Cardiovascular disease (CVD) is responsible for more

Epidemiology and Prevention Impact of Blood Pressure and Blood Pressure Change During Middle Age on the Remaining Lifetime Risk for Cardiovascular Disease The Cardiovascular Lifetime Risk Pooling Project Norrina Allen, PhD; Jarett D. Berry, MD, MS; Hongyan Ning, MD, MS; Linda Van Horn, PhD, RD; Alan Dyer, PhD; Donald M. Lloyd-Jones, MD, ScM Background Prior estimates of lifetime risk (LTR) for cardiovascular disease (CVD) examined the impact of blood (BP) at the index age and did not account for changes in BP over time. We examined how changes in BP during middle age affect LTR for CVD, coronary heart disease, and stroke. Methods and Results Data from 7 diverse US cohort studies were pooled. Remaining LTRs for CVD, coronary heart disease, and stroke were estimated for white and black men and women with death free of CVD as a competing event. LTRs for CVD by BP strata and by changes in BP over an average of 14 years were estimated. Starting at 55 years of age, we followed up 61 585 men and women for 700 000 person-years. LTR for CVD was 52.5% (95% confidence interval, 51.3 53.7) for men and 39.9% (95% confidence interval, 38.7 41.0) for women. LTR for CVD was higher for blacks and increased with increasing BP at index age. Individuals who maintained or decreased their BP to normal levels had the lowest remaining LTR for CVD, 22% to 41%, compared with individuals who had or developed by 55 years of age, 42% to 69%, suggesting a dose-response effect for the length of time at high BP levels. Conclusions Individuals who experience increases or decreases in BP in middle age have associated higher and lower remaining LTR for CVD. Prevention efforts should continue to emphasize the importance of lowering BP and avoiding or delaying the incidence of to reduce the LTR for CVD. (Circulation. 2012;125:37-44.) Key Words: cardiovascular diseases coronary disease risk factors stroke Cardiovascular disease (CVD) is responsible for more than one third of all deaths in the United States. 1 Although CVD mortality rates have decreased over the past 4 decades, it remains the leading cause of death and one of the top causes of functional disability. 1 Given the burden of CVD, further efforts are needed to improve CVD prevention. Clinical practice guidelines focused on prevention have recommended developing lifetime risk (LTR) estimates as a method for improved long-term prediction for individuals, better estimation of the burden in the population, and more appropriate comparisons of risk between diseases. 2,3 In addition, LTR estimates appear to be particularly useful in education campaigns because they may be more easily understood than traditional epidemiological measures. 3,4 Clinical Perspective on p 44 The LTR for CVD among whites has been estimated to be 1 in 2 for men and 1 in 3 for women. 5 Data from a single cohort with a limited number of blacks suggest that black men and women have LTRs for CVD death at least as high as those for whites of the same sex. 6 In addition, the LTR for CVD increases with increasing risk factor burden. 6,7 Blood (BP) represents one of the most important modifiable risk factors for CVD. The LTR for CVD, particularly stroke, increases dramatically with increasing BP for both blacks and whites such that the remaining LTR for CVD was twice as high among individuals with stage 2 in middle age compared with those with optimal BP levels. 7 Previous studies of the LTR for CVD have been limited to single measures of BP during middle age. 8 However, to be more useful for individualized risk prediction and communication, detailed, risk factor specific LTR estimates must be developed. 4 Systolic BP generally increases linearly with age, and individuals with higher systolic BP experience the largest increases. 9 However, little is known about how changes in BP Received October 15, 2010; accepted November 11, 2011. From the Department of Preventive Medicine (N.A., H.N., L.V.H., A.D., D.M.L.-J.) and Bluhm Cardiovascular Institute, Department of Medicine (D.M.L.-J.), Northwestern University Feinberg School of Medicine, Chicago, IL, and UT Southwestern Medical Center, Department of Medicine, Division of Cardiology, Dallas, TX (J.D.B.). The online-only Data Supplement is available with this article at http://circ.ahajournals.org/lookup/suppl/doi:10.1161/circulationaha. 110.002774/-/DC1. Correspondence to Donald M. Lloyd-Jones, MD, ScM, Department of Preventive Medicine and Division of Cardiology, Feinberg School of Medicine, Northwestern University, 680 N Lake Shore Dr, Ste 1102, Chicago, IL 60611. E-mail dlj@northwestern.edu 2011 American Heart Association, Inc. Circulation is available at http://circ.ahajournals.org DOI: 10.1161/CIRCULATIONAHA.110.002774 37

38 Circulation January 3/10, 2012 during middle age may affect the LTR for CVD. Accordingly, we examined the impact of BP at specific ages and BP change over 10 years on the LTR for all CVD and for fatal and nonfatal coronary heart disease (CHD) and stroke among black and white men and women in the Lifetime Risk Pooling Project. Methods Study Sample Data for this study came from the Cardiovascular Lifetime Risk Pooling Project. Detailed methods for the Pooling Project have previously been published. 