Abstract
The Minnesota Code (MC) and Novacode (Nova) are the most widely used electrocardiographic (ECG) classification systems. The comparative strengths of their classifications for Q- and ST-T–wave abnormalities in predicting coronary heart disease (CHD) events and total mortality have not been evaluated separately by gender. We studied standard 12-lead electrocardiograms at rest from 4,988 participants in the Cardiovascular Health Study. Average age at baseline was 73 years, 60% of participants were women 85% were white, and 22% had a history of cardiovascular disease or presence of ECG myocardial infarction by MC or Nova. Starting in 1989 with an average 17-year follow-up, 65% of participants died and 33% had incident CHD in a cohort free of cardiovascular disease at baseline. Of these, electrocardiograms with major Q-wave or major ST-T abnormalities by MC or Nova predicted increased risk for CHD events and total mortality with no significant differences in predictability between men and women. The study also found that women had fewer major Q-wave changes but more major ST-T abnormalities than men. However, there were no gender differences in predicting CHD events and total mortality. In conclusion, ECG classification systems for myocardial infarction/ischemia abnormalities by MC or Nova are valuable and useful for men and women in clinical trials and epidemiologic studies.
Electrocardiographic (ECG) criteria for myocardial infarction and ischemia have been used as evidence for coronary heart disease (CHD) in epidemiologic studies and clinical trials and are a powerful index for assessment of cardiovascular disease (CVD) risk and an important predictor for total mortality.1–8 Comparative strengths of their classification by the Minnesota Code (MC)9,10 and the Novacode (Nova),10–12 have not been evaluated separately by gender. The present study evaluated the value of current MC and Nova for myocardial infarction/ischemia for prediction of incident fatal and nonfatal cardiac events and total mortality and compared the predictive power of each coding system in men and women.
Methods
The Cardiovascular Health Study (CHS) is a population-based prospective cohort study of risk factors for CHD and stroke in men and women ≥65 years of age in 4 United States communities that started in 1989. Eligible participants were sampled from Medicare-eligibility lists in each area. Details of the study design have been described previously.13,14 Eligible participants gave informed consent, and study protocols were approved by institutional review boards at participating institutions. Of 5,888 ECG recordings available from baseline examinations of the CHS cohort, those with incomplete data or inadequate quality and QRS duration ≥120 ms were excluded. The remaining 4,988 participants at baseline were classified into 2 groups, a CVD group (n = 1,113) and a CVD-free group (n = 3,875). CVD at baseline was defined by the presence of ECG myocardial infarction evidence by MC or Nova, a history of clinical myocardial infarction, angina pectoris that was confirmed by retrospective review of hospitalization medical records, or self-report of coronary angioplasty or bypass surgery. Average follow-up was 17 years (maximum 18). Three outcomes were considered in the present analysis: incident CHD events (fatal and nonfatal), CHD death, and all-cause mortality. After baseline, deaths and hospitalization events were ascertained by semiannual follow-up telephone calls to cohort members, review of vital records, and community surveillance of hospitalized and fatal events. The CHS mortality review committee reviewed and adjudicated all fatal events.15,16
Identical electrocardiographs (MAC PC, Marquette Electronics, Inc., Milwaukee, Wisconsin) were used in all clinic sites. Standard 12-lead electrocardiograms at rest were recorded in all participants using strictly standardized procedures.14 All electrocardiograms were processed in a central laboratory (Epidemiologic Cardiology Research Center, Wake Forest University, Winston-Salem, North Carolina) and were classified by the Nova and MC using the 2001 version of the GE Marquette 12-SL program. ECG classification criteria for myocardial infarction/ischemia by Nova/MC in the present study were the same as previously described.8,10,12 Nova myocardial infarction was identified by Nova codes 5.1 to 5.4 and ECG major Q/ST-T changes by Nova was identified as Nova codes 5.1 to 5.6. Similarly, MC myocardial infarction was identified by categories C.1 to C.4 to match Nova myocardial infarction. ECG major Q/ST-T change by MC was identified as categories C.1 to C.6 (Table 1).
Table 1.
