Fatherhood and Cardiovascular Health, Disease, and Mortality: Associations From the Multi-Ethnic Study of Atherosclerosis

HIGHLIGHTS • Cardiovascular health in older age was worse for fathers than for nonfathers.• Cardiovascular health was worse for men who became fathers at a young age (i.e., <25 years).• Fatherhood's association with cardiovascular health and mortality differed by race.• Black fathers had lower age-adjusted mortality rates than Black nonfathers.• Incident cardiovascular disease did not differ between fathers and nonfathers.


INTRODUCTION
In 2021, the life expectancy for men in the U.S. was on average 6 years less than the life expectancy for women, and this discrepancy is most significant for racial and ethnic minorities. 1 Cardiovascular disease (CVD) is the leading cause of death among men, 2 and men have worse cardiovascular health (CVH) than women. 3CVH is a global summary of health factors and behaviors that have been proven to influence the development of CVD and is measured by the American Heart Association's Life's Essential 8 (LE8). 4en's CVH declines most rapidly from late adolescence through their 30s, 5 and during this time, most men in the U. S. become fathers. 6Notably, fatherhood is increasingly recognized as an influencer of health 7 and a key experience in the life course of men. 8he transition to fatherhood is associated with both positive and negative changes in CVH metrics, such as a decrease in smoking rates, 9 an increase in BMI, 10,11 an increase in fat intake, 12 and a decrease in physical activity. 9n addition, becoming a father at a younger age is associated with CVH metrics 13 and all-cause mortality. 14,15Being a father has been associated with higher rates of CVD in Chinese men 16 and lower CVD in cohorts of primarily White participants, suggesting potential differences by race and culture. 17,18Prior studies have found that fatherhood is also associated with lower risk of all-cause mortality 17,18 and multiple other health-related factors, such as depression, 19 marriage, 20 male infertility, 21 testosterone level, 22 and occupation and income, 6 which highlights the complex relationship between fatherhood and health.
Previous studies that evaluated fatherhood, CVH, CVD, and mortality have not included racially and ethnically diverse populations and lacked comprehensive CVH evaluation. 17,18,21,23The Multi-Ethnic Study of Atherosclerosis (MESA) is a unique scientific resource that has collected diverse participant population data on fatherhood and the information needed to calculate LE8.In addition, MESA's longitudinal study has relatively high retention and a large number of recorded disease events. 24This study's objectives were to measure the associations between fatherhood onset and status with (1) CVH scores measured by the LE8 framework and (2) incident CVD and all-cause mortality and assess for differences among race and ethnicity using MESA's diverse sample.The authors hypothesized that the association between fatherhood and CVH, CVD, and mortality would differ by race and ethnicity.

Study Sample
MESA is a prospective cohort study of 6,814 persons (47.2% men) aged 45−84 years without known CVD at baseline. 25MESA was designed to evaluate subclinical CVD in a diverse sample of Black, Chinese, Hispanic, and White individuals.Participants were recruited in 2000−2002 from 6 communities across the U.S.: Forsyth County (Winston-Salem), NC; New York, NY; Baltimore, MD; St. Paul, MN; Chicago, IL; and Los Angeles, CA.The study team excluded females from this study.They included all men from MESA who completed the family history interview during Exam 2  (2002−2004) and had complete data for comorbidities and CVH, resulting in a sample size of 2,814 males (Figure 1).The 399 men who were excluded from this study for missing data or not participating in Exam 2 differed from those included in the analytic sample by race/ethnicity (excluded participants were a higher proportion Black and lower proportion White than those included in the sample, p<0.001), marital status (excluded sample had lower percentage married men, p=0.02), and depressive symptoms (excluded sample had a higher average depression score, p=0.01) (Appendix Table 1, available online).This study follows STROBE guideline reporting for cohort studies.The study was approved by the IRBs of each site, and all participants provided informed consent.