7 In brief, data from 17 US epidemiological cohort studies were pooled to create the Cardiovascular Lifetime Risk Pooling Project. All cohorts had to meet several criteria: (1) community- or population-based sampling or large volunteer cohort; (2) availability of at least 1 baseline examination at which participants provided demographic, personal, and medical history information and underwent direct measurement of physiological or anthropometric variables (eg, BP, weight); (3) longitudinal follow-up of at least 10 years with complete or near-complete ascertainment of vital status; and (4) availability of cause-specific or cardiovascular mortality data with or without ascertainment of nonfatal cardiovascular events. Because one of the chief aims of the Cardiovascular Lifetime Risk Pooling Project is to compare LTRs between different race/ethnic groups, we specifically sought to obtain data from cohort studies that included nonwhite race/ethnic groups. We did not include cohorts of individuals that participated in clinical trials, had only self-report of risk factors for important covariates without direct physical examination by study investigators, or had few deaths or events. Data sets that included all cardiovascular outcomes, including fatal and nonfatal stroke, were included in this study. These cohorts include the Atherosclerosis Risk in Communities (ARIC) Study, Framingham Heart Study (FHS), Framingham Offspring Study (FOS), Honolulu Heart Program (HHP), Kaiser Permanente Study of the Oldest Old, Cardiovascular Health Study (CHS), and Women s Health Initiative (WHI) Observational Study. Limited-access data sets were obtained from the National Heart, Lung, and Blood Institute (NHLBI) for the ARIC Study, 10 FHS, 11 FOS, 12 and HHP. 13,14 Data from the Kaiser Permanente Study of the Oldest Old 15 were obtained from the Inter-University Consortium for Political and Social Research at the University of Michigan (http://www.icpsr.umich.edu/icpsr/index.html). Complete data sets were obtained directly by agreement from the CHS 16 and WHI Observational Study 17 investigators. After all of the data were obtained, variables of interest from each data set were cleaned and renamed with the use of a standardized protocol to allow ease of use in the pooling project analyses. Given that data were obtained from cohort studies spanning 6 decades from the 1940s to the 2000s, there were differences between studies in measurement techniques. Whenever possible, values obtained with the most similar techniques were used for analyses. All data were appropriately deidentified, and all study protocols and procedures were approved by the IRB at Northwestern University. BP Measures For all cohorts except the Kaiser Permanente Study, BP was recorded as the average of 2 or 3 seated measurements taken by trained study personnel on manual mercury sphygmomanometers (Table I in the online-only Data Supplement). BP data for participants of the Kaiser Permanente Study of the Oldest Old were taken from clinical measurements recorded in the participants medical record. Participants were grouped according to their index BP as measured within 5 years of each index age. For example, BPs measured in individuals between 40 and 50 years of age were included in the analyses for the index age of 45 years. According to the Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure 18 criteria, the average of at least 2 BP measurements was categorized as follows: normal, 120/ 80 mm Hg untreated; pre, systolic BP of 120 to 139 mm Hg or diastolic BP of 80 to 89 mm Hg untreated with medication; stage 1, systolic BP of 140 to 159 mm Hg or diastolic BP of 90 to 99 mm Hg untreated with medication; and stage 2 or treated, systolic BP 160 mm Hg, diastolic BP 100 mm Hg, or treated with antihypertensive medication. The prevalence of treatment in this study is low because of the time period during which these cohorts were initiated. In addition, although the risk for CVD is reduced for hypertensive individuals who receive antihypertensive medication, their risk remains higher than that for individuals with untreated, ideal BP levels. 19 Therefore, to be conservative, we included treated individuals in the highest-risk group. BP changes were examined among individuals who had at least 2 separate BP measurements 10 years apart. The baseline BP measurement had to have been after 20 years of age and at least 10 years before the BP measurement at the index age, as defined above. On average, baseline BP measurements were recorded 14 years before the index age; ie, on average, baseline BPs were taken at 41 years of age for the index age of 55 years. To classify an individual s change in BP, we collapsed BP into 3 categories: normal, 120/ 80 mm Hg untreated; pre, untreated systolic BP of 120 to 139 mm Hg or diastolic BP of 80 to 89 mm Hg; or, systolic BP 140 mm Hg, diastolic BP 90 mm Hg, or treated. BP change was categorized into 9 categories based on the change from the first to second measurement. Event Ascertainment and Definition For all cohorts except the Kaiser Permanente Study of the Oldest Old, CVD was defined as fatal CHD, hospitalized myocardial infarction, nonhospitalized myocardial infarction, and fatal and nonfatal stroke by trained physician-adjudicators using all available medical records. For death events, many cohorts used linkage to the National Death Index for underlying cause of death from death certificate data, whereas others used adjudicated cause of death by study investigators after review of all available medical records and/or autopsy data. For the present analyses, deaths resulting from CVD (as adjudicated or indicated by International Classification of Disease, 8th revision [ICD-8], ICD-9 codes 390 458, and ICD-10 codes I00-I99) or CHD (as adjudicated or indicated by ICD-8 and ICD-9 codes 410 414 and ICD-10 codes I20-I25) were included. Nonfatal events of interest, including myocardial infarction and stroke, were obtained only from studies that adjudicated events with 1 exception (the Kaiser Permanente Study). Criteria for adjudication of myocardial infarction and stroke have been published for these cohorts: ARIC, 10 CHS, 16 FHS and FOS, 5,8 HHP, 20 and WHI. 17 Diagnosis of myocardial infarction generally required at least two of the following criteria for diagnosis: typical chest discomfort, evolution of ECG abnormalities consistent with myocardial infarction, and serological evidence of myocardial necrosis. For most cohorts, stroke was diagnosed as a persistent central neurological deficit(s) lasting 24 hours and unexplained by other causes, as adjudicated by trained physician adjudicators using standard criteria. 