Childhood Psychological Distress and Healthy Cardiovascular Lifestyle 17–35 Years Later: The Potential Role of Mental Health in Primordial Prevention

Childhood Psychological Distress and Healthy Cardiovascular Lifestyle 17–35 Years Later: The... Abstract Background Maintaining a healthy lifestyle in adulthood has been shown to significantly reduce cardiovascular disease risk. Increasing evidence suggests that behavioral risk factors for cardiovascular disease are established in childhood; however, limited research has evaluated whether childhood psychological factors play a role. Purpose To evaluate the association between childhood psychological distress and young to mid adulthood healthy lifestyle. Methods Using prospective data from the 1958 British Birth Cohort, we assessed whether psychological distress in childhood (captured by internalizing and externalizing symptoms at ages 7, 11, and 16 years) predicted healthy lifestyle at ages 33 (N = 10,748) and 42 (N = 9,581) years. Healthy lifestyle was measured using an index previously demonstrated to predict cardiovascular disease, consisting of five components: absence of smoking, moderate alcohol consumption, regular physical activity, healthy diet, and ideal body weight. Results Few participants (3.8% at age 33 years and 2.8% at age 42 years) endorsed all five healthy lifestyle components. Linear regression models, adjusting for potential child- and family-level confounders, revealed that higher distress levels in childhood were negatively associated with healthy lifestyle at age 33 years (β = −0.11, SE = 0.01, p < .001) and 42 years (β = −0.13, SE = 0.01, p < .001). Higher distress was also associated with significantly lower odds of endorsing each lifestyle component, except physical activity, at both ages. Additional analyses indicated that childhood distress levels were highest among those whose lifestyle scores were low at age 33 and further declined between ages 33 and 42. Conclusions Psychological distress in childhood may indicate children at risk of less healthy lifestyle practices later in life. Although our findings are preliminary, psychological distress may also provide an important target for public health interventions aimed at preventing cardiovascular disease. Psychological distress, Lifestyle, Prospective cohort study, Epidemiology, Cardiovascular disease prevention Introduction Maintaining a healthy lifestyle reliably reduces risk of chronic disease, including coronary heart disease (CHD) [1, 2], stroke [3], diabetes [4], and cancer [5, 6], as well as all-cause mortality [7]. Clusters of lifestyle factors have a multiplicative impact on mortality compared with individual behaviors [8], and growing evidence suggests greater potential for public health impact of multiple- versus single-behavior interventions [9]. Thus, researchers are increasingly clustering multiple components into an overall “healthy lifestyle” index, which typically includes five components: absence of smoking, moderate alcohol consumption, regular physical activity, healthy diet, and body mass index (BMI) under 25 kg/m2 [1, 2, 10]. Particular attention has been paid to the impact of a healthy lifestyle on cardiovascular disease (CVD) outcomes. For example, researchers followed 84,129 initially disease-free women from the Nurses’ Health Study for 14 years and found that women with three, four, or five components of a healthy lifestyle (defined as above), respectively, had 57%, 66%, and 83% lower risk for incident CHD compared with all other women [1]. A similar graded relationship was found among 42,847 men in the Health Professionals Follow-up Study [2]. Furthermore, this study showed that men who adopted at least two additional healthy lifestyle factors during follow-up, compared to those who did not make any lifestyle changes, had a 27% reduction in CHD risk [2]. A strong reduction in risk has also been demonstrated for stroke, with one study reporting that over 50% of the ischemic stroke cases in each of two prospective cohorts were attributable to lack of adherence to a healthy lifestyle (less than all five health behaviors) [3]. While most cardiovascular and biomedical research to date has considered healthy lifestyle as a predictor of reduced risk [1–3, 7, 10, 11], less has considered the early-life determinants of healthy lifestyle as an outcome. With recognition that childhood factors set the stage for adulthood cardiovascular health [12], CVD primordial prevention—that is, prevention of disease risk factors before they begin to develop—may benefit greatly from identifying factors in childhood that offer some ability to predict who will adopt and maintain healthier lifestyle practices and therefore have better cardiovascular health as adults [13, 14]. Identification of such factors will enable both targeted surveillance and strategic intervention well before disease processes are initiated. One potential early-life contributor to a healthy adult lifestyle is childhood psychological distress. Psychological distress in childhood has been linked with a number of important cardiovascular health- related outcomes in adulthood, including increased inflammation [15], obesity [16], and cardiometabolic risk [17]. Although several lines of evidence suggest that better childhood mental health may lead to increased likelihood of engaging in healthier behaviors, most research looks at individual unhealthy behaviors as the end point, as opposed to an overall healthy lifestyle. Furthermore, nearly all the research linking mental health to behaviors has been conducted with adult populations and is cross-sectional in nature, so directions of effect cannot be established. Some prospective research has found that specific childhood disorders (e.g., anxiety, conduct disorder) are associated with certain cardiovascular health-related factors, such as cigarette smoking, problem drinking, and obesity [18–20]. However, to the best of our knowledge, no previous research has investigated the prospective relationship between a spectrum of child psychological distress (including levels that would not met criteria for any disorder) and an overall healthy lifestyle (as captured by multiple health components). Child psychological distress may provide an early indicator of risk for less healthy lifestyle and subsequently poorer health in adulthood. Using data from the 1958 British Birth Cohort Study, we evaluated whether psychological distress in childhood was associated with an overall healthy cardiovascular lifestyle 17–35 years later in young to mid adulthood. We hypothesized that children with higher levels of distress would have be less likely to engage in healthy behaviors in adulthood, even when adjusting for potential child- and family-level confounders of the relationship. We selected covariates based on previous evidence of their association with psychological distress and (at least one component of) a healthy lifestyle. For example, maternal education and marital circumstances have been shown to predict earlier onset of smoking in offspring [21]. Additionally, children of mothers who smoked during pregnancy are more likely, as adults, to have alcohol use disorders [22] and be obese [23]. Lower social class has also been found to predict less healthy lifestyles [24, 25]. These family factors may also be associated with child psychological distress [26, 27]. In addition, both childhood