalities. In contrast, a subanalysis of the Multi-Ethnic Study of Atherosclerosis (MESA) study, involving participants without known CVD (mean age of 62 years), found no association between major ECG abnormalities and subclinical atherosclerosis, here defined as increased common carotid intima-media thickness or coronary artery calcium.24 This discrepancy may be due to age variations between populations, the inclusion of participants with prior CVD in VISP, the use of intima-media thickness as a marker in MESA and variations in the definitions of major ECG abnormalities. Rodin et al also reported an association between AAA and major ECG abnormalities; however, this relationship was observed only in univariate analyses and faded in multivariate analyses.25 In our cohort, unadjusted univariate analyses showed a non-significant trend (OR 1.81, 95% CI 0.24 to 13.36). The low prevalence of both screen-detected AAA (1.0%) and newly identified ECG abnormalities (1.3%) prevented us from performing meaningful multivariate analyses. Unlike the studies by Denes et al6 and Lee et al,7 we found no increased prevalence of screen-detected DM or HT in participants with major ECG abnormalities, though differences in measuring methods and sample sizes might explain this discrepancy. Adjusted analyses also showed that participants using lipid-lowering medicine were less likely to have screen-detected major ECG abnormalities, which might be described by the effect of statins on AF.26 While guidelines recommend CVD risk assessment,13 no such consensus exists for major ECG abnormalities.8 Our results show that people with major ECG abnormalities may also benefit from screening for CP, AAA and LEAD, emphasising the potential added value of multimodal CVD screening. More specifically, people with rhythm and rate disorders may benefit from additional screening for CP and LEAD. Rhythm and rate disorders The most prevalent major ECG abnormalities observed were AF and atrial flutter, affecting 2.1% of the VISP cohort. This prevalence aligns with other studies, which report rates ranging from 0.5% to 5.5% in individuals around the age of 67 years.7 10–12 22 26 Among the 92 participants with rhythm and rate disorders, 18 (19.6%) were newly identified during screening. This detection rate is higher than that reported in a substudy of the systematic screening for AF (STROKESTOP) study, where only 3% was discovered through ECG screening.11 Consistent with previous literature, we found a higher prevalence of rhythm and rate disorders in men2 10–12 22 and in individuals with elevated BMI or high waist circumference.2 9 Unlike our findings, the MESA study24 did not show a higher prevalence of subclinical atherosclerosis in participants with AF. Additionally, our finding of a higher prevalence of AF in participants with LEAD has, to our knowledge, not been described before. Contrary to other studies,1 9 we found no significant differences in the prevalence of DM or HT. Signs of myocardial damage We found ECG-verified significant myocardial damage, both past and present, in 0.6% of all VISP participants. Previous studies have reported a prevalence of signs of myocardial damage ranging from 0.5% to 11%2 3 7 despite substantial heterogeneity across age groups and varying definitions of significant myocardial damage based on ECG findings.8 Although CVD is more frequent among men,9 we found no sex differences according to signs of myocardial damage, which aligns with existing literature.22 23 De Bacquer et al reported a higher prevalence of DM and obesity in participants with ECGs with myocardial damage,2 a finding we confirmed. Even though ECG abnormalities associated with myocardial damage are not significantly linked to subclinical atherosclerosis,24 71.4% of participants with signs of myocardial damage were newly diagnosed through screening, and 42.9% required intervention after cardiology consultation. This suggests that this subpopulation represents an asymptomatic group with the potential for life-saving preventive intervention. Conduction disorders Overall, conduction disorders were seen in 0.7% of our cohort, matching prevalence rates ranging from 0.1% to 0.8% in previous studies,2 7 9 22 with LBBB being the most common conduction disorder. Unlike most studies, which report an overrepresentation of women with conduction disorders,3 22 23 we found no significant sex differences. Most VISP participants with screen-detected conduction disorders were referred for cardiac follow-up, but only 15.6% required intervention. Notably, no participants with screen-detected LBBB required intervention beyond secondary medical prophylaxis for HT or high serum cholesterol, suggesting that LBBB may represent a major ECG abnormality with limited clinical impact on overall health and a high risk of overdiagnosis. Referral for cardiac work-up To our knowledge, this study is the first to describe the frequency of referral for cardiac work-up including interventions among screening participants with major ECG abnormalities. Out of the 152 participants with major ECG abnormalities, 59 (1.3% of the total screened population) were