10 17 In the Kaiser Permanente Study, 15 incident stroke events were ascertained from hospitalization discharge codes. For this cohort, we counted only the first hospitalization for which stroke (ICD-8 and ICD-9 codes 430 438 and ICD-10 codes I60-I69) was the first-listed discharge diagnosis. Statistical Analyses LTR was calculated by a modified technique of survival analyses 21 22 in which participants contribute information on the incidences of CVD, CHD, and stroke for each age they attain during follow-up. This approach has been used in numerous prior studies of LTR. 23 24 Each subject in the study was followed up from study entry until the occurrence of a first CVD event, death, or age 95. Remaining LTR estimates for CVD, CHD, and stroke were stratified by BP level and were calculated separately for black and white men and women beginning at 45, 55, 65, and 75 years of age. LTR estimates for CVD, CHD, and stroke were also stratified by change in BP category for the index age of 55 years. Because of the limited number of blacks and overall similar patterns of association with BP, blacks and whites were combined in estimates of LTR by change in

Allen et al Changes in Blood Pressure and Lifetime Risk for CVD 39 Table 1. The 10-, 20-, 30-, and 40- Year Cumulative Risks for Cardiovascular Disease Adjusted for Competing Risk Starting at the Index Age of 55 Years by Sex and Race Cumulative Adjusted Risk (95% Confidence Interval) Specified Follow-Up Starting at Index Age of 55 y Men Women Whites Normal blood 8.7 (7.6 9.8) 21.3 (18.9 23.6) 39.8 (35.9 43.6)... 2.3 (2.1 2.6) 9.2 (7.9 10.5) 27.8 (24.9 30.7) 46.1 (4.3 50.8) Pre 9.6 (8.8 10.4) 27.7 (26.4 29.0) 50.8 (49.1 52.5) 61.8 (59.7 64.0) 3.3 (3.0 3.6) 13.9 (13.0 14.8) 38.1 (36.4 39.6) 56.2 (53.9 58.6) Stage 1 12.4 (11.0 13.8) 32.7 (30.7 34.7) 57.1 (54.7 59.4)... 5.2 (4.4 6.0) 19.6 (17.9 21.3) 45.9 (43.4 48.3) 64.7 (61.3 68.1) Stage 2 15.5 (13.0 18.0) 38.8 (35.4 42.2) 59.3 (55.6 62.9)... 9.8 (7.7 11.9) 30.5 (27.1 33.9) 53.7 (49.6 57.8)... Blacks Normal blood 11.7 (7.5 15.8) 23.5 (15.2 31.8)...... 5.2 (3.8 6.5) 19.0 (11.3 26.6)...... Pre 14.3 (11.6 16.9) 29.9 (24.9 34.8) 49.3 (39.8 58.8)... 8.3 (7.1 9.5) 15.2 (12.5 18.0) 35.3 (25.8 44.8)... Stage 1 16.1 (11.5 20.7) 38.4 (31.0 45.7) 53.3 (44.1 62.6)... 16.1 (13.0 19.2) 34.8 (26.5 43.0)...... Stage 2 20.8 (13.8 27.8) 41.2 (31.9 50.4)...... 20.6 (15.0 26.2) 37.1 (28.1 46.1)......... Indicates limited person-years of follow-up, precluding a stable estimate. BP. LTRs to 85 years of age by strata of change in BP were estimated. Given the uncertainty associated with small numbers, LTR estimates are reported only when the adjusted person-time exceeded 100 person-years at the extremes of age. All statistical analyses were performed with SAS statistical software version 9.1 (SAS Institute, Inc, Cary, NC). Results Study Sample and Remaining LTR for CVD, CHD, and Stroke A total of 61 585 men and women were followed up from 55 years of age and contributed 695 394 person-years of followup. The numbers of participants, events, and person-years of follow-up for stroke, CHD, and CVD at the index ages of 45, 55, 65, and 75 years are shown in Table II in the online-only Data Supplement. Given the large number of participants and years of follow-up at the index age of 55 years and the relatively few events that occurred before this age, we focused our findings on the remaining LTR starting at the index age of 55 years. Overall, starting at the age of 55 years, the remaining LTR for CVD was 52.5% (95% confidence interval [CI], 51.3 53.7) for men and 39.9% (95% CI, 38.7 41.0) for women, for CHD was 30.9% (95% CI, 29.8 31.9) for men and 17.5% (95% CI, 16.6 18.3) for women, and for stroke was 11.2% (95% CI, 10.3 12.1) for men and 14.7% (95% CI, 13.6 15.8) for women. Remaining LTR by BP at Index Age When participants were categorized according to their BP at the index age of 55 years, 25.7% of men and 40.8% of women had normal BPs, 49.4% of men and 47.5% of women had pre, 18.1% of men and 9.6% of women had stage 1, and 6.8% of men and 2.2% of women had stage 2 or treated. The LTR for CVD, CHD, and stroke increased with increasing BP category for both men and women (Tables 1 3, the Figure, and Table IV in the online-only Data Supplement). LTR for CVD by BP was similar for the index age of 45 years (Figure I in the online-only Data Supplement). As shown in Table 4, the higher the BP level was, the younger the age was at which participants reached a LTR for CVD, CHD, and stroke of 8% (a measure of early burden and approximately half of the remaining LTR to 85 years of age). In general, black women had the greatest age difference (9 21 years) between when the lowest and highest BP strata reached an LTR of 8%; the difference was largest for stroke. The age at which blacks reached a higher LTR was younger than for whites at almost all BP levels. These patterns were similar when cohorts were examined individually and when results were examined with and without the relatively large number of WHI participants (Table IV in the online-only Data Supplement). BP Changes During Middle Age More than half of all men and women remained in consistent BP strata from 41 through 55 years of age (Table III in the online-only Data Supplement). Almost 20% of men experienced decreases in their BP and 30% experienced increases before 55 years of age. In contrast, almost 40% of women experienced an increase in BP and only 10% experienced a decrease before 55 years of age. At baseline, men tended to have higher systolic BPs; however, by the index age of 55 years, systolic BPs were higher among women owing to larger increases in BP observed among women during middle age. Individuals who became hypertensive during middle age experienced the largest increases in systolic BP with an increase of 32 mm Hg. Women had lower diastolic BPs in all categories. Changes in diastolic BP over middle age were similar for both men and women. Participants who experienced decreases in BP during middle age had smaller increases in total cholesterol and body mass index compared with participants who maintained their BP or experienced increasing BP.