distress and adult health behaviors have been linked with characteristics of the child, including sex, low birth weight (LBW), academic performance, BMI, and physical health status [28, 29]. Thus, the present study included consideration of these relevant characteristics of the child and the child’s family. Methods Sample Population Data come from the 1958 British Birth Cohort Study (also known as the NCDS) [30]—an ongoing longitudinal study of people born in Great Britain during a single week in March 1958 (N = 18,558). Information on various aspects of the cohort members’ physical and mental health was collected at birth and during six follow-up waves between 1964 and 2000 (at ages 7, 11, 16, 23, 33, and 42 years). Informed consent was obtained from all individual participants included in the study. The cohort is described in detail elsewhere [30]. Outcomes for this study were derived from measurements at Wave 5 (age 33) and Wave 6 (age 42). At Wave 5, 11,469 of the original cohort members participated in the study. Participants missing data on any of the five healthy lifestyle components at age 33 years were excluded from the age 33 analyses (N = 588). We excluded an additional 133 people who did not have any measures of psychological distress in childhood, leaving an analytic sample of 10,748 at age 33 (Wave 5). At Wave 6 (age 42), 11,419 of the original cohort members participated in the study. Most (N = 9,890) of the Wave 6 participants also participated in Wave 5. Participants missing data on any of the five healthy lifestyle components at age 42 years were excluded from the age 42 analyses (N = 1,723). The majority (N = 1,641) of those excluded were missing data on only one behavior (predominantly alcohol consumption, N = 1,531). We excluded an additional 115 people who did not have any measures of psychological distress in childhood, leaving an analytic sample of 9,581 at age 42 (Wave 6). (See Fig. 1 for a flow chart of sample selection and exclusions.) Most participants (N = 7,998) were in the analytic samples at both waves. Generally, as expected, those in the analytic samples tended to be healthier and better off than those excluded from the original birth cohort sample (Supplementary Table A). Fig. 1. View largeDownload slide Flow chart of sample selection and exclusions. Fig. 1. View largeDownload slide Flow chart of sample selection and exclusions. Measures Childhood psychological distress Childhood psychological distress was ascertained from validated measures of psychopathology and emotional/behavioral functioning: the Bristol Social Adjustment Guide (BSAG) [31] at ages 7 and 11 and the Rutter Behaviour Scale (RBS) [32] at age 16; all measures were completed by participants’ teachers. Items from the BSAG and the RBS were designed to capture levels of emotional disturbance and social maladjustment [33]. The overall score derived from the 146-item teacher-rated BSAG has been found to have good inter-rater reliability (r = .76) [34]. Similarly, the overall score derived from the 26-item teacher-rated version of the RBS has been shown to have good inter-rater reliability (r = .72) and test-retest reliability (r = .89) [32]. Within the National Child Development Study (NCDS) population, the BSAG scores at ages 7 and 11 and the RBS score at age 16 were all significantly correlated (Rs = 0.26–0.39; p-values all <.001). The two measures derived from the overall scores appear to be fairly comparable [35] and have each been shown to predict increased risk of adulthood psychiatric disorders [36]. Prior research has more specifically demonstrated two primary forms of childhood distress—externalizing and internalizing—associated with increased risk of adverse developmental and health outcomes [37, 38]. These two types of distress share some common characteristics, but they can also be distinguished from one another on the basis of distinctive specific features [39]. To ensure that each form of distress was adequately captured, and to be consistent with previous work with these measures in this cohort [36, 38], we first created separate scores for each type of distress at each age by summing across symptoms related to externalizing (e.g., restlessness, aggression) and those related to internalizing (e.g., worry, depression). If participants were missing at least half of either the externalizing or the internalizing items, we set their corresponding subscale score to missing; otherwise, for approximately 1% of the sample, we substituted missing values (typically only one scale item) with the mean of their completed items. We then created a continuous total distress score at each age in childhood (7, 11, and 16 years), by computing the mean of the standardized externalizing and internalizing scores. Internal consistency reliabilities of the total distress measures were high: Cronbach’s alphas = 0.79 at age 7, 0.76 at age 11, and 0.86 at age 16. Finally, we computed the mean score across the three time points to create the overall childhood psychological distress measure, where higher scores indicate greater psychological distress. For ease of interpretation, we standardized this summary score to have a mean of zero and a SD of 1. Healthy lifestyle index We created separate healthy lifestyle index scores at ages 33 and 42 years, based on criteria used in previous work demonstrated to predict disease outcomes [1–3, 10]. The healthy lifestyle index comprised five components, all self-reported: (i) absence of current cigarette smoking; (ii) moderate alcohol consumption (≤7 drinks/week for women, ≤14 for men; excluding abstention); (iii) regular physical activity (≥4–5 times/week); (iv) healthy diet (top 40% of diet score); and (v) ideal body weight (BMI <25 kg/m2). Following prior work that developed this measure, we created a dichotomous measure of each healthy lifestyle component (0/1, with 1 indicating presence of healthy levels of the component), and then, summed the components to create a continuous score. Thus, the healthy lifestyle index ranged from 0 to 5, with higher scores indicating healthier lifestyles. Smoking status was self-reported at ages 33 and 42. We assigned all current nonsmokers, including both never smokers and former smokers, a score of 1 on the absence of smoking component. We derived alcohol consumption from a self-reported measure of overall frequency of drinking (with response options ranging from never to most days) and from a series of self-report questions on the number of specific alcoholic beverages (e.g., beer, wine, spirits) consumed in the last 7 days. In addition, we assigned a problem drinking score based on endorsement of two or more out of four items on the CAGE questionnaire, a widely used and validated method for identifying alcoholism [40]. Women who reported drinking more than seven drinks in the past 7 days, men who reported drinking more than 14 drinks in the past 7 days, and all those who reported a past-year drinking problem, were given a score of 0. Consistent with several measures of healthy lifestyle (e.g., 1, 2, 3, 7, 11), those who reported never drinking were also given a score of 0, as abstention is not considered optimal with regard to cardiovascular health [41]. All others were considered to be moderate drinkers and, thus, were given a score of 1 on the moderate alcohol consumption component. If participants reported regularly engaging in any sport of leisure