newly diagnosed during VISP participation. Furthermore, 63 (1.4% of the total screened population) were referred for cardiac work-up, and 34 (0.8%) were offered an intervention, with medication adjustments being the most frequent course of action. The referral rate of 1.4% is comparable to that observed in cancer screening programmes, whereas the false-positive rate of 0.5% is markedly lower than that reported for mammography screening, which is offered in many countries.27 28 Population-based ECG screening Our findings indicate that screening for ECG abnormalities supports early detection and timely cardiac interventions, advocating for the incorporation of a single ECG into population-based screening as part of a nationally funded, multimodal CVD screening programme. Recently, there has been a call for screening and prevention strategies to reduce the incidence of sudden cardiac deaths.29 A Danish study analysing 54 025 deaths found that sudden cardiac death accounted for 12.6% of all deaths. Notably, 44.5% of these occurred in individuals without a prior hospital-based diagnosis of CVD, although the incidence was higher among those with known CVD.30 Ultimately, such a strategy could reduce morbidity, hospitalisations and mortality, thereby improving overall public health outcomes. The optimal age for ECG screening remains unknown.15 The DANCAVAS study showed a cost benefit when 65- to 74-year-old participants were invited to combined CVD screening.17 Additionally, both systematic and opportunistic screening for AF are described as being more cost-effective than usual practice for screening participants ≥65 years of age.15 31 Screening for AF with 12-lead ECGs performed by cardiac nurses in a set-up like VISP showed a sensitivity of 97.1% and a specificity of 100%.31 However, the sensitivity of screening for other major ECG abnormalities remains unknown. The ESC emphasises that pulse palpation (sensitivity 87%–97% and specificity 70%–81%) could replace single ECGs (sensitivity 94%–98% and specificity 75%–95%) as a simple and cheap alternative in screening for AF.13 26 While AF was the most common major ECG abnormality in our study (92 out of 152), approximately 40% (60 out of 152) of the detected findings were other major ECG findings abnormalities that can be diagnosed only using a 12-lead ECG. The use of a single 12-lead ECG in the VISP may underdiagnose paroxysmal AF and flutter in an estimated 10%–40% of cases.11 12 Thus, while continuous monitoring is expected to improve AF detection rates, it would require significant financial and logistical resources. Additionally, the procedure would probably be less acceptable to participants and is unlikely to be cost-effective. Invasive examinations and changes in medication were initiated with the intention of achieving a net clinical benefit, yet they also carry potential risks. Moreover, being referred for cardiac outpatient follow-up can induce considerable stress in screened individuals, which is particularly concerning when resulting from a false-positive finding. However, among participants who underwent cardiac interventions based on newly detected ECG abnormalities in VISP, no procedure-related complications were observed. No adverse effects related to screening were recorded. Strengths and limitations A major strength of our study is the detailed cardiovascular profile of 67-year-old VISP participants, including both men and women, and its approach to addressing the persistent underrepresentation of women in cardiovascular studies.31–33 The uniform age of all participants represents a strength as it eliminates potential confounding effects of age-related ECG changes on the observed prevalence of abnormalities and CVD. Though other studies have found ECG screening may be optimal in individuals aged ≥65 years,4 12 17 this age-specific design may limit population representativeness and generalisability to broader screening populations. Several other limitations need to be acknowledged. First, a key limitation to our study, although perhaps inevitable in any screening study, is that the non-attending part of the population may be those in most need of the screening interventions. Although the uptake in VISP is high19 an earlier study has shown that invitees who decline screening are more likely to have CVD.33 Follow-up is warranted to investigate whether disparities exist in characteristics and outcomes between participants and non-participants. Second, left ventricular hypertrophy (LVH) was not included as a major ECG abnormality in this study, even though LVH is associated with HT, heart failure, CVD and death.34 Underlying causes of LVH are treatable9 and therefore targetable by prophylactic initiatives if identified during screening.9 34 Third, variability in ECG classification may limit comparability across studies, but we used the standardised and commonly used Minnesota coding system and transparent definitions to support reproducibility. Fourth, self-reported comorbidities and medication use are prone to recall bias, potentially overestimating these comorbidities compared with data from the national health