40 Circulation January 3/10, 2012 Table 2. The 10-, 20-, 30-, and 40-Year Cumulative Risks for Coronary Heart Disease Adjusted for Competing Risk Starting at the Index Age of 55 Years by Sex and Race Cumulative Adjusted Risk (95% Confidence Interval) Specified Follow-Up Starting at Index Age of 55 y Men Women Whites Normal blood 5.9 (4.9 6.8) 13.6 (11.7 15.5) 22.9 (19.9 26.0) 29.0 (24.8 33.1) 1.1 (0.9 1.3) 4.3 (3.4 5.2) 11.0 (9.2 12.9) 18.3 (15.0 21.6) Pre 7.0 (6.3 7.6) 18.0 (16.9 19.1) 30.1 (28.6 31.5) 35.4 (33.6 37.2) 1.7 (1.5 1.9) 6.8 (6.2 7.5) 16.3 (15.1 17.4) 23.4 (21.7 25.1) Stage 1 9.0 (7.8 10.2) 21.2 (19.5 23.0) 34.8 (32.7 37.0) 39.6 (37.0 42.2) 2.7 (2.2 3.3) 9.7 (8.4 10.9) 22.1 (20.2 24.1) 28.8 (26.4 31.3) Stage 2 10.6 (8.5 12.7) 25.4 (22.2 28.1) 37.5 (34.1 41.0)... 4.2 (2.8 5.6) 14.7 (12.2 17.3) 26.6 (23.2 30.1)... Blacks Normal blood 5.9 (2.9 8.9) 10.9 (4.6 17.3)...... 1.8 (1.0 2.6) 4.8 (0.5 9.0)...... Pre 6.7 (4.8 8.6) 13.0 (9.6 16.4) 25.7 (17.5 34.0)... 2.9 (2.2 3.7) 4.9 (3.4 6.4) 14.8 (8.1 21.5)... Stage 1 7.3 (4.2 10.5) 17.3 (11.9 22.8) 25.4 (18.0 32.9)... 5.4 (3.6 7.3) 12.2 (6.8 17.7)...... Stage 2 7.6 (3.0 12.3) 18.0 (11.0 25.0)...... 8.8 (5.0 12.6) 13.3 (7.6 19.0)......... Indicates limited person-years of follow-up, precluding a stable estimate. Change in BP and Remaining LTR for Stroke Men and women who consistently had BPs 120/ 80 mm Hg and those whose BP decreased to 120/ 80 mm Hg over time tended to have the lowest remaining LTR for CVD, CHD, and stroke (Table 5). The remaining LTRs were similar for individuals whose BP was consistently 120/ 80 mm Hg and those whose BP decreased to this level. Individuals whose BP increased over time before 55 years of age had higher LTRs compared with those whose BP remained stable or decreased before 55 years of age. Although the findings were similar for men and women, men who had throughout middle life experienced the highest remaining LTR for stroke, whereas men whose BP increased to hypertensive levels had the highest LTRs for CVD and CHD. In contrast, women whose BP increased to hypertensive levels by 55 years of age had the highest LTR for stroke, but those who consistently had hypertensive levels before 55 years of age had the highest LTRs for CVD and CHD. Discussion Among the cohorts included in the Lifetime Risk Pooling Project, the remaining LTR for CVD was 52.5% in men and Table 3. The 10-, 20-, 30-, and 40-Year Cumulative Risks for Stroke Adjusted for Competing Risk Starting at the Index Age of 55 Years by Sex and Race Cumulative Adjusted Risk (95% Confidence Interval) Specified Follow-Up Starting at Index Age of 55 y Men Women Whites Normal blood 1.2 (0.8 1.7) 3.4 (2.4 4.4) 7.7 (5.7 9.7) 10.8 (7.8 13.8) 0.5 (0.4 0.7) 2.4 (1.6 3.1) 7.3 (5.6 8.9) 12.8 (9.8 15.7) Pre 1.7 (1.4 2.1) 5.3 (4.6 6.0) 9.7 (8.8 10.7) 11.4 (10.2 12.6) 0.8 (0.7 1.0) 3.1 (2.6 3.5) 8.3 (7.4 9.2) 14.1 (12.6 15.7) Stage 1 2.4 (1.7 3.1) 5.4 (4.4 6.4) 9.6 (8.2 10.9) 11.5 (9.6 13.4) 1.7 (1.2 2.2) 5.5 (4.5 6.5) 11.1 (9.1 12.5) 16.7 (14.4 19.0) Stage 2 3.5 (2.2 4.9) 6.8 (5.0 8.6) 9.1 (7.0 11.2)... 4.1 (2.7 5.6) 9.9 (7.6 12.1) 15.8 (12.9 18.7)... Blacks Normal blood 3.7 (0.9 6.4)......... 1.5 (0.7 2.3) 7.7 (1.4 14.1)...... Pre 5.8 (3.6 7.9) 8.2 (5.4 10.9) 8.2 (5.4 10.9)... 2.8 (2.0 3.5) 4.5 (3.2 5.8) 9.0 (3.8 14.2)... Stage 1 8.3 (4.4 12.3) 13.8 (8.2 19.3) 15.0 (9.1 21.0)... 5.7 (3.6 7.9) 7.3 (4.7 10.0)...... Stage 2 8.5 (2.8 14.1) 12.1 (5.4 18.9)...... 9.1 (4.1 14.1) 13.2 (6.5 20.0)......... Indicates limited person-years of follow-up, precluding a stable estimate.