activity that involved physical exercise (e.g., sports, aerobics classes, weight training running, swimming cycling, walking, dancing) [42], they were asked a follow-up question about how frequently they took part in the activities. Six response options ranged from less than twice a month to every day. Anyone who reported regularly exercising at least 4 to 5 days a week was characterized as engaging in regular physical activity (score = 1). To capture quality of diet, we followed an approach previously used in this cohort [43]. Participants self-reported their average weekly frequency of consumption of a variety of food types, with response options ranging from never to more than once a day. We coded food consumption frequency on a scale from zero to five, giving higher scores to frequent “healthy” food consumption (fresh fruit and salad/raw vegetables) and lower scores to frequent “unhealthy” food consumption (chips, fried food, sweets/chocolates, and cakes/biscuits). We then summed the scores over the six food types to create a continuous measure of overall diet quality, with higher scores indicating better diet quality. Following previous literature on a healthy lifestyle [2, 3], we dichotomized the continuous diet score to define healthy diet as the top 40% of the healthy diet scores. We calculated BMI from weight and height measurements, which were self-reported at age 33 and 42 and obtained from a medical examination at age 45. If self-reported measurements were not available at age 42 (N = 951), we substituted weight and height measurements from the medical examination at age 45. BMI measures at ages 42 and 45 were highly correlated (r = .87, p<.001). Following prior work, we defined ideal body weight as BMI <25 kg/m2 (i.e., not overweight or obese) [3, 13]. While this categorization of ideal body weight includes underweight (BMI <18.5 kg/m2), being underweight has relatively few implications for overall cardiovascular health [13]. Less than 2.5% of our analytic samples were underweight. Covariates Child covariates consisted of sex (male, female); LBW (<5.5 pounds vs. ≥ = 5.5 pounds); cognition at age 7 (assessed using a using a 10-problem arithmetic test [44] and the Southgate Reading Test [45], respectively, with the lowest 10% of scores in both tests counting as low performance) [46]; overweight at age 7 (medical-examiner-recorded BMI reached or exceeded cut points laid out by the International Obesity Task Force: 17.92 kg/m2 for boys and 17.75 kg/m2 for girls [47]); and presence of any physical health problems by age 7 (any asthma, chronic illness, or signs of heart problems; yes/no). Family covariates consisted of father’s occupation at child’s birth (or age 7 if missing; manual vs. nonmanual), mother’s education level (left school at/before vs. after minimum leaving age), mother’s cigarette smoking during pregnancy (yes/no), and mother’s marital status at child’s birth (single/divorced/widowed vs. married). For all covariates, we included a “missing” category, to maintain the full available sample size at each time point in adjusted models. For regression analyses, covariates were dummy coded, with the typically more favorable category counting as the reference category (e.g., normal birth weight, normal math and reading scores, not overweight, no physical health problems, nonmanual paternal occupation, beyond minimal maternal education, maternal nonsmoking, married). Statistical Analysis Because healthy lifestyle scores were normally distributed, we used linear regression models to estimate the relationship between childhood psychological distress and adulthood healthy lifestyle, running separate models for the age 33 and 42 outcomes. (Note: we came to the same conclusions using Poisson regression models.) In a minimally adjusted model (Model 1), we adjusted for sex, and in a second model (Model 2), we adjusted for sex as well as all child and family covariates, to assess potential confounding. As a sensitivity analysis, we further adjusted for concurrent adult distress, to assess the independent effect of childhood distress. Adult distress was measured at ages 33 and 42 years, using the self-report psychological distress subscale of the Malaise Inventory [48]. The Inventory includes 15 items measuring psychological distress (e.g., “do you often feel miserable or depressed?”) with yes/no response options. We dealt with missing items in the same manner as in childhood. In additional analyses, we examined change in healthy lifestyle scores over time, among those who had healthy lifestyle measures at both time points (N = 7,998). We categorized lifestyle as improving (i.e., higher score at age 42 compared with 33), declining, or staying the same, in conjunction with whether people started with unhealthy (0 to 2 healthy lifestyle components at age 33) or healthy (3 to 5 components) lifestyle. This yielded six categories: unhealthy and declined, unhealthy constant, unhealthy but improved, healthy but declined, healthy constant, and healthy and improved. We used generalized linear models to assess whether childhood psychological distress was associated with the nature of the change in healthy lifestyle between age 33 and 42 years. To see if the child psychological functioning had a particularly strong impact on any one of the healthy lifestyle components, we explored associations with each component separately (at both ages) using logistic regression models, adjusting for sex (Model 1) plus all covariates (Model 2). While ideal body weight (BMI < 25 kg/m2) is typically included in healthy lifestyle indices [1, 3, 10, 11], unlike the other index components, it is not, strictly speaking, a health behavior. In sensitivity analyses, we reran the main models excluding BMI from the index. To minimize potential bias due to participant attrition, we applied inverse probability weights to all analyses, estimating weights based on key factors associated with attrition: sex, LBW, math and reading scores, internalizing and externalizing symptoms, overweight status, physical health problems, father’s occupation, mother’s education, and mother’s marital status. All analyses were performed using SAS version 9.3 software (SAS Institute, Inc., Cary, NC). Access to the data, including a special license agreement for use of biomedical data, was granted by the Economic and Social Data service council, the UK Data Archive, and the Center for Longitudinal Studies. Ethical approval of this study was given by the Harvard T.H. Chan School of Public Health Institutional Review Board. Results Baseline Characteristics and Healthy Lifestyle in Adulthood In the age 33 sample, half of the participants were female (50.7%, Table 1). At birth, 4.7% of these participants were low birth weight, the majority of their fathers had a manual occupation (70.0%), and the majority of their mothers were married (92.3%), had not smoked cigarettes during pregnancy (63.4%), and had left school at or before the minimum leaving age (70.2%). By age 7, 6.0% of respondents had experienced physical health problems. At age 7, 5.4% had low math scores, 7.1% had low reading scores, and 8.6% were overweight. The distribution of characteristics was almost identical in the sample at age 42 years. All covariates were significantly associated with childhood distress. Table 1 Sample Characteristics at Wave 5, and Mean Childhood Psychological Distress Z-score by Covariates