Electrocardiographic criteria for myocardial infarction and ischemia by Novacode and Minnesota Code
| Category | Novacode | Minnesota Code* |
|---|---|---|
| Major Q waves | 5.1 | C.1 |
| Moderate Q waves with ST-T abnormalities | 5.2 | C.2 |
| Moderate Q waves without ST-T abnormalities | 5.3 | C.3 |
| Minor Q waves with ST-T abnormalities | 5.4 | C.4 |
| Isolated/major ST abnormalities | 5.5 | C.5 |
| Isolated/major T wave abnormalities | 5.6 | C.6 |
| Minor Q waves | 5.7 | C.7 |
| Minor ST-T abnormalities | 5.8 | C.8 |
| No significant Q waves or ST-T | 5.0 | C.0 |
| Electrocardiographic myocardial infarction by Novacode and Minnesota Code |
5.1–5.4 | C.1–C.4 |
| ECG major Q or major ST-T abnormalities by Novacode and Minnesota Code |
5.1–5.6 | C.1–C.6 |
See Zhang et al‡ for detailed definition.
C standard for combined code of Minnesota Codes 1, 4, and 5 to match hierarchical codes in Novacode 5.
CVD and CVD-free groups at baseline were analyzed separately. Cox proportional hazards regression models were used to test and compare the MC and Nova for myocardial infarction/ischemia and ECG abnormality classifications as predictors of each of the 3 study end points after adjustment for demographic and clinical variables. Clinical and demographic characteristics that were significantly associated with risk of CHD event or total mortality were included in the multivariate model to determine if ECG predictors for risk of CHD events and total mortality differed between women and men and between CVD and CVD-free groups at baseline, which are listed in Tables 2 and 3. All analyses were performed with SAS 9.1.3 (SAS Institute, Cary, North Carolina).
Table 2.
Baseline electrocardiographic findings by Novacode and Minnesota Code criteria and outcomes during average 17-year follow-up
| Characteristics | Total (n = 4,988) |
Prevalent CVD Group |
CVD-Free Group |
||||
|---|---|---|---|---|---|---|---|
| Women (n = 561) |
Men (n = 552) |
p Value* | Women (n = 2,416) |
Men (n = 1,459) |
p Value* | ||
| Age (years) | 73 ±6 | 74±6 | 74± 6 | 0.5846 | 72 ±5 | 73± 6 | 0.0001 |
| Body mass index (kg/m2) | 27 ±5 | 27±5 | 27± 4 | 0.0414 | 27 ±5 | 26± 4 | 0.0364 |
| Systolic blood pressure (mm Hg) | 136 ± 22 | 141 ± 24 | 135 ± 21 | <0.0001 | 136 ± 22 | 135 ± 21 | 0.5097 |
| Diastolic blood pressure (mm Hg) | 71 ± 11 | 69 ± 12 | 71 ± 11 | 0.0416 | 70 ± 11 | 73 ± 11 | <0.0001 |
| African-American | 15.4% | 20.7% | 10.0% | <0.0001 | 15.7% | 14.9% | 0.5563 |
| Current smoking | 12.2% | 13.0% | 9.1% | <0.0001 | 12.8% | 12.0% | <0.0001 |
| Hypertension | 57.8% | 72.7% | 61.4% | <0.0001 | 56.4% | 52.9% | 0.0356 |
| Diabetes mellitus | 15.6% | 20.6% | 25.5% | 0.0542 | 12.2% | 15.5% | 0.0030 |
| Total mortality | 64.9% | 73.8% | 83.3% | 0.0001 | 56.5% | 68.3% | <0.0001 |
| Coronary heart disease death | 15.6% | 22.1% | 29.7% | 0.0037 | 10.6% | 16.2% | <0.0001 |
| Incident coronary heart disease† | 32.6% | 28.9% | 38.6% | <0.0001 | |||
| Myocardial infarction/ischemia by Novacode/Minnesota Code‡ | |||||||
| Myocardial infarction by Novacode/Minnesota Code | 8.9% | 35.7% | 44.0% | 0.0043 | |||
| Major Q wave by Novacode/Minnesota Code | 6.7% | 24.2% | 35.7% | <0.0001 | |||
| Major ST-T change by Novacode/Minnesota Code | 15.2% | 35.1% | 29.4% | 0.0396 | 10.7% | 9.7% | 0.3484 |
| Major Q/ST-T change by Novacode/Minnesota Code | 20.2% | 52.6% | 56.5% | 0.2085 | 10.7% | 9.7% | 0.3484 |
| Major Q/ST-T change by Novacode‡ | |||||||
| Myocardial infarction by Novacode | 6.3% | 23.5% | 33.0% | 0.0005 | |||
| Major Q wave by Novacode | 4.1% | 13.4% | 23.0% | <0.0001 | |||
| Major ST-T change by Novacode | 14.4% | 33.7% | 28.8% | 0.0789 | 9.6% | 9.5% | 0.9160 |
| Major Q/ST-T change by Novacode | 17.5% | 43.5% | 47.1% | 0.2271 | 9.6% | 9.5% | 0.9160 |
| Major Q/ST-T change by Minnesota Code‡ | |||||||
| Myocardial infarction by Minnesota Code | 7.8% | 31.2% | 39.0% | 0.0067 | |||
| Major Q wave by Minnesota Code | 6.4% | 23.7% | 33.7% | 0.0002 | |||
| Major ST-T change by Minnesota Code | 12.2% | 28.9% | 22.8% | 0.0212 | 8.9% | 7.2% | 0.0564 |
| Major Q/ST-T change by Minnesota Code | 17.8% | 49.0% | 52.5% | 0.2411 | 8.9% | 7.2% | 0.0564 |
Between men and women in baseline groups with and without cardiovascular disease.