registry, as discussed in the VISP protocol paper.20 Fifth, our cohort size is too small to detect significant differences in screen-detected CVD within the conduction disorders and myocardial damage subgroups, increasing the risk of type II errors in comparative analysis. A larger cohort would provide more detailed data, allowing for improved CVD stratification within subgroups and potentially revealing significant differences in CVD prevalence associated with individual major ECG abnormalities. Finally, screening in a community-based setting like VISP requires a robust logistical setup with trained screening personnel and a cardiology department prepared to evaluate ECGs and provide cardiac follow-up. This requirement may impact the feasibility of replicating the VISP model in other settings. In the VISP baseline article,18 the prevalence of cardiac arrhythmia was 2.8%, whereas this study found a prevalence of 3.4%. This discrepancy arises from differences in data sources and cohort size; our study was cross-referenced with participants’ electronic patient journals and included only those participants who underwent an ECG, leading to a more specific dataset. Although the detection rate of newly identified ECG abnormalities was low, screening should be viewed considering the criteria outlined by national health authorities, where even low-prevalence conditions may justify screening if early detection enables timely intervention in otherwise asymptomatic individuals at risk of severe outcomes. For example, national breast cancer screening is offered despite a prevalence of 0.6%.35 Similarly, ECG screening may contribute to timely prevention and treatment. As in any screening programme, feasibility, costs, resource use and potential harms should be carefully evaluated.36 In VISP, these aspects will be assessed and reported in forthcoming studies. Conclusion In a Danish cohort of 67-year-old men and women, we found that 3.4% had major ECG abnormalities, with 1.3% of participants having newly screen-detected major ECG abnormalities. Overall, 0.8% were offered an intervention during cardiac follow-up, with medication modifications as the most frequent intervention. Participants with major ECG abnormalities were more likely to be men and have coexisting screen-detected CP, LEAD or AAA. This study offers a comprehensive overview of the point prevalence and clinical follow-up of major ECG abnormalities in a population-based cardiovascular screening setting. While the yield of ECG screening was low, our findings may suggest that prevalence alone should not determine the relevance of screening. In accordance with established screening principles, timely detection and treatment remain central considerations36 when evaluating the overall value of adding a simple and low-cost tool like ECG to multimodal cardiovascular screening programmes, ultimately contributing to improved population health outcomes. Supplementary Material Reviewer comments Author's manuscript Acknowledgments The authors take this opportunity to express their gratitude to TJ Lynggaard, PB Damsgaard, PL Greve and CS Christensen for their assistance with data collection. Furthermore, we are indebted to the cardiology nurses AM Lundsgaard and JL Jensen for assessing all ECGs over the years. Footnotes Contributors: JSL and MD designed the VISP. DVDG collected the data and categorised ECGs together with DS. MD, AH and JSL conceived and designed the analyses, which were performed and interpreted by MD and DVDG. DVDG drafted the original work, and the coauthors critically revised sequential versions of the manuscript. All approved the final version for publication and agreed to be accountable for all aspects of the work, ensuring its integrity and accuracy. DVDG is the guarantor. Funding: VISP is funded by Viborg Municipality, the Vascular Research Unit and Department of Cardiology, Viborg Regional Hospital, Denmark. The authors are employed within the funding departments, which had no influence on the study design, conduct, analysis or reporting. Competing interests: None declared. Patient and public involvement: Patients and/or the public were not involved in the design, or conduct, or reporting or dissemination plans of this research. Provenance and peer review: Not commissioned; externally peer reviewed. Data availability statement Data are available upon reasonable request. Ethics statements Patient consent for publication Consent obtained directly from patient(s). Ethics approval This study involves human participants but VISP is a preventive initiative launched by the public administration of Viborg Municipality (Committee for Senior Citizens and Health Affairs). It is approved by the Danish Data Protection Agency (1-16-02-232-15). This study involves human participants but an Ethics Committee exempted the study (Journal no. 1-10-72-163-19). Contact: komite@rm.dk exempted this study. Participants gave informed consent to participate in the study before taking part. References Associated Data Supplementary Materials Reviewer comments Author's manuscript Data Availability Statement Data are available upon reasonable request.