Allen et al Changes in Blood Pressure and Lifetime Risk for CVD 41 Figure. Cumulative lifetime risk (percent) of cardiovascular disease (CVD), coronary heart disease (CHD), and stroke adjusted for competing risk of non-cvd death by blood (BP) category for men (A) and women (B). 39.9% in women. The LTR for CVD was higher among blacks compared with whites of the same sex and increased with increasing BP at the index age. Changes in BP during middle age appear to have a substantial impact on the LTR for stroke. Almost 70% of all men who develop in middle age will experience a CVD event by 85 years of age. Among women, those who have from 41 through 55 years of age have the greatest LTR for CVD at 49.4%. Individuals who are able to maintain a normal BP or to decrease their BP to normal levels during middle age have Table 4. Age at Which Cumulative Lifetime Risk for Cardiovascular Disease, Coronary Heart Disease, and Stroke Exceeds 8% by Blood Pressure Category at the Index Age of 55 Years Blood Pressure Category Normal Pre Stage 1 Hypertension Stage 2 Hypertension Cardiovascular disease 8% lifetime risk, y White men 63 63 62 60 Black men 63 60 59 59 White women 73 71 67 63 Black women 67 64 60 58 Coronary heart disease 8% lifetime risk, y White men 68 66 63 62 Black men 68 66 65 65 White women 80 76 73 69 Black women... 81 72 63 Stroke 8% lifetime risk, y White men 86 82 82 79 Black men... 69 65 64 White women 87 85 80 71 Black women... 85... 64... Indicates limited person-years of follow-up, precluding a stable estimate.

42 Circulation January 3/10, 2012 Table 5. Cumulative Lifetime Risk for Fatal/Nonfatal Stroke, Coronary Heart Disease, and Cardiovascular Disease by Blood Pressure Categories at Baseline and Index Ages BP Category at Baseline ( 10 y Before Remaining Lifetime Risk (95% CI) 85 Years of Age Index Age) BP Category at Index Age, mm Hg Stroke CHD CVD Men 120/ 80 120/ 80 12.5 (6.4 18.6) 24.8 (16.9 32.7) 41.0 (31.9 50.2) 120 139/80 89 19.8 (11.3 28.3) 33.6 (24.2 42.9) 52.2 (41.3 63.1) 140/ 90/Rx 23.5 (8.0 39.0) 30.5 (16.3 44.7) 69.0 (47.5 90.6) 120 139/80 89 120/ 80 10.1 (5.1 15.2) 24.6 (17.4 31.8) 34.5 (26.3 42.6) 120 139/80 89 18.4 (14.1 22.7) 34.7 (25.3 35.3) 50.7 (44.9 56.5) 140/ 90/Rx 19.2 (14.1 24.4) 45.0 (38.3 51.7) 65.3 (58.4 72.3) 140/ 90/Rx 120/ 80......... 120 139/80 89 22.6 (15.1 30.0) 32.3 (23.9 40.8) 59.4 (49.9 69.0) 140/ 90/Rx 23.6 (19.2 27.9) 43.3 (38.1 48.5) 65.3 (59.9 70.6) Women 120/ 80 120/ 80 10.8 (7.0 14.7) 5.8 (3.1 8.5) 21.8 (16.9 26.8) 120 139/80 89 13.6 (9.1 18.1) 14.6 (10.1 19.1) 34.1 (27.8 40.5) 140/ 90/Rx 26.0 (16.7 35.4) 13.5 (6.9 20.0) 42.2 (30.0 54.4) 120 139/80 89 120/ 80 10.2 (5.3 15.1) 12.2 (7.1 17.3) 24.7 (17.8 31.6) 120 139/80 89 17.6 (13.5 21.6) 16.6 (12.7 20.5) 38.9 (33.6 44.2) 140/ 90/Rx 22.4 (17.2 27.6) 21.7 (16.8 26.7) 47.2 (40.8 53.6) 140/ 90/Rx 120/ 80......... 120 139/80 89 17.8 (9.6 26.0) 15.8 (8.3 23.3) 36.5 (26.2 46.7) 140/ 90/Rx 19.4 (15.1 23.7) 24.9 (20.3 29.5) 49.4 (43.8 62.6) BP indicates blood ; CI, confidence interval; CHD, coronary heart disease; CVD, cardiovascular disease; Rx; treatment; and..., limited person-years of follow-up, precluding a stable estimate. significantly lower LTRs for CVD, 35% to 40% and 22% to 25% for men and women, respectively. Among the white participants of the FHS (40 59 years of age), previous estimates for the LTR of CVD, 51.7% for men and 39.2% for women, were similar to those in this study, 52.5% and 39.9%, respectively. 25 LTR estimates for CHD were lower in this study compared with previous estimates from the FHS, 30.9% versus 40.8% for men and 17.5% versus 24.6% in women. 5 Similarly, LTR estimates for stroke were slightly lower in this study compared with previous estimates from the FHS, 11.2% versus 14.5% for men and 14.7% versus 18.1% in women, although participants in the FHS were followed up to 106 years of age compared with 95 years of age in this study. 8 Birth-cohort effects may explain the lower LTRs for stroke observed in this study because FHS participants were recruited into the original FHS cohort in 1948 compared with some of the more contemporary cohorts included in the Lifetime Risk Pooling Project. Significant declines in BP levels have occurred over the past 50 years as a result of improvements in population levels of untreated BP and increased treatment and control of. 26 28 Even within the FHS, Carandang et al 29 found a trend toward decreased LTR of stroke from 1950 to 2004. Although decreasing population BP levels may have resulted in slightly lower overall LTR estimates than those previously published by the FHS, analyses within the Lifetime Risk Pooling Project have demonstrated that, regardless of changes in prevalence, the association between BP and LTR for CVD remains the same over time within strata. 7 Prior estimates for the LTR for CVD have been based on single BP measurements taken at or near the index age. However, individuals are exposed to different BP levels throughout their life; thus, a more detailed examination of the effect of BP exposure is warranted. Systolic BP tends to increase linearly throughout middle age, although there is variability in the slope of the increase. 