Child variables Wave 5 (age 33)a Child psychological distress Z-scoreb N = 10,748 N (%) Mean (SD)  Sex   Female 5,446 (50.67) −0.15 (0.92)   Male 5,302 (49.33) 0.14 (1.05)  Low birth weight   No 9,692 (90.17) −0.02 (0.99)   Yes 500 (4.65) 0.17 (1.04)   Missing 556 (5.17) 0.08 (1.12)  Math performance   Normal 9,061 (84.30) −0.07 (0.93)   Low 584 (5.43) 0.76 (1.07)   Missing 1,103 (10.26) 0.12 (1.23)  Reading performance   Normal 8,905 (82.85) −0.11 (0.90)   Low 766 (7.13) 0.88 (1.06)   Missing 1,077 (10.02) 0.12 (1.25)  Overweight   No 7,886 (73.37) −0.04 (0.96)   Yes 925 (8.61) −0.06 (0.91)   Missing 1,937 (18.02) 0.15 (1.16)  Physical health problems   No 8,880 (82.62) −0.03 (0.96)   Yes 641 (5.96) 0.04 (1.02)   Missing 1,227 (11.42) 0.13 (1.19) Family variables  Father’s occupation   Nonmanual 2,939 (27.34) −0.25 (0.86)   Manual 7,524 (70.00) 0.08 (1.02)   Missing 285 (2.65) 0.16 (1.24)  Mother’s education level   Stayed after min leaving age 2,657 (24.72) −0.23 (0.88)   Left at/before min leaving age 7,549 (70.24) 0.07 (1.01)   Missing 542 (5.04) 0.09 (1.13)  Mother’s smoking during pregnancy   No 6,814 (63.40) −0.08 (0.95)   Yes 3,298 (30.68) 0.13 (1.04)   Missing 636 (5.92) 0.07 (1.13)  Mother’s marital status   Married 9,919 (92.29) −0.02 (0.99)   Unmarried/ Divorced/Widowed 310 (2.88) 0.29 (1.06)   Missing 519 (4.83) 0.09 (1.13) Child variables Wave 5 (age 33)a Child psychological distress Z-scoreb N = 10,748 N (%) Mean (SD)  Sex   Female 5,446 (50.67) −0.15 (0.92)   Male 5,302 (49.33) 0.14 (1.05)  Low birth weight   No 9,692 (90.17) −0.02 (0.99)   Yes 500 (4.65) 0.17 (1.04)   Missing 556 (5.17) 0.08 (1.12)  Math performance   Normal 9,061 (84.30) −0.07 (0.93)   Low 584 (5.43) 0.76 (1.07)   Missing 1,103 (10.26) 0.12 (1.23)  Reading performance   Normal 8,905 (82.85) −0.11 (0.90)   Low 766 (7.13) 0.88 (1.06)   Missing 1,077 (10.02) 0.12 (1.25)  Overweight   No 7,886 (73.37) −0.04 (0.96)   Yes 925 (8.61) −0.06 (0.91)   Missing 1,937 (18.02) 0.15 (1.16)  Physical health problems   No 8,880 (82.62) −0.03 (0.96)   Yes 641 (5.96) 0.04 (1.02)   Missing 1,227 (11.42) 0.13 (1.19) Family variables  Father’s occupation   Nonmanual 2,939 (27.34) −0.25 (0.86)   Manual 7,524 (70.00) 0.08 (1.02)   Missing 285 (2.65) 0.16 (1.24)  Mother’s education level   Stayed after min leaving age 2,657 (24.72) −0.23 (0.88)   Left at/before min leaving age 7,549 (70.24) 0.07 (1.01)   Missing 542 (5.04) 0.09 (1.13)  Mother’s smoking during pregnancy   No 6,814 (63.40) −0.08 (0.95)   Yes 3,298 (30.68) 0.13 (1.04)   Missing 636 (5.92) 0.07 (1.13)  Mother’s marital status   Married 9,919 (92.29) −0.02 (0.99)   Unmarried/ Divorced/Widowed 310 (2.88) 0.29 (1.06)   Missing 519 (4.83) 0.09 (1.13) aDistributions are shown for sample at Wave 5, which were almost identical to distributions at Wave 6. bHigher (positive) scores indicate more distress. All p-values <.001 for tests of between group differences in child distress Z-scores. View Large Table 1 Sample Characteristics at Wave 5, and Mean Childhood Psychological Distress Z-score by Covariates Child variables Wave 5 (age 33)a Child psychological distress Z-scoreb N = 10,748 N (%) Mean (SD)  Sex   Female 5,446 (50.67) −0.15 (0.92)   Male 5,302 (49.33) 0.14 (1.05)  Low birth weight   No 9,692 (90.17) −0.02 (0.99)   Yes 500 (4.65) 0.17 (1.04)   Missing 556 (5.17) 0.08 (1.12)  Math performance   Normal 9,061 (84.30) −0.07 (0.93)   Low 584 (5.43) 0.76 (1.07)   Missing 1,103 (10.26) 0.12 (1.23)  Reading performance   Normal 8,905 (82.85) −0.11 (0.90)   Low 766 (7.13) 0.88 (1.06)   Missing 1,077 (10.02) 0.12 (1.25)  Overweight   No 7,886 (73.37) −0.04 (0.96)   Yes 925 (8.61) −0.06 (0.91)   Missing 1,937 (18.02) 0.15 (1.16)  Physical health problems   No 8,880 (82.62) −0.03 (0.96)   Yes 641 (5.96) 0.04 (1.02)   Missing 1,227 (11.42) 0.13 (1.19) Family variables  Father’s occupation   Nonmanual 2,939 (27.34) −0.25 (0.86)   Manual 7,524 (70.00) 0.08 (1.02)   Missing 285 (2.65) 0.16 (1.24)  Mother’s education level   Stayed after min leaving age 2,657 (24.72) −0.23 (0.88)   Left at/before min leaving age 7,549 (70.24) 0.07 (1.01)   Missing 542 (5.04) 0.09 (1.13)  Mother’s smoking during pregnancy   No 6,814 (63.40) −0.08 (0.95)   Yes 3,298 (30.68) 0.13 (1.04)   Missing 636 (5.92) 0.07 (1.13)  Mother’s marital status   Married 9,919 (92.29) −0.02 (0.99)   Unmarried/ Divorced/Widowed 310 (2.88) 0.29 (1.06)   Missing 519 (4.83) 0.09 (1.13) Child variables Wave 5 (age 33)a Child psychological distress Z-scoreb N = 10,748 N (%) Mean (SD)  Sex   Female 5,446 (50.67) −0.15 (0.92)   Male 5,302 (49.33) 0.14 (1.05)  Low birth weight   No 9,692 (90.17) −0.02 (0.99)   Yes 500 (4.65) 0.17 (1.04)   Missing 556 (5.17) 0.08 (1.12)  Math performance   Normal 9,061 (84.30) −0.07 (0.93)   Low 584 (5.43) 0.76 (1.07)   Missing 1,103 (10.26) 0.12 (1.23)  Reading performance   Normal 8,905 (82.85) −0.11 (0.90)   Low 766 (7.13) 0.88 (1.06)   Missing 1,077 (10.02) 0.12 (1.25)  Overweight   No 7,886 (73.37) −0.04 (0.96)   Yes 925 (8.61) −0.06 (0.91)   Missing 1,937 (18.02) 0.15 (1.16)  Physical health problems   No 8,880 (82.62) −0.03 (0.96)   Yes 641 (5.96) 0.04 (1.02)   Missing 1,227 (11.42) 0.13 (1.19) Family variables  Father’s occupation   Nonmanual 2,939 (27.34) −0.25 (0.86)   Manual 7,524 (70.00) 0.08 (1.02)   Missing 285 (2.65) 0.16 (1.24)  Mother’s education level   Stayed after min leaving age 2,657 (24.72) −0.23 (0.88)   Left at/before min leaving age 7,549 (70.24) 0.07 (1.01)   Missing 542 (5.04) 0.09 (1.13)  Mother’s smoking during pregnancy   No 6,814 (63.40) −0.08 (0.95)   Yes 3,298 (30.68) 0.13 (1.04)   Missing 636 (5.92) 0.07 (1.13)  Mother’s marital status   Married 9,919 (92.29) −0.02 (0.99)   Unmarried/ Divorced/Widowed 310 (2.88) 0.29 (1.06)   Missing 519 (4.83) 0.09 (1.13) aDistributions are shown for sample at Wave 5, which were almost identical to distributions at Wave 6. bHigher (positive) scores indicate more distress. All p-values <.001 for tests of between group differences in child distress Z-scores. View Large Distress Z-scores at each age in childhood (7, 11, and 16 years) were positively correlated with one another, with correlations ranging from 0.26 to 0.38 (all p-values were <.001). Child distress was also modestly correlated with adult distress at age 33 (0.18, p < .001) and 42 (0.15, p < .001). The healthy lifestyle index scores at age 33 were positively correlated with healthy lifestyle index at age 42 (polychoric correlation = 0.55, asymptotic standard error [ASE] = 0.009). Correlations between childhood psychological distress, overall healthy lifestyle score, and each dichotomous component of the healthy lifestyle index are presented in Supplementary Table B. The proportion of nonsmokers increased slightly between ages 33 and 42, while the proportion of participants reporting moderate alcohol consumption decreased over time (Table 2). The proportions of the other three lifestyle factors remained fairly stable between the two time points. Only 3.8% and 2.8% of 33 and 42 year olds, respectively, endorsed all five components of a healthy lifestyle. The majority of the people reported having two or three healthy lifestyle components at each age. Table 2 Number (and Percent) of People, and Mean (and SD) of Childhood Psychological Distress Z-score, by Healthy Lifestyle Component and Score Age 33 (Wave 5) N = 10,748 Age 42 (Wave 6) N = 9,581 People Child distress Z-scorea People Child distress