Measures
Men were categorized as either fathers or nonfathers on the basis of information from the family history interview that was completed at Exam 2. In this interview, participants were asked to list any children's ages and medical conditions.Men who did not list any children were categorized as nonfathers.The authors calculated fathers' age at birth of their oldest child (fatherhood onset) as the age of the participant at Exam 2 minus the current age of their oldest child at Exam 2. Because becoming a father at older ages is associated with improved health relative to becoming a father at a younger age, the authors used age categories that were used in previous articles and made age ≥35 years the reference group. 13,15One participant had a child between Exams 1 and 2 and was excluded from the analysis.
CVH was measured utilizing the American Heart Association's LE8 framework. 4The 8 metrics consist of 4 health behaviors (healthy diet, participation in physical activity, avoidance of nicotine, and healthy sleep) and 4 health factors (healthy weight, level of blood lipids, levels of blood glucose, and blood pressure).There were no data for sleep at Exam 1, and the LE8 framework is appropriate to use even without all 8 metrics, 26 so this metric was excluded from our analysis.Participant information from Exam 1 was used to calculate a CVH score from 0 to 100 for each of the 7 metrics and average total CVH score of the 7 metrics; a higher score indicates a healthier metric.Data for diet, 27,28 physical activity, 29 and nicotine exposure 30 were collected through questionnaires adopted from validated instruments.BMI (kg/m 2 ) was calculated from measurements of weight and height.Total cholesterol and plasma glucose were measured from blood samples obtained after a 12-hour fast.Resting seated blood pressure was measured 3 times, and the mean of the last 2 measurements was used.
On the basis of previous literature, 3,6,19,20,22,31−33 the covariates assessed were age, race/ethnicity, marital status, education, having a primary care clinician, total gross family income level, depressive symptoms, total serum testosterone, and alcohol consumption.All covariate data were collected from Exam 1.For 68 individuals with missing income data at Exam 1, Exam 2 data were used.Depression was defined as a Center for Epidemiologic Studies Depression (CESD) Scale score ≥16.
A detailed description of the follow-up methods is available at https://internal.mesa-nhlbi.org.Briefly, follow-up for medical events occurred for 18 years (2000 −2018) at 9-to-12-month intervals through interviews and review of hospital records.Two physician members of the MESA morbidity and mortality review committee independently classified significant clinical events.If they disagreed, then the full committee made the final classification.CVD events were defined as definite or probable myocardial infarction, death due to coronary heart disease, resuscitated cardiac arrest, and coronary revascularization; fatal or nonfatal stroke (ischemic and hemorrhagic subtypes); definite or probable heart failure; peripheral arterial disease; and CVD death.CVD death was defined as death due to atherosclerotic heart disease, stroke, atherosclerotic disease other than coronary disease, and other CVD.

Statistical Analysis
The associations between fatherhood status and demographic characteristics and CVH metrics were examined using 2-sided chi-square and t-test procedures.Multivariable linear regression and general linear models were used to assess the relationship of fatherhood status and onset with CVH.Multivariable Cox proportional hazard regression models were used to evaluate the association of fatherhood status and onset with time to incident CVD events, CVD death, and all-cause mortality.There was no evidence of violations of the proportionality assumption (p>0.05).The covariates were included sequentially with age only and then all of our covariates: marital status, education, having a primary care clinician, income, total serum testosterone, alcohol consumption, and CESD depression score.For the CVD and mortality analysis, the authors ran a model with all the covariates and then a second model with the covariates and total CVH score at Exam 1.Because there were no changes to significant results with CVH added, the authors present the fully adjusted model only.They assessed for effect modification by race/ethnicity and marital status on the association of fatherhood with each outcome.In addition, they ran their analyses as an overall sample and stratified by race/ethnicity.Analysis was performed using SAS, Version 9.4 (SAS Institute, Cary, NC).