Incident coronary heart disease includes fatal and nonfatal coronary heart disease as myocardial infarction, coronary death, hospitalized angina pectoris, and coronary revascularization.
See Table 1 for criteria of electrocardiographic myocardial infarction and major Q/ST-T change.
Table 3.
Baseline cardiovascular disease group separately by gender: prevalent electrocardiographic myocardial infarction/ischemia for prediction of coronary heart disease death and total mortality by Novacode and Minnesota Code criteria
| Total (n = 1,113) |
Event (rate)* |
HR† (95% CI) |
p Value‡ | ||
|---|---|---|---|---|---|
| Women | Men | Women | Men | ||
| Coronary heart disease death (n = 124/561 for women, n = 164/552 for men) |
|||||
| Myocardial infarction by Novacode§ | 35 (27%) | 61 (34%) | 1.79 (1.12–2.84) | 1.72 (1.22–2.44) | 0.6984 |
| Myocardial infarction by Minnesota Code§ | 38 (22%) | 64 (30%) | 1.11 (0.71–1.74) | 1.19 (0.84–1.68) | 0.8201 |
| Major Q/ST-T change by Novacode§ | 73 (30%) | 91 (35%) | 2.19 (1.44–3.33) | 2.11 (1.50–2.97) | 0.5321 |
| Major Q/ST-T change by Minnesota Code§ | 71 (26%) | 96 (33%) | 1.52 (1.00–2.31) | 1.68 (1.19–2.37) | 0.6563 |
| Total mortality (n = 414/561 for women, n = 460/552 for men) | |||||
| Myocardial infarction by Novacode | 107 (81%) | 160 (88%) | 1.47 (1.14–1.90) | 1.46 (1.19–1.80) | 0.9280 |
| Myocardial infarction by Minnesota Code | 133 (76%) | 184 (86%) | 1.31 (1.04–1.65) | 1.30 (1.05–1.60) | 0.6246 |
| Major Q/ST-T change by Novacode | 199 (82%) | 229 (88%) | 1.47 (1.17–1.84) | 1.47 (1.20–1.79) | 0.8741 |
| Major Q/ST-T change by Minnesota Code | 218 (79%) | 251 (87%) | 1.40 (1.12–1.75) | 1.39 (1.13–1.70) | 0.5221 |
Event rate is number of events divided by number in that category (percentage) of electrocardiographic myocardial infarction/ischemia by Minnesota Code/Novacode.
Adjusted for key demographic and clinical variables of age, race, education, smoking status, alcohol use, diabetes, hypertension, cancer, increased angina, body mass index, systolic blood pressure, hematocrit, white blood cell count, ankle– brachial index, baseline glucose, insulin, and creatinine.
Assessing equivalence of hazard ratios for men and women.
CI = confidence interval; HR = hazard ratio.
Results
In the present study, of a total of 4,988 participants 63 to 100 years old (60% women and 85% non-Hispanic), there were 22% with a history of CVD or presence of ECG myocardial infarction by the MC or Nova, leaving 78% of participants who were free of CVD at baseline. During an average 17-year follow-up, 33% in the baseline CVD-free group developed incident CHD and 65% in the combined cohort died (Table 2). Data showed that global agreements between classifications of myocardial infarction/ischemia by the MC and Nova were high with a kappa coefficient as high as 0.82 for the combined codes for major Q/ST-T wave abnormalities between the 2 coding systems with no significant difference between men and women. Of those with CVD at baseline, men had more major Q waves than women (24% for women and 36% for men), but women had more major ST-T abnormalities than men by the Nova or MC (35% for women and 29% for men). In contrast, there was no significant difference for combined codes of major Q/ST-T wave abnormalities between women and men (53% for women and 57% for men).