Z-scorea N (%) M (SD) N (%) M (SD) Healthy lifestyle component Nonsmoker  No 3,561 (33.13) 0.13 (1.03) 2,809 (29.32) 0.11 (1.00)  Yes 7,187 (66.87) −0.21 (0.84) 6,772 (70.68) −0.23 (0.83) Moderate alcohol consumption  No 2,994 (27.86) −0.06 (0.96) 4,211 (43.95) −0.09 (0.94)  Yes 7,754 (72.14) −0.11 (0.91) 5,370 (56.05) −0.16 (0.86) Regular physical activity  No 7,988 (74.32) −0.11 (0.92) 7,112 (74.23) −0.14 (0.89)  Yes 2,760 (25.68) −0.07 (0.94) 2,469 (25.77) −0.11 (0.93) Healthy diet  No 6,759 (62.89) −0.04 (0.95) 6,126 (63.94) −0.07 (0.93)  Yes 3,989 (37.11) −0.20 (0.87) 3,455 (36.06) −0.24 (0.83) Ideal body weight (body mass index <25)  No 4,664 (43.39) −0.03 (0.95) 4,323 (45.12) −0.04 (0.93)  Yes 6,084 (56.61) −0.15 (0.90) 5,258 (54.88) −0.20 (0.86) Total number of healthy lifestyle components  0 295 (2.74) 0.17 (1.08) 318 (3.32) 0.24 (1.03)  1 1,450 (13.49) 0.12 (1.01) 1,666 (17.39) 0.06 (0.99)  2 3,262 (30.35) −0.02 (0.96) 3,029 (31.61) −0.08 (0.90)  3 3,569 (33.21) −0.16 (0.88) 2,944 (30.73) −0.21 (0.85)  4 1,767 (16.44) −0.29 (0.80) 1,352 (14.11) −0.33 (0.75)  5 405 (3.77) −0.28 (0.77) 272 (2.84) −0.38 (0.71) Age 33 (Wave 5) N = 10,748 Age 42 (Wave 6) N = 9,581 People Child distress Z-scorea People Child distress Z-scorea N (%) M (SD) N (%) M (SD) Healthy lifestyle component Nonsmoker  No 3,561 (33.13) 0.13 (1.03) 2,809 (29.32) 0.11 (1.00)  Yes 7,187 (66.87) −0.21 (0.84) 6,772 (70.68) −0.23 (0.83) Moderate alcohol consumption  No 2,994 (27.86) −0.06 (0.96) 4,211 (43.95) −0.09 (0.94)  Yes 7,754 (72.14) −0.11 (0.91) 5,370 (56.05) −0.16 (0.86) Regular physical activity  No 7,988 (74.32) −0.11 (0.92) 7,112 (74.23) −0.14 (0.89)  Yes 2,760 (25.68) −0.07 (0.94) 2,469 (25.77) −0.11 (0.93) Healthy diet  No 6,759 (62.89) −0.04 (0.95) 6,126 (63.94) −0.07 (0.93)  Yes 3,989 (37.11) −0.20 (0.87) 3,455 (36.06) −0.24 (0.83) Ideal body weight (body mass index <25)  No 4,664 (43.39) −0.03 (0.95) 4,323 (45.12) −0.04 (0.93)  Yes 6,084 (56.61) −0.15 (0.90) 5,258 (54.88) −0.20 (0.86) Total number of healthy lifestyle components  0 295 (2.74) 0.17 (1.08) 318 (3.32) 0.24 (1.03)  1 1,450 (13.49) 0.12 (1.01) 1,666 (17.39) 0.06 (0.99)  2 3,262 (30.35) −0.02 (0.96) 3,029 (31.61) −0.08 (0.90)  3 3,569 (33.21) −0.16 (0.88) 2,944 (30.73) −0.21 (0.85)  4 1,767 (16.44) −0.29 (0.80) 1,352 (14.11) −0.33 (0.75)  5 405 (3.77) −0.28 (0.77) 272 (2.84) −0.38 (0.71) aHigher (positive) scores indicate more distress. View Large Table 2 Number (and Percent) of People, and Mean (and SD) of Childhood Psychological Distress Z-score, by Healthy Lifestyle Component and Score Age 33 (Wave 5) N = 10,748 Age 42 (Wave 6) N = 9,581 People Child distress Z-scorea People Child distress Z-scorea N (%) M (SD) N (%) M (SD) Healthy lifestyle component Nonsmoker  No 3,561 (33.13) 0.13 (1.03) 2,809 (29.32) 0.11 (1.00)  Yes 7,187 (66.87) −0.21 (0.84) 6,772 (70.68) −0.23 (0.83) Moderate alcohol consumption  No 2,994 (27.86) −0.06 (0.96) 4,211 (43.95) −0.09 (0.94)  Yes 7,754 (72.14) −0.11 (0.91) 5,370 (56.05) −0.16 (0.86) Regular physical activity  No 7,988 (74.32) −0.11 (0.92) 7,112 (74.23) −0.14 (0.89)  Yes 2,760 (25.68) −0.07 (0.94) 2,469 (25.77) −0.11 (0.93) Healthy diet  No 6,759 (62.89) −0.04 (0.95) 6,126 (63.94) −0.07 (0.93)  Yes 3,989 (37.11) −0.20 (0.87) 3,455 (36.06) −0.24 (0.83) Ideal body weight (body mass index <25)  No 4,664 (43.39) −0.03 (0.95) 4,323 (45.12) −0.04 (0.93)  Yes 6,084 (56.61) −0.15 (0.90) 5,258 (54.88) −0.20 (0.86) Total number of healthy lifestyle components  0 295 (2.74) 0.17 (1.08) 318 (3.32) 0.24 (1.03)  1 1,450 (13.49) 0.12 (1.01) 1,666 (17.39) 0.06 (0.99)  2 3,262 (30.35) −0.02 (0.96) 3,029 (31.61) −0.08 (0.90)  3 3,569 (33.21) −0.16 (0.88) 2,944 (30.73) −0.21 (0.85)  4 1,767 (16.44) −0.29 (0.80) 1,352 (14.11) −0.33 (0.75)  5 405 (3.77) −0.28 (0.77) 272 (2.84) −0.38 (0.71) Age 33 (Wave 5) N = 10,748 Age 42 (Wave 6) N = 9,581 People Child distress Z-scorea People Child distress Z-scorea N (%) M (SD) N (%) M (SD) Healthy lifestyle component Nonsmoker  No 3,561 (33.13) 0.13 (1.03) 2,809 (29.32) 0.11 (1.00)  Yes 7,187 (66.87) −0.21 (0.84) 6,772 (70.68) −0.23 (0.83) Moderate alcohol consumption  No 2,994 (27.86) −0.06 (0.96) 4,211 (43.95) −0.09 (0.94)  Yes 7,754 (72.14) −0.11 (0.91) 5,370 (56.05) −0.16 (0.86) Regular physical activity  No 7,988 (74.32) −0.11 (0.92) 7,112 (74.23) −0.14 (0.89)  Yes 2,760 (25.68) −0.07 (0.94) 2,469 (25.77) −0.11 (0.93) Healthy diet  No 6,759 (62.89) −0.04 (0.95) 6,126 (63.94) −0.07 (0.93)  Yes 3,989 (37.11) −0.20 (0.87) 3,455 (36.06) −0.24 (0.83) Ideal body weight (body mass index <25)  No 4,664 (43.39) −0.03 (0.95) 4,323 (45.12) −0.04 (0.93)  Yes 6,084 (56.61) −0.15 (0.90) 5,258 (54.88) −0.20 (0.86) Total number of healthy lifestyle components  0 295 (2.74) 0.17 (1.08) 318 (3.32) 0.24 (1.03)  1 1,450 (13.49) 0.12 (1.01) 1,666 (17.39) 0.06 (0.99)  2 3,262 (30.35) −0.02 (0.96) 3,029 (31.61) −0.08 (0.90)  3 3,569 (33.21) −0.16 (0.88) 2,944 (30.73) −0.21 (0.85)  4 1,767 (16.44) −0.29 (0.80) 1,352 (14.11) −0.33 (0.75)  5 405 (3.77) −0.28 (0.77) 272 (2.84) −0.38 (0.71) aHigher (positive) scores indicate more distress. View Large Childhood Psychological Distress and Healthy Lifestyle Adjusting for sex, childhood distress was negatively associated with healthy lifestyle at age 33 (β = −0.14, SE = 0.01, p<.001) and 42 (β = −0.15, SE = 0.01, p<.001) (Table 3). The effect of distress on healthy lifestyle attenuated slightly in fully adjusted models: Holding all other variables constant, a 1 SD increase in distress was associated, on average, with a 0.11 lower healthy lifestyle index score at age 33 (SE = 0.01, p<.001) and a 0.13 lower score at age 42 (SE = 0.01, p<.001). These estimates are similar in magnitude, for example, to the significant effect on age 42 healthy lifestyle of maternal smoking versus nonsmoking during pregnancy (β = −0.10) and manual versus nonmanual paternal occupation (β = −0.12), both established predictors of health behaviors [49, 50]. When comparing mean healthy lifestyle levels at age 42 between the groups with lower (bottom quartile, mean healthy lifestyle score = 2.6) versus higher (top quartile, mean healthy lifestyle score = 2.1) psychological distress, the mean difference in lifestyle score was approximately 0.5 on a five-point scale, equivalent to a 10% decrease in level of healthy lifestyle (or half a health behavior). Table 3 Linear Regression of the Association Between Childhood Psychological Distress and Healthy Lifestylea at Ages 33 and 42 Years Model 1 Model 2 β (SE) p-value β (SE) p-value Age 33 years (N = 10,748) −0.14 (0.01) <.001 −0.11 (0.01) <.001 Age 42 years (N = 9,581) −0.15 (0.01) <.001 −0.13 (0.01) <.001 Model 1 Model 2 β (SE) p-value β (SE) p-value Age 33 years (N = 10,748) −0.14 (0.01) <.001 −0.11 (0.01) <.001 Age 42 years (N = 9,581) −0.15 (0.01) <.001 −0.13 (0.01) <.001 a“Healthy lifestyle” consists of 5 components: absence of smoking, moderate alcohol consumption, regular physical activity, healthy diet, and ideal body weight. We modeled childhood psychological distress, where higher scores indicate more distress. Model 1 adjusts for sex; Model 2 adjusts for sex plus child covariates (LBW, math performance, reading performance, overweight, and physical health problems) and family covariates (father’s occupation, and mother’s education, smoking during pregnancy, and marital status). View Large Table 3 Linear Regression of the Association Between Childhood Psychological