9 Individuals with the highest levels of systolic BP at baseline appear to have the largest age-related linear increases in systolic BP. 9 Above and beyond current BP, antecedent BP has been found to be an important determinant in risk for CVD. 30 Specifically examining stroke risk, studies done among Japanese populations have found that, in addition to current BP, BP increases over the prior 20 years represent an independent risk factor for stroke. 31 Individuals who were found to have at multiple time points had the greatest risk for stroke. 32 The present study is the first to examine how changes in BP throughout middle age (average change from 41 to 55 years of age) affect subsequent LTRs for CVD, CHD, and stroke among a large, diverse US population. We found that individuals who maintained or reduced their BP to normal levels by 55 years of age had the lowest LTRs. Decreases in BP may have been due to lifestyle changes, as suggested by the changes in body mass index and total cholesterol, although it is possible that differences were due to random variation or regression to the mean. The highest LTR for CVD among men was among those who developed during middle age. In contrast, among women, individuals who had from 41 through 55 years of age had the highest LTR for CVD. Although it was not consistently evident that a longer

Allen et al Changes in Blood Pressure and Lifetime Risk for CVD 43 duration of high BP was associated with increased LTRs, it is possible that a greater duration of is associated with increased competing risk from non-cvd mortality. Our findings suggest the hypothesis that there may be a dose-response effect for the length of time at higher BPs. Buck et al 33 found that the cardiovascular risk for hypertensive patients was highest for those who developed at younger ages and declined as age of onset increased from 40 to 69 years of age. Similarly, patterns in our findings suggest that women who had developed by early middle age (mean age, 41 years) had a higher LTR for CVD and CHD compared with those who had developed later in middle age, at 55 years of age. A similar pattern was seen among men when examining the remaining LTR of stroke. This study has several strengths. First, we examined the LTR for CVD among a large, diverse sample of populationbased cohorts. These analyses included almost 700 000 person-years of follow-up. Many of the cohorts included in the Lifetime Risk Pooling Project are more contemporary than those used to derive prior estimates of the LTR for stroke, thereby providing more current LTR estimates. Importantly longitudinal BP measurements were available, allowing us to examine how BP changes over an average of 14 years during middle age affect the LTR for CVD. However, some individuals may not have had a second BP available because of death or, more likely, the cohort structure and timing of visits. These individuals therefore would not have been included in our analyses of BP change. Despite pooling, there was limited follow-up for blacks; therefore, we had to combine races in our analyses on change in BP and were unable to determine racial differences in the association between BP changes and LTR for CVD. By examining stratified estimates, we have adjusted for any confounding by age, race, sex, and BP; however, the methods used to determine LTR do not allow for adjustment, so there may be potential confounding by other factors such as socioeconomic status, body mass index, and other behavioral factors that could not be included in the model. In addition, we pooled multiple population-based cohorts, and there is a potential for heterogeneity resulting from birth-cohort effects and/or secular trends in the LTR estimates. Although there have been secular trends in the CVD event rates and treatment patterns, 29 most of the follow-up we have is before the widespread use of antihypertensives and lipid-lowering therapies. In the Lifetime Risk Pooling Project, the prevalence of risk factors has changed over birth cohorts; however, the relationships between risk factors and outcomes have remained remarkably consistent across cohorts and birth cohorts in the Lifetime Risk Pooling Project. In addition, changes in an individual s treatment status during follow-up may have affected that person s remaining LTR for CVD. However, the data we used to estimate LTRs for stroke are similar to what would be available to patients and clinicians at a given index age, without knowledge of subsequent changes in treatment strategy. Given the limited follow-up duration, it is not possible to estimate LTRs using only data from cohorts recruited in the last decade, when antihypertensive use has become prevalent. Our results suggest that there may be a dose-response effect for the number of years with high BP on the LTR of CVD, CHD, and stroke. Individuals who are able to maintain or to decrease their BP to normal BP levels during middle age have the lowest LTR for CVD. In contrast, individuals who experience an increase in BP have higher LTRs for CVD. Taking BP changes into account can provide more accurate estimates for LTR for CVD and represent a step forward in developing individualized risk prediction strategies. Therefore, avoiding before middle age and delaying the onset of the development of both appear to have a significant impact on an