As presented in Table 3, electrocardiograms with major Q/ST-T wave codes for myocardial infarction/ischemia by MC/Nova was a strong predictor for CHD death and total mortality. ECG myocardial infarction by Nova in the baseline CVD group had ≥70% increased risk for CHD death with no difference in predictability between women and men, although the increased risk by ECG myocardial infarction by MC did not reach statistical significance. Presence of major Q wave or major ST-T abnormality was a significant predictor of CHD death for Nova and MC with no difference between men and women. For total mortality, ECG myocardial infarction alone or combined with major Q/ST-T abnormalities showed an increased risk with no differences between the 2 classification systems and no significant gender differences, which had ≥30% significant increased risk for total mortality.
Risk of incident CHD by ECG markers was examined in cohort members who were free of clinical CVD at baseline. Combined ischemic codes by Nova/MC were equally predictive (i.e., no statistically significant difference between men and women) for incident CHD in women and men. Any major or minor ST-T change had ≥36% increased risk for incident CHD in women and men. In addition, electrocardiograms with minor ST-T change (Nova 5.8 or MC C.8) predicted 39% to 85% increased risk for CHD death in men and 43% to 67% increased risk for women, where MC has a higher risk than Nova. For total mortality, electrocardiograms with major ST-T change by Nova/MC predicted 24% to 26% increased risk for total mortality in women, whereas the 22% to 42% increased risk that was found for electrocardiograms with minor ST-T change was more apparent for MC than for Nova with no significant difference between men and women (details are shown in the Appendix, available online).
Discussion
This is the first study to evaluate and compare the Nova and MC ECG classification systems for myocardial infarction/ischemia by gender in a large cohort with long-term follow-up. The present findings are nonetheless consistent with and thus extend findings previously described in recent publications.4–8 In this study, women had fewer major Q-wave abnormalities but more major ST-T abnormalities than men. However, combined codes for major Q/ST-T changes for myocardial infarction/ischemia by MC or Nova had similar predictive value for CHD events and total mortality in women and men. In summary, these results show that MC and Nova are valuable classification systems for ECG myocardial infarction or ischemia with no significant gender differences for prediction of CHD events and total mortality.
| Total (n = 3,875) |
Event (rates)* |
HR (95% CI)† |
p Value‡ | ||
|---|---|---|---|---|---|
| Women | Men | Women | Men | ||
| Incident coronary heart disease (n = 699/2,416 for women, n = 563/1,459 for men) |
|||||
| Novacode 5 for ischemia§ | |||||
| 5.5/5.6 | 103 (45%) | 63 (46%) | 1.98 (1.55–2.53) | 1.36 (1.01–1.85) | 0.0738 |
| 5.7 | 48 (32%) | 46 (38%) | 1.32 (0.95–1.84) | 1.04 (0.75–1.44) | |
| 5.8 | 226 (33%) | 139 (46%) | 1.39 (1.15–1.67) | 1.40 (1.12–1.73) | |
| 5.0 | 322 (24%) | 315 (35%) | reference | reference | |
| Combined Minnesota Code for ischemia§ | |||||
| C.5/5.6 | 94 (44%) | 52 (50%) | 1.83 (1.43–2.34) | 1.62 (1.18–2.24) | 0.7447 |
| C.7 | 41 (29%) | 46 (41%) | 1.21 (0.86–1.70) | 1.23 (0.88–1.72) | |
| C.8 | 148 (37%) | 95 (44%) | 1.51 (1.23–1.86) | 1.62 (1.27–2.07) | |
| C.0 | 416 (25%) | 370 (36%) | reference | reference | |
| Coronary heart disease death (n = 255/2,416 for women, n = 236/1,459 for men) |
|||||
| Novacode 5 for ischemia | |||||
| 5.5/5.6 | 36 (16%) | 28 (20%) | 1.42 (0.92–2.18) | 1.31 (0.82–2.10) | 0.8110 |