RESULTS
Among the 2,814 men, mean age at CVH assessment was 62.2 years, 24% self-identified as Black, 13% selfidentified Chinese, 22% self-identified Hispanic, and 41% self-identified White (Table 1).Most men (82.4%) were fathers at the time of CVH assessment.Fathers were less likely to report White race (37.5% vs 56.7% for nonfathers, p<0.001) and a bachelor's degree or higher level of education (40.3% vs 53.6%, p<0.001) but were not more likely to report total gross family income ≥$50,000 (48.5% vs 44.1%, p=0.08) (Table 1) than nonfathers.Fathers reported lower levels of depressive symptoms at Exam 1 than nonfathers (7.2% vs 11.7% Fathers had worse overall CVH (adjusted mean score=63.2vs 64.7, p=0.03) and more nicotine exposure (63.1 vs 66.6, p=0.04) than nonfathers.Otherwise, there was no difference in the other CVH metrics at Exam 1 between fathers and nonfathers (Table 2).After stratifying by race and ethnicity, the authors found that White fathers had worse total CVH than White nonfathers (65.8 vs 68.9, p=0.01), but there was no significant difference between fathers and nonfathers for Black, Chinese, or Hispanic participants (Appendix Table 3, available online).
Fathers who were aged <20 years had more nicotine exposure (p<0.01) and worse total CVH score (p=0.01)than fathers who were aged >35 years at the birth of their first child; fathers who were aged 20−24 years had worse nicotine exposure (p=0.03),BMI (p<0.01),blood sugar scores (p=0.049), and total CVH (p<0.001); and fathers who were aged 25−29 years had worse BMI (p<0.01) and blood lipid scores (p=0.0)(Table 2).In the stratified analysis, Black fathers who were aged <20 years (p=0.03)or 20−24 years (p<0.01)and Hispanic fathers who were aged <20 years (p=0.01)or 20−24 years (p<0.01) at the birth of their oldest child had worse total CVH than fathers who were aged >35 years at the birth of their oldest child (Appendix Table 3, available online).
During a median 17.6 years of follow-up, there were 608 CVD events (28.0%cumulative incidence), 214 deaths due to CVD (10.0%cumulative incidence), and 854 deaths of any causes (34.4% cumulative incidence).In age-adjusted models, fathers who were aged <20 years (hazard ratio [HR]=1.82;95% CI=1.27, 2.62) and 20 −24 years (HR=1.39;95% CI=1.07, 1.81) at the birth of their oldest child had a higher rate of all-cause mortality than fathers who were aged >35 years at the birth of their first child (Appendix Table 4, available online).In the fully adjusted models, there was no association between father's age at the birth of their first child and CVD events, CVD death, and all-cause mortality (Appendix Table 4, available online).
In age-adjusted and fully adjusted models, fathers and nonfathers did not differ in hazards for CVD and CVD death (Table 3).In age-adjusted models, fathers had a lower rate of all-cause mortality (HR=0.82;95% CI=0.69, 0.98), but this relationship was not significant in our fully adjusted model (HR=0.86;95% CI=0.72, 1.03) (Table 3).The relationship between fatherhood and all-cause mortality in an unadjusted survival curve differs for White men (Appendix Figure 1, available online), but this was not significant in adjusted models.Similarly, the authors observed evidence for statistical interaction between fatherhood and race/ethnicity on all-cause mortality before adjustment (interaction p=0.00) (Appendix Table 5, available online).After adjustment, this interaction was attenuated (p=0.36), and the authors did not observe evidence for an interaction between fatherhood and race/ethnicity on incident CVD or CVD death (all p for moderation >0.05).3][14][15]34 This study found that Black fathers, Hispanic fathers, and fathers overall who had their first child at ages <20 and 20 −24 years had worse CVH later in life than fathers who were aged >35 years at the birth of the first child. Similrly, a British cohort study found that men who were aged <20 years at the birth of their first child had higher mid-life blood pressure and BMI, even after controlling for multiple social and behavior factors.13 Early-timed fatherhood is associated with lower financial stability, 35 increased likelihood of multiple partnerships, 35 and depressive symptoms, 36 all of which may increase the risk of poor health behaviors and outcomes throughout the life course.
Moreover, in age-adjusted models, the authors found that men overall and Black men who transitioned to fatherhood at ages <20 and 20−24 years had a higher rate of all-cause mortality than those who entered fatherhood from age >35 years.Similar findings have been seen in other populations.Among a sample of Norwegian men, fathers who were aged <23 years at the birth of their first child had higher mortality than the other age groups. 15Among a Finnish cohort of siblings, men who entered fatherhood aged <24 years had higher mortality than their brothers who entered fatherhood at ages 25 −26 years. 14This higher mortality may be due to poor health behaviors, differences in cognitive abilities and sociodemographic characteristics, or other mechanisms that are not completely understood. 23,34Therefore, future research along with clinical and public health interventions that focus on young fathers may have important health impacts for men across their life course.
This study found that White fathers and fathers overall had worse total CVH than nonfathers.These findings are similar to prior observations that fatherhood has been associated with CVD risk factors such as weight gain, 10,11 a decrease in physical activity, 9 and an increase in fat intake. 12The changes in health metrics suggest that the additional commitment of child care responsibilities and stress during the transition to fatherhood may make it difficult for men to maintain a healthy diet and exercise for weight management. 37A study following men as they transition to fatherhood found that men had the largest weight gain in the first 6 months after child birth. 38Thus, to further define these relationships, future studies should focus on the time men transition to fatherhood in longitudinal cohorts to measure the life course trajectory of CVH metrics.This study's finding that fathers were more likely to use nicotine than nonfathers was in contrast to a prior evaluation from a sample of National Health and Nutrition Examination Survey men aged 20−59 years that reported that men who were living with children had lower odds of smoking than men living without children. 9The present study differs because it is comparing fathers with nonfathers rather than with those living with children, and the MESA population was significantly older (average age of 62.2 years) at the time of data collection.The transition to fatherhood has been associated with an increase in depression 36,39 and anxiety, 40 and these factors may contribute to nicotine use habits.Because smoking is the leading cause of preventable death in the U.S., 41 a better understanding of the relationship between fatherhood and tobacco use may help inform public health and clinical smoking-cessation programs.
This study found that fathers have lower age-adjusted all-cause mortality but no difference in CVD or CVD events.This study is not the first to observe an association between fatherhood and lower mortality 17,18 ; however, this study found that fathers actually have worse CVH, suggesting that there may be factors outside of CVH that contribute to differences in all-cause mortality between fathers and nonfathers.First, fathers may have a more robust social support system, and social connectedness has been associated with mortality. 42Second, fathers may be more likely to have someone as their future caretaker (i.e., their children) and help them attend medical appointments and manage medications and treatments.Finally, MESA fathers had lower rates of depressive symptoms than nonfathers.Because depression is associated with mortality in men, 43 it may be an important mediator in the relationship between fatherhood and mortality.This study found that the relationship between mortality and fatherhood was attenuated with covariates; therefore, future studies are needed to evaluate the social factors, such as social support, that may mediate the relationship between fatherhood and mortality.
This study's finding of an interaction between race/ ethnicity and fatherhood on all-cause mortality is a new development.The stratified age-adjusted analysis found that Black fathers had lower rates of all-cause mortality than nonfathers.Fatherhood has been associated with positive behavioral changes among Black men. 44Black men suffer from one of the lowest life expectancies of any demographic group in the U.S. 1 ; future studies that focus on understanding the influence of fatherhood on health outcomes can illuminate opportunities for health promotion for Black men with and without children.