Distress and Healthy Lifestylea at Ages 33 and 42 Years Model 1 Model 2 β (SE) p-value β (SE) p-value Age 33 years (N = 10,748) −0.14 (0.01) <.001 −0.11 (0.01) <.001 Age 42 years (N = 9,581) −0.15 (0.01) <.001 −0.13 (0.01) <.001 Model 1 Model 2 β (SE) p-value β (SE) p-value Age 33 years (N = 10,748) −0.14 (0.01) <.001 −0.11 (0.01) <.001 Age 42 years (N = 9,581) −0.15 (0.01) <.001 −0.13 (0.01) <.001 a“Healthy lifestyle” consists of 5 components: absence of smoking, moderate alcohol consumption, regular physical activity, healthy diet, and ideal body weight. We modeled childhood psychological distress, where higher scores indicate more distress. Model 1 adjusts for sex; Model 2 adjusts for sex plus child covariates (LBW, math performance, reading performance, overweight, and physical health problems) and family covariates (father’s occupation, and mother’s education, smoking during pregnancy, and marital status). View Large Excluding BMI from the healthy lifestyle index yielded similar results (Supplementary Table C). When simultaneously adjusting for concurrent adult distress, in addition to all other covariates, the effect of child distress on healthy lifestyle remained significant (age 33: β = −0.09, SE = 0.01, p<.001; age 42: β = −0.10, SE = 0.01, p<.001). Concurrent adult distress was also independently associated with healthy lifestyle, in fully adjusted models, at both age 33 (β = −0.06, SE = 0.005, p<.001) and 42 (β = −0.06, SE = 0.01, p<.001). Analysis of Change in Lifestyle Of those who had outcome measures at both time points (N = 7998), 8.6% started unhealthy and declined, 18.8% started unhealthy and remained constant, 18.2% started unhealthy and improved, 27.1% started healthy and declined, 20.2% started healthy and remained constant, and 7.2% started healthy and improved. Fully adjusted generalized linear models revealed significant between group differences in psychological distress in childhood (F = 12.31, p<.001). Distress Z-scores were highest among those who started unhealthy and declined (mean = 0.12) and lowest among those who started healthy and improved (mean = −0.28). Those who started unhealthy and declined had significantly higher distress than those in all other groups, except for those who started unhealthy and remained unhealthy (Fig. 2). Fig. 2. View largeDownload slide Childhood psychological distress Z-scores by category of lifestyle change between ages 33 and 42: unhealthy and declined (UD), unhealthy constant (UC), unhealthy but improved (UI), healthy but declined (HD), healthy constant (HC), and healthy and improved (HI). Fig. 2. View largeDownload slide Childhood psychological distress Z-scores by category of lifestyle change between ages 33 and 42: unhealthy and declined (UD), unhealthy constant (UC), unhealthy but improved (UI), healthy but declined (HD), healthy constant (HC), and healthy and improved (HI). Analysis of Individual Lifestyle Components The overall child distress score was associated with almost all the individual components of a healthy lifestyle, at both ages 33 and 42 (Table 4). Apart from regular physical activity, a standard deviation increase in distress was associated with lower odds of endorsing each of the healthy lifestyle components. The largest reduction in odds was observed for smoking. Each 1 SD increase in childhood distress score corresponded to about a 30% decrease in the odds of being a nonsmoker (at both ages). Child distress was associated with slightly higher odds of reporting regular physical activity at ages 33 and 42 years. Table 4 Fully Adjusted Logistic Regression of Individual Dichotomous Healthy Lifestyle Components on Child Psychological Distress Age 33 (Wave 5) Age 42 (Wave 6) N = 10,748 N = 9,581 OR (95% CI) p-value OR (95% CI) p-value Healthy lifestyle componentsa Nonsmoker 0.71 (0.69–0.74) <.001 0.70 (0.68–0.73) <.001 Moderate alcohol consumption 0.96 (0.92–0.99) .009 0.92 (0.89–0.95) <.001 Regular physical activity 1.04 (1.01–1.08) .03 1.05 (1.01–1.09) .007 Healthy diet 0.90 (0.87–0.93) <.001 0.88 (0.85–0.91) <.001 Ideal body weight (body mass index <25) 0.96 (0.93–0.99) .009 0.94 (0.91–0.96) <.001 Age 33 (Wave 5) Age 42 (Wave 6) N = 10,748 N = 9,581 OR (95% CI) p-value OR (95% CI) p-value Healthy lifestyle componentsa Nonsmoker 0.71 (0.69–0.74) <.001 0.70 (0.68–0.73) <.001 Moderate alcohol consumption 0.96 (0.92–0.99) .009 0.92 (0.89–0.95) <.001 Regular physical activity 1.04 (1.01–1.08) .03 1.05 (1.01–1.09) .007 Healthy diet 0.90 (0.87–0.93) <.001 0.88 (0.85–0.91) <.001 Ideal body weight (body mass index <25) 0.96 (0.93–0.99) .009 0.94 (0.91–0.96) <.001 aEach healthy lifestyle component is dichotomized (0/1, with 1 indicating presence of healthy levels of the component). Each healthy lifestyle component is modeled separately, at both ages 33 and 42 years. Cell entries are odds ratios (ORs) and 95% confidence intervals (CIs). ORs show the odds of endorsing each healthy lifestyle component associated with a 1-SD increase in distress score. ORs < 1 indicate lower odds of having the health component. All models adjust for all child and family covariates. View Large Table 4 Fully Adjusted Logistic Regression of Individual Dichotomous Healthy Lifestyle Components on Child Psychological Distress Age 33 (Wave 5) Age 42 (Wave 6) N = 10,748 N = 9,581 OR (95% CI) p-value OR (95% CI) p-value Healthy lifestyle componentsa Nonsmoker 0.71 (0.69–0.74) <.001 0.70 (0.68–0.73) <.001 Moderate alcohol consumption 0.96 (0.92–0.99) .009 0.92 (0.89–0.95) <.001 Regular physical activity 1.04 (1.01–1.08) .03 1.05 (1.01–1.09) .007 Healthy diet 0.90 (0.87–0.93) <.001 0.88 (0.85–0.91) <.001 Ideal body weight (body mass index <25) 0.96 (0.93–0.99) .009 0.94 (0.91–0.96) <.001 Age 33 (Wave 5) Age 42 (Wave 6) N = 10,748 N = 9,581 OR (95% CI) p-value OR (95% CI) p-value Healthy lifestyle componentsa Nonsmoker 0.71 (0.69–0.74) <.001 0.70 (0.68–0.73) <.001 Moderate alcohol consumption 0.96 (0.92–0.99) .009 0.92 (0.89–0.95) <.001 Regular physical activity 1.04 (1.01–1.08) .03 1.05 (1.01–1.09) .007 Healthy diet 0.90 (0.87–0.93) <.001 0.88 (0.85–0.91) <.001 Ideal body weight (body mass index <25) 0.96 (0.93–0.99) .009 0.94 (0.91–0.96) <.001 aEach healthy lifestyle component is dichotomized (0/1, with 1 indicating presence of healthy levels of the component). Each healthy lifestyle component is modeled separately, at both ages 33 and 42 years. Cell entries are odds ratios (ORs) and 95% confidence intervals (CIs). ORs show the odds of endorsing each healthy lifestyle component associated with a 1-SD increase in distress score. ORs < 1 indicate lower odds of having the health component. All models adjust for all child and family covariates. View Large Discussion Using prospective data from the 1958 British Birth Cohort Study, we found that individuals with higher psychological distress between ages 7 and 16 years were less likely to maintain a healthy lifestyle at both ages 33 and 42 years. Greater distress in childhood was associated with less healthy lifestyle, even when controlling for concurrent adult distress. Childhood distress was also associated with deterioration in lifestyle between ages 33 and 42, suggesting that the effects of childhood distress may continue to compound over the life course with potentially substantial impacts on CVD development. Despite evidence of correlations between psychological distress and individual health behaviors, surprisingly little