| 5.7 | 21 (14%) | 20 (17%) | 1.35 (0.81–2.25) | 1.04 (0.62–1.73) | |
| 5.8 | 90 (13%) | 63 (21%) | 1.43 (1.06–1.93) | 1.39 (1.00–1.93) | |
| 5.0 | 108 (8%) | 125 (14%) | reference | reference | |
| Combined Minnesota Code for ischemia | |||||
| C.5/5.6 | 35 (16%) | 23 (22%) | 1.43 (0.94–2.18) | 1.52 (0.93–2.49) | 0.9969 |
| C.7 | 16 (11%) | 22 (20%) | 1.06 (0.59–1.89) | 1.17 (0.72–1.91) | |
| C.8 | 63 (16%) | 49 (23%) | 1.67 (1.21–2.29) | 1.85 (1.29–2.65) | |
| C.0 | 141 (8%) | 142 (14%) | reference | reference | |
| Total mortality (n = 1,366/2,416 for women, n = 996/1,459 for men) |
|||||
| Novacode 5 for ischemia | |||||
| 5.5/5.6 | 164 (71%) | 103 (75%) | 1.24 (1.03–1.50) | 1.15 (0.91–1.45) | 0.9847 |
| 5.7 | 87 (59%) | 86 (71%) | 1.01 (0.79–1.28) | 1.00 (0.78–1.28) | |
| 5.8 | 411 (61%) | 233 (77%) | 1.11 (0.98–1.27) | 1.22 (1.04–1.44) | |
| 5.0 | 704 (52%) | 574 (64%) | reference | reference | |
| Combined Minnesota Code for ischemia | |||||
| C.5/5.6 | 153 (71%) | 80 (76%) | 1.26 (1.05–1.52) | 1.19 (0.92–1.54) | 0.2293 |
| C.7 | 87 (61%) | 75 (68%) | 1.14 (0.89–1.45) | 0.90 (0.70–1.17) | |
| C.8 | 256 (64%) | 173 (81%) | 1.21 (1.04–1.40) | 1.42 (1.18–1.71) | |
| C.0 | 870 (52%) | 668 (65%) | reference | reference | |
Event rate is number of events divided by number in that category (percentage) of electrocardiographic myocardial infarction/ischemia by Minnesota Code/Novacode.
Adjusted for key demographic and clinical variables of age, race, education, smoking status, alcohol use, diabetes, hypertension, cancer, increased angina, body mass index, systolic blood pressure, hematocrit, white blood cell count, ankle– brachial index, baseline glucose, insulin, and creatinine.
Assessing equivalence of hazard ratios for men and women.
See Table 1 and Zhang et al8 for ischemia criteria. Nova 5.5/5.6 or Minnesota Code C.5/C.6 as major ST-T change, Nova 5.8 or Minnesota Code C.8 as minor ST-T change, and Nova 5.7 or Minnesota Code C.7 as minor Q change.
Abbreviations as in Table 3.
Acknowledgment
We acknowledge the contributions of CHS investigators (http://www.chs-nhlbi.org).
This study was supported by Contracts NHLBI N01-HC-85079 through N01-HC-85086, N01-HC-35129, N01-HC-15103, N01-HC-55222, N01-HC-75150, N01-HC-45133, U01-HL-080295, R01-HL-087652, and R01-HL-088456 from the National Heart, Lung, and Blood Institute, Bethesda, Maryland.
Appendix 1.
Prevalent electrocardiographic ischemia for prediction of coronary heart disease incidence, coronary heart disease death, and total mortality by Novacode and Minnesota Code criteria in baseline cardiovascular disease–free group separately for gender
References
- 1.Denes P, Larson JC, Lloyd-Jones DM, Prineas RJ, Greenland P. Major and minor ECG abnormalities in asymptomatic women and risk of cardiovascular events and mortality. JAMA. 2007;298:978–985. doi: 10.1001/jama.297.9.978. [DOI] [PubMed] [Google Scholar]
- 2.Prineas RJ, Grandits G, Rautaharju PM, Cohen J, Zhang ZM, Crow RS. Long-term prognostic significance of isolated minor electrocardiographic T-wave abnormalities in middle aged men free of clinic cardiovascular disease. The Multiple Risk Factor Study (MRFIT) Am J Cardiol. 2002;90:1391–1395. doi: 10.1016/s0002-9149(02)02881-3. [DOI] [PubMed] [Google Scholar]
- 3.Zhang ZM, Prineas RJ, Case D, Soliman EZ, Rautaharju PM, for ARIC Research Group Comparison of the prognostic significance of the electrocardiographic QRS/T angles in predicting incident coronary heart disease and total mortality (from the Atherosclerosis Risk In Communities Study) Am J Cardiol. 2007;100:844–849. doi: 10.1016/j.amjcard.2007.03.104. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Rautaharju PM, Prineas RJ, Wood J, Zhang ZM, Crow R, Heiss G. Electrocardiographic predictors of new-onset heart failure in men and in women free of coronary heart disease (from the Atherosclerosis Risk in Communities [ARIC] Study) Am J Cardiol. 2007;100:1437–1441. doi: 10.1016/j.amjcard.2007.06.036. [DOI] [PubMed] [Google Scholar]