individual s remaining LTR for CVD. Acknowledgments The investigators of the Cardiovascular Lifetime Risk Pooling Project wish to extend their deepest gratitude to the investigators of all of the included cohort studies for their hard work and dedication in collecting the underlying data and especially to the participants in the cohorts, whose dedication and commitment have formed the basis of profound observations on health and disease that have contributed to improved health, longevity, and quality of life for millions of individuals. Sources of Funding Dr Lloyd-Jones 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. Dr Lloyd-Jones and this work are supported in part by grant R21 HL085375 from the NHLBI. As part of his research training, Dr Allen has received support from a Ruth Kirschstein NRSA/NHLBI fellowship at Northwestern University Feinberg School of Medicine (T32 HL 069771 07). The ARIC Study, FHS, FOS, and HHP are conducted and supported by the NHLBI in collaboration with the study investigators. This manuscript was prepared using limited-access data sets obtained from the NHLBI for these studies and does not necessarily reflect the opinions or views of the study investigators or the NHLBI. The research reported in this article from the CHS was supported by contracts N01-HC-85079 through N01-HC-85086, N01-HC-35129, N01 HC-15103, N01 HC-55222, N01-HC-75150, and N01-HC-45133 and grant U01 HL080295 from the NHLBI, with additional contribution from the National Institute of Neurological Disorders and Stroke. A full list of principal CHS investigators and institutions can be found at http://www.chs-nhlbi.org/pi.htm. The WHI program is funded by the NHLBI, National Institutes of Health, US Department of Health and Human Services through contracts N01WH22110, 24152, 32100 2, 32105 6, 32108 9, 32111 13, 32115, 32118 32119, 32122, 42107 26, 42129 32, and 44221. We thank the WHI investigators and staff for their dedication and the study participants for making the program possible. A full listing of WHI investigators can be found at http://www.whiscience.org/publications/whi_investigators_shortlist.pdf. None. Disclosures References 1. Roger VL, Go AS, Lloyd-Jones DM, Adams RJ, Berry JD, Brown TM, Carnethon MR, Dai S, de Simone G, Ford ES, Fox CS, Fullerton HJ, Gillespie C, Greenlund KJ, Hailpern SM, Heit JA, Ho PM, Howard VJ, Kissela BM, Kittner SJ, Lackland DT, Lichtman JH, Lisabeth LD, Makuc DM, Marcus GM, Marelli A, Matchar DB, McDermott MM, Meigs JB, Moy CS, Mozaffarian D, Mussolino ME, Nichol G, Paynter NP, Rosamond WD, Sorlie PD, Stafford RS, Turan TN, Turner MB, Wong ND, Wylie-Rosett J. Heart disease and stroke statistics 2011 update: a report from the American Heart Association. Circulation. 2011;123: e18 e209. 2. Mosca L, Banka CL, Benjamin EJ, Berra K, Bushnell C, Dolor RJ, Ganiats TG, Gomes AS, Gornik HL, Gracia C, Gulati M, Haan CK, Judelson DR, Keenan N, Kelepouris E, Michos ED, Newby LK, Oparil S, Ouyang P, Oz MC, Petitti D, Pinn VW, Redberg RF, Scott R, Sherif K, Smith SC Jr, Sopko G, Steinhorn RH, Stone NJ, Taubert KA, Todd BA, Urbina E, Wenger NK. Evidence-based guidelines for cardiovas-

44 Circulation January 3/10, 2012 cular disease prevention in women: 2007 update. Circulation. 2007; 115:1481 1501. 3. Wilkins J, Lloyd-Jones D. Short-term versus lifetime risk assessment for cardiovascular disease: pros, cons, and clinical implications. Curr Cardiovasc Risk Rep. 2009;3:144 149. 4. Seshadri S, Wolf PA. Lifetime risk of stroke and dementia: current concepts, and estimates from the Framingham study. Lancet Neurol. 2007;6:1106 1114. 5. Lloyd-Jones DM, Larson MG, Beiser A, Levy D. Lifetime risk of developing coronary heart disease. Lancet. 1999;353:89 92. 6. Lloyd-Jones DM, Dyer AR, Wang R, Daviglus ML, Greenland P. Risk factor burden in middle age and lifetime risks for cardiovascular and non-cardiovascular death (Chicago Heart Association Detection Project in Industry). Am J Cardiol. 2007;99:535 540. 7. Berry JD, Dyer A, Cai X, Garside DB, Ning H, Thomas A, Greenland P, Horn LV, Tracy RP, Lloyd-Jones DM. Lifetime risks for cardiovascular disease by age and risk factor burden. N Engl J Med. In Press. 8. Seshadri S, Beiser A, Kelly-Hayes M, Kase CS, Au R, Kannel WB, Wolf PA. The lifetime risk of stroke: estimates from the Framingham study. Stroke. 2006;37:345 350. 9. Franklin SS, Gustin W 4th, Wong ND, Larson MG, Weber MA, Kannel WB, Levy D. Hemodynamic patterns of age-related changes in blood : the Framingham Heart Study. Circulation. 1997;96:308 315. 10. ARIC Investigators. The Atherosclerosis Risk in Communities (ARIC) study: design and objectives. Am J Epidemiol. 1989;129:687 702. 11. Dawber TR, Kannel WB, Lyell LP. An approach to longitudinal studies in a community: the Framingham study. Ann N Y Acad Sci. 1963;107: 539 556. 12. Kannel WB, Manning F, McNamara P, Garrison RJ, Castelli WP. An investigation of coronary heart disease in families: the Framingham Offspring Study. Am J Epidemiol. 1979;110:281 290. 13. Kagan A, Harris BR, Winkelstein W Jr, Johnson KG, Kato H, Syme SL, Rhoads GG, Gay ML, Nichaman MZ, Hamilton HB, Tillotson J. Epidemiologic studies of coronary heart disease and stroke in Japanese men living in Japan, Hawaii and California: demographic, physical, dietary and biochemical characteristics. J Chronic Dis. 1974;27:345 364. 