research has investigated these relationships prospectively or gone beyond considering a single behavior to evaluate a set of factors that comprise an overall healthy lifestyle. This is the first study to our knowledge to explore the prospective association between childhood psychological distress and an overall healthy lifestyle in adulthood. Our findings are consistent with those from other community-based longitudinal studies, reporting that aspects of child distress predict unhealthy behaviors later in adolescence or adulthood, such as cigarette smoking and/or problematic alcohol use [18, 19]. In the current study, when exploring the individual components of a healthy lifestyle as separate outcomes we also found that, with the exception of regular physical activity, more childhood distress was associated with lower odds of being healthy on each lifestyle component. The physical activity finding is somewhat surprising given prior evidence, among adults, of a relationship between depression and greater likelihood of sedentary lifestyle [51]. However, the impact of childhood distress on adult physical activity may depend on the specific aspect and manifestation of childhood distress. For example, individuals experiencing attention-deficit hyperactivity symptoms may exercise as a way to alleviate those symptoms [52]. The size of associations in this study was relatively small. However, determinants of health behaviors are likely multifactorial, and any one variable will likely contribute only modestly to such outcomes. Given the robust link between overall healthy lifestyle and reduced risk of chronic disease [1–7, 10, 11], even small changes in healthy lifestyle, associated with childhood distress, may be clinically relevant. For example, results from analyses using a similar measure of healthy lifestyle among women in the Nurses Health Study indicate a 16.20% increased risk of stroke, on average, for each 1-point decrease in lifestyle ([3], [53]). Assuming a comparable sample, the 0.5-point decrease in healthy lifestyle score assuming a comparable sample, would reflect an approximate 8% increase in stroke risk. High levels of distress in childhood could plausibly be linked with less healthy lifestyle components in adulthood in a number of ways. For example, smoking, drinking, or “comfort eating” may be used to self-medicate or cope with psychological distress or affective symptoms [54–56]. Distress in childhood could lead to impaired social relationships or alienation from mainstream peers, fostering affiliation with more “deviant” peer groups, which may influence substance use behaviors [21]. Distress in childhood may impair educational or professional attainment [57], thereby limiting resources to access healthy food or safe exercise spaces. Distress may also contribute to poorer self-regulatory capacity, due to demands on self-control resources, which is associated with physical activity, smoking, alcohol consumption, and eating behaviors [58]. These and other potential mechanisms should be explored further in future research. Moreover, as the distribution of child psychological distress in the population is not random, attention should also be given to its determinants. Though some aspects of child psychological functioning (e.g., hyperactivity) may have a genetic component [59], a great deal of research has suggested that child and adolescent psychopathology is strongly shaped by early-life experiences, with adversity [60], social stress [61], and lower socioeconomic status [62] each contributing to greater likelihood of high levels of distress. Other research suggests that childhood mental health is modifiable, with prevention and intervention strategies targeting children, caregivers, and schools [63–67]. The current study has some limitations. Despite having measured distress prospectively, we cannot completely rule out the possibility of reverse causation, given many unhealthy behaviors have their origins in adolescence and some in childhood [13]. Indeed, prior research suggests that the relationships between mental health and health behaviors is likely bidirectional, although these effects may be more apparent as individuals enter into late adolescence and young adulthood [68]. Although we adjusted for many potential confounders, there may be others that were unmeasured, including genetic variation [69] and parental factors such as child-rearing style [70] and psychopathology [71]. Another limitation is that the available measures for several behaviors were somewhat constrained. For example, we had information only on frequency of consumption of various foods but not portion size. However, other work has suggested that the consumption frequency explains most of the variation in food intake [43]. While we may have lost some information by dichotomizing the healthy lifestyle components, this approach allowed us to condense information into a meaningful index and is consistent with measures of healthy lifestyle that have previously been shown to predict disease risk. For other ways to operationalize health behavior composites, and methodological considerations regarding combining information across multiple health behaviors, see the study by Prochaska et al. [9]. Finally, like many longitudinal studies, the 1958 British Birth Cohort Study is limited by participant attrition; however, the remaining samples at ages 33 and 42 were fairly representative of the original cohort, and our use of inverse probability weights in analyses reduces biases due to attrition. This study also has a number of important strengths. It is one of the first studies to measure psychological distress in childhood and consider it in relation to a set of subsequent health-related behaviors over a 17- to 35-year follow-up period. Childhood distress was reported by teachers when children were ages 7, 11, and 16 years, alleviating concerns that the participants’ adult health status or behavior could bias their memory or reporting of early symptoms; a potential concern with retrospective designs. Furthermore, we could evaluate and account for a broad range of potential confounders. With increased appreciation for the childhood origins of adult CVD [72], and the growing interest in primordial prevention [14], it is essential to evaluate whether and what factors in childhood may provide insight into whether individuals will initiate and maintain key healthy lifestyle practices. The small percentage of participants with all five components of a healthy lifestyle in young and mid adulthood (3.8% at age 33 and 2.8% at age 42) highlights the importance of identifying opportunities to promote a healthy lifestyle; primordial prevention efforts may benefit from greater consideration of the role of childhood mental health in setting trajectories of healthy lifestyle across the life course. Our findings suggest that even at a relatively young age, psychological distress may signal increased risk for engaging in an array of less healthy lifestyle practices later in life. Critical future steps will be to evaluate whether improving child mental health indeed improves subsequent health risk factors in adulthood and correspondingly decreases risk for CVD. CVD primordial prevention efforts would benefit from consideration of the child psychological distress and the role it plays in relation to known/traditional risk factors for CVD. Elucidating these