- 5.Rautaharju PM, Grace S, Nelson JC, Marino-Larsen EK, Psaty BM, Furberg CD, Zhang ZM, Robbins J, Gottdiener JS, Chaves P. Comparison of mortality risk for electrocardiographic abnormalities in men and women with and without coronary heart disease (from the Cardiovascular Health Study) Am J Cardiol. 2006;97:309–315. doi: 10.1016/j.amjcard.2005.08.046. [DOI] [PubMed] [Google Scholar]
- 6.Kumar A, Prineas RJ, Arnold AM, Psaty BM, Furberg CD, Robbins J, Lloyd-Jones DM. Prevalence, prognosis, and implication of isolated minor nonspecific ST-segment and T-wave abnormalities in older adults: Cardiovascular Health Study. Circulation. 2008;118:2790–2796. doi: 10.1161/CIRCULATIONAHA.108.772541. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Rautaharju PM, Kooperberg C, Larson JC, LaCroix A. Electrocardiographic abnormalities that predict coronary heart disease events and mortality in postmenopausal women. The Women’s health Initiative. Circulation. 2006;113:473–480. doi: 10.1161/CIRCULATIONAHA.104.496091. [DOI] [PubMed] [Google Scholar]
- 8.Zhang ZM, Prineas RJ, Eaton CB. Evaluation and comparison of the Minnesota code and Novacode for electrocardiographic Q-ST wave abnormalities for the independent prediction of incident coronary heart disease and total mortality (from the Women’s Health Initiative) Am J Cardiol. 2010;106:18–25. doi: 10.1016/j.amjcard.2010.02.007. [DOI] [PubMed] [Google Scholar]
- 9.Prineas RJ, Crow RS, Blackburn H. The Minnesota Code Manual of Electrocardiographic Findings. John Wright PSB; Boston, MA: 1982. p. 203. [Google Scholar]
- 10.Prineas RJ, Crow RS, Zhang ZM. The Minnesota Code Manual of Electrocardiographic Findings. 2nd Ed Springer; London: 2009. pp. 277–324. [Google Scholar]
- 11.Rautaharju PM, MacInnis PJ, Warren JW, Wolf HK, Rykers PM, Calhoun HP. Methodology of ECG interpretation in the Dalhousie program: NOVACODE ECG classification procedures for clinical trials and population health surveys. Methods Inf Med. 1990;29:362–374. [PubMed] [Google Scholar]
- 12.Rautaharju PM, Park LP, Chaitman BR, Rautaharju F, Zhang ZM. The Novacode criteria for classification of ECG abnormalities and their clinically significant progression and regression. J Electrocardiol. 1998;31:157–187. [PubMed] [Google Scholar]
- 13.Fried LP, Borhani NO, Enright P, Furberg CD, Gardin JM, Kronmal RA, Kuller LH, Manolio TA, Mittelmark MB, Newman A, O’Leary DH, Psaty BM, Rautaharju PM, Tracy RP, Weiler PG. The Cardiovascular Health Study: design and rationale. Ann Epidemiol. 1991;1:263–276. doi: 10.1016/1047-2797(91)90005-w. [DOI] [PubMed] [Google Scholar]
- 14.Furberg CD, Manolio TA, Psaty BM, Bild DE, Borhani NO, Newman A, Tabatznik B, Rautaharju PM. Major ECG abnormalities in persons aged 65 years and older (The Cardiovascular Health Study) Am J Cardiol. 1992;69:1329–1335. doi: 10.1016/0002-9149(92)91231-r. [DOI] [PubMed] [Google Scholar]
- 15.Psaty BM, Kuller LH, Bild D, Burke GL, Kittner SJ, Mittelmark M, Price TR, Rautaharju PM, Robbins J. Methods of assessing prevalent cardiovascular disease in the Cardiovascular Health Study. Ann Epidemiol. 1995;5:270–277. doi: 10.1016/1047-2797(94)00092-8. [DOI] [PubMed] [Google Scholar]
- 16.Ives DG, Fitzpatrick AL, Bild DE, Psaty BM, Kuller LH, Crowley PM, Cruise RG, Theroux S. Surveillance and ascertainment of cardiovascular events: the Cardiovascular Health Study. Ann Epidemiol. 1995;5:278–285. doi: 10.1016/1047-2797(94)00093-9. [DOI] [PubMed] [Google Scholar]