14. Worth RM, Kagan A. Ascertainment of men of Japanese ancestry in Hawaii through World War II Selective Service registration. J Chronic Dis. 1970;23:389 397. 15. Haan MN, Selby JV, Rice DP, Quesenberry CP Jr, Schofield KA, Liu J, Fireman BH. Trends in cardiovascular disease incidence and survival in the elderly. Ann Epidemiol. 1996;6:348 356. 16. Fried LP, Borhani NO, Enright P, Furberg CD, Gardin JM, Kronmal RA, Kuller LH, Manolio TA, Mittelmark MB, Newman A, et al. The Cardiovascular Health Study: design and rationale. Ann Epidemiol. 1991;1: 263 276. 17. Women s Health Initiative Study Group. Design of the Women s Health Initiative clinical trial and observational study. Control Clin Trials. 1998; 19:61 109. 18. Chobanian AV, Bakris GL, Black HR, Cushman WC, Green LA, Izzo JL Jr, Jones DW, Materson BJ, Oparil S, Wright JT Jr, Roccella EJ. The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure: the JNC 7 report. JAMA. 2003;289:2560 2572. 19. Lloyd-Jones DM, Evans JC, Levy D. Hypertension in adults across the age spectrum: current outcomes and control in the community. JAMA. 2005;294:466 472. 20. Yano K, Grove JS, Reed DM, Chun HM. Determinants of the prognosis after a first myocardial infarction in a migrant Japanese population: the Honolulu Heart Program. Circulation. 1993;88:2582 2595. 21. Beiser A, D Agostino RB Sr, Seshadri S, Sullivan LM, Wolf PA. Computing estimates of incidence, including lifetime risk: Alzheimer s disease in the Framingham Study: the Practical Incidence Estimators (PIE) macro. Stat Med. 2000;19:1495 1522. 22. Gaynor J, Feuer E, Tan C, Wu D, Little C, Straus D, Clarkson B, Brennan M. On the use of cause-specific failure and conditional failure probabilities. J Am Stat Assoc. 1993;88:402 409. 23. Berry JD, Garside DB, Cai X, Dyer AR, Lloyd-Jones DM. Remaining lifetime risks for cardiovascular disease death by risk factor burden at selected ages in black and white men and women [abstract]. Circulation. 2007;116:II-832a. 24. Lloyd-Jones DM, Larson MG, Leip EP, Beiser A, D Agostino RB, Kannel WB, Murabito JM, Vasan RS, Benjamin EJ, Levy D. Lifetime risk for developing congestive heart failure: the Framingham Heart Study. Circulation. 2002;106:3068 3072. 25. Lloyd-Jones DM, Leip EP, Larson MG, D Agostino RB, Beiser A, Wilson PW, Wolf PA, Levy D. Prediction of lifetime risk for cardiovascular disease by risk factor burden at 50 years of age. Circulation. 2006;113:791 798. 26. Luepker RV, Arnett DK, Jacobs DR Jr, Duval SJ, Folsom AR, Armstrong C, Blackburn H. Trends in blood, control, and stroke mortality: the Minnesota Heart Survey. Am J Med. 2006;119: 42 49. 27. Hajjar I, Kotchen TA. Trends in prevalence, awareness, treatment, and control of in the United States, 1988 2000. JAMA. 2003; 290:199 206. 28. Kumanyika SK, Landis JR, Matthews-Cook YL, Almy SL, Boehmer SJ. Systolic blood trends in US adults between 1960 and 1980: influence of antihypertensive drug therapy. Am J Epidemiol. 1998;148: 528 538. 29. Carandang R, Seshadri S, Beiser A, Kelly-Hayes M, Kase CS, Kannel WB, Wolf PA. Trends in incidence, lifetime risk, severity, and 30-day mortality of stroke over the past 50 years. JAMA. 2006;296:2939 2946. 30. Vasan RS, Massaro JM, Wilson PW, Seshadri S, Wolf PA, Levy D, D Agostino RB. Antecedent blood and risk of cardiovascular disease: the Framingham Heart Study. Circulation. 2002;105:48 53. 31. Shimizu Y, Kato H, Lin CH, Kodama K, Peterson AV, Prentice RL. Relationship between longitudinal changes in blood and stroke incidence. Stroke. 1984;15:839 846. 32. Kikumura T, Omae T, Ueda K, Takeshita M, Hirota Y. Long-term changes in blood prior to the development of cerebral infarction: the Hisayama study. J Chronic Dis. 1981;34:239 248. 33. Buck C, Baker P, Bass M, Donner A. The prognosis of according to age at onset. Hypertension. 1987;9:204 208. CLINICAL PERSPECTIVE Prior estimates of lifetime risk for cardiovascular disease (CVD) examined the impact of blood (BP) at the index age and did not account for changes in BP over time. Using pooled data from 7 diverse US cohort studies, we examined how changes in BP during middle age affect the lifetime risk for CVD, coronary heart disease, and stroke. Remaining lifetime risk from 55 years of age for CVD, coronary heart disease, and stroke was estimated for 61 585 white and black men and women by BP strata and by changes in BP over an average of 14 years. Lifetime risk for CVD was 52.5% (95% confidence interval, 51.3 53.7) for men and 39.9% (95% confidence interval, 38.7 41.0) for women. Lifetime risk for CVD was higher for blacks and increased with increasing BP at index age. Individuals who maintained or decreased their BP to normal levels had the lowest remaining lifetime risk for CVD, 21% to 41%, compared with individuals who had or developed by 55 years of age, 49.4% to 69.0%. Individuals who experience increases or decreases in BP in middle age have associated higher and lower remaining lifetime risk for CVD, and there may be a dose-response effect for the length of time at high BP levels. Prevention efforts should continue to emphasize the importance of lowering BP and avoiding or delaying the incidence of to reduce the lifetime risk for CVD.