Limitations
This study has several limitations.First, the average age of participants in this sample at the time of baseline health assessment was 62 years, so this study cannot measure the relationship between the transition to fatherhood and health; rather, the focus is on being a father and the association with CVH, CVD, and mortality.Second, MESA enrolled individuals who were free of CVD at baseline, and there is potential for selection bias if there is a difference between early-onset CVD among fathers and nonfathers.Third, it is possible that individuals became fathers after Exam 2, which would cause them to be miscategorized; however, in the U.S., only 3% of men become fathers over the age of 40 years. 6Fourth, data on male fertility were not available.Male infertility is associated with poor health outcomes, 21 and this may be an important unmeasured confounder in this analysis.Fifth, the authors did not have information on whom men live with or time spent caring for children.Therefore, they cannot assess the relationship of CVH and mortality with caring for, living with, or being engaged with children.Sixth, variables measured at Exam 1 do not necessarily represent the participants' characteristics at the time men became fathers, which could influence the relationship between fatherhood and health over the life course; for example, men might have been unmarried when they had children but recently married.Finally, there may be limits to this study's generalizability if the trends of fatherhood have evolved over the past several decades 45

CONCLUSIONS
In this multiethnic longitudinal sample, fatherhood status and age at entry to fatherhood were associated with CVH and age-adjusted mortality outcomes.This study highlights that men who transition to fatherhood at younger ages, especially Black and Hispanic men, may be at higher risk for poor health outcomes and may benefit from focused clinical and public health attention.The interaction between race/ethnicity and fatherhood on mortality underscores the complex influence of fatherhood on the health of men and suggests possible additional benefits of fatherhood for Black men.The life expectancy of men has stagnated over the past decade and actually decreased during the coronavirus disease 2019 (COVID-19) pandemic. 1ecause most men in the U.S. are fathers, 6 identifying some of the explanations for the associations between health, disease, and fatherhood could have important health implications for men, especially for men of color.

Figure 1 .
Figure 1.Flow diagram for determining sample size.
because most of the participants in the MESA sample became fathers in the 1970−1980s.

Table 1 .
Characteristics of MESA Sample at Baseline a Characteristics recorded atExam1, except the number and age of children, which was recorded at Exam 2. b Computed with chi-square across categories.cHigherCVHscore indicates a healthier metric.CESD, Center for Epidemiologic Studies Depression; CVH, cardiovascular health; HDL, high-density lipoprotein; MESA, Multi-Ethnic Study of Atherosclerosis; n/a, not available.hadCESDscore≥16, p<0.001).The mean age of fathers at the birth of their oldest child was 27.6 years (SD=6.3),and the average number of children among fathers was 2.9 (SD=1.7).There was a significant difference in the average age of fathers at the birth of their oldest child by race and ethnicity: 25.8 (SD=6.5)forBlackmen, 30.7 (SD=6.2) for Chinese men, 26.6 (SD=6.1)forHispanicmen, and 28.3 (SD=5.8)forWhitemen (p<0.001)(AppendixTable2,available online).

Table 2 .
Association Between Fatherhood Status and Onset With CVH Score Using Multivariate Linear Regression Boldface indicates statistical significance (p<0.05) between fathers and nonfathers or between men who became fathers at age >35 years and those of other age categories.
Note: a Individuals CVH metrics taken from Exam 1. b LS mean.Covariates include age, race/ethnicity, marital status, education, having a primary care clinician, income, total serum testosterone, former and current alcohol consumption, and CESD depression score.c Higher CVH score indicates a healthier metric.CESD, Center for Epidemiologic Studies Depression; CVH, cardiovascular health; LS, least square.