mechanisms of risk may provide important targets for prevention efforts in children, which may have long-lasting effects on adult cardiovascular health. Supplementary Material Supplementary material is available at Annals of Behavioral Medicine online. Acknowledgments Dr. Winning was supported by the Julius B. Richmond Fellowship at the Harvard Center on the Developing Child and by the Martha May Eliot Fund at the Harvard T. H. Chan School of Public Health, and Dr. Gilsanz was supported by the Yerby Postdoctoral Fellowship; however, no direct funding was received or set aside for the writing of this paper. Compliance with Ethical Standards Authors’ Statement of Conflict of Interest and Adherence to Ethical Standards The authors declare that they have no conflict of interest. Primary Data: We performed secondary data analysis of data from the 1958 British Birth Cohort Study. Authors' Contributions: All authors contributed to the ideas in this paper. AW led the writing and analyses, and all authors contributed to drafts and approved the final version. Ethical approval: All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Informed consent: Informed consent was obtained from all individual participants included in the 1958 British Birth Cohort study. References 1. Stampfer MJ , Hu FB , Manson JE , Rimm EB , Willett WC . Primary prevention of coronary heart disease in women through diet and lifestyle . N Engl J Med . 2000 ; 343 : 16 – 22 . Google Scholar CrossRef Search ADS PubMed 2. Chiuve SE , McCullough ML , Sacks FM , Rimm EB . Healthy lifestyle factors in the primary prevention of coronary heart disease among men: Benefits among users and nonusers of lipid-lowering and antihypertensive medications . Circulation . 2006 ; 114 : 160 – 167 . Google Scholar CrossRef Search ADS PubMed 3. Chiuve SE , Rexrode KM , Spiegelman D et al. Primary prevention of stroke by healthy lifestyle . Circulation . 2008 ; 118 : 947 – 954 . Google Scholar CrossRef Search ADS PubMed 4. Hu FB , Manson JE , Stampfer MJ et al. Diet, lifestyle, and the risk of type 2 diabetes mellitus in women . N Engl J Med . 2001 ; 345 : 790 – 797 . Google Scholar CrossRef Search ADS PubMed 5. Platz EA , Willett WC , Colditz GA et al. Proportion of colon cancer risk that might be preventable in a cohort of middle-aged US men . Cancer Causes Control . 2000 ; 11 : 579 – 588 . Google Scholar CrossRef Search ADS PubMed 6. Dartois L , Fagherazzi G , Boutron-Ruault MC , Mesrine S , Clavel-Chapelon F . Association between five lifestyle habits and cancer risk: Results from the E3N cohort . Cancer Prevent Res . 2014 ; 7 : 516 – 525 . Google Scholar CrossRef Search ADS 7. Knoops KT , de Groot LC , Kromhout D et al. Mediterranean diet, lifestyle factors, and 10-year mortality in elderly European men and women: The HALE project . JAMA . 2004 ; 292 : 1433 – 1439 . Google Scholar CrossRef Search ADS PubMed 8. Shaw BA , Agahi N . A prospective cohort study of health behavior profiles after age 50 and mortality risk . BMC Public Health . 2012 ; 12 : 803 . Google Scholar CrossRef Search ADS PubMed 9. Prochaska JJ , Spring B , Nigg CR . Multiple health behavior change research: An introduction and overview . Prev Med . 2008 ; 46 : 181 – 188 . Google Scholar CrossRef Search ADS PubMed 10. Liu K , Daviglus ML , Loria CM et al. Healthy lifestyle through young adulthood and the presence of low cardiovascular disease risk profile in middle age: The Coronary Artery Risk Development in (Young) Adults (CARDIA) study . Circulation . 2012 ; 125 : 996 – 1004 . Google Scholar CrossRef Search ADS PubMed 11. Chomistek AK , Chiuve SE , Eliassen AH et al. Healthy lifestyle in the primordial prevention of cardiovascular disease among young women . J Am Coll Cardiol . 2015 ; 65 : 43 – 51 . Google Scholar CrossRef Search ADS PubMed 12. Kuh D , Ben Shlomo Y. A life course approach to chronic disease epidemiology: Tracing the origins of ill-health from early to adult life . Oxford : Oxford University Press ; 1997 . 13. Lloyd-Jones DM , Hong Y , Labarthe D et al. Defining and setting national goals for cardiovascular health promotion and disease reduction: The American Heart Association’s strategic Impact Goal through 2020 and beyond . Circulation . 2010 ; 121 : 586 – 613 . Google Scholar CrossRef Search ADS PubMed 14. Lloyd-Jones DM . Cardiovascular health and protection against CVD. More than the sum of the parts ? Circulation . 2014 ; 130 : 1671 – 1673 . Google Scholar CrossRef Search ADS PubMed 15. Appleton AA , Buka SL , McCormick MC et al. Emotional functioning at age 7 years is associated with C-reactive protein in middle adulthood . Psychosom Med . 2011 ; 73 : 295 – 303 . Google Scholar CrossRef Search ADS PubMed 16. Goodwin RD , Sourander A , Duarte CS et al. Do mental health problems in childhood predict chronic physical conditions among males in early adulthood? Evidence from a community-based prospective study . Psychol Med . 2009 ; 39 : 301 – 311 . Google Scholar CrossRef Search ADS PubMed 17. Winning A , Glymour MM , McCormick MC , Gilsanz P , Kubzansky LD . Psychological distress across the life course and cardiometabolic risk: Findings from the 1958 British Birth Cohort Study . J Am Coll Cardiol . 2015 ; 66 : 1577 – 1586 . Google Scholar CrossRef Search ADS PubMed 18. Charach A , Yeung E , Climans T , Lillie E . Childhood attention-deficit/hyperactivity disorder and future substance use disorders: Comparative meta-analyses . J Am Acad Child Adolesc Psychiatry . 2011 ; 50 : 9 – 21 . Google Scholar CrossRef Search ADS PubMed 19. Marmorstein NR , White HR , Loeber R , Stouthamer-Loeber M . Anxiety as a predictor of age at first use of substances and progression to substance use problems among boys . J Abnorm Child Psychol . 2010 ; 38 : 211 – 224 . Google Scholar CrossRef Search ADS PubMed 20. von Stumm S , Deary IJ , Kivimaki M et al. Childhood behavior problems and health at midlife: 35-year follow-up of a Scottish birth cohort . J Child Psychol Psychiatry . 2011 ; 52 : 992 – 1001 . Google Scholar CrossRef Search ADS PubMed 21. Hayatbakhsh R , Mamun AA , Williams GM , O’Callaghan MJ , Najman JM . Early childhood predictors of early onset of smoking: A birth prospective study . Addictive Behaviors . 2013 ; 38 : 2513 – 2519 . Google Scholar CrossRef Search ADS PubMed 22. Nomura Y , Gilman SE , Buka SL . Maternal smoking during pregnancy and risk of alcohol use disorders among adult offspring . J Stud Alcohol Drugs . 2011 ; 72 : 199 – 209 . Google Scholar CrossRef Search ADS PubMed 23. Power C , Jefferis BJ . Fetal environment and subsequent obesity: A study of maternal smoking . Int J Epidemiol . 2002 ; 31 : 413 – 419 . Google Scholar CrossRef Search ADS PubMed 24. Grzywacz JG , Marks NF . Social inequalities and exercise during adulthood: Toward an ecological perspective . J Health Soc Behav . 2001 ; 42 : 202 – 220 . Google Scholar CrossRef Search ADS PubMed 25. Snead MC , Cockerham WC . Health lifestyles and social class in the deep South . In Kronenfeld JJ , (ed), Social inequalities, health and health care delivery . Vol. 20 . Bingley, UK: Emerald Group Publishing Limited ; 2002 : 107 – 122 . Google Scholar CrossRef Search ADS 26. Melchior M , Moffitt TE , Milne BJ , Poulton R , Caspi A . Why do children from socioeconomically disadvantaged families suffer from poor health when they reach adulthood? A life-course study . Am J Epidemiol . 2007 ; 166 : 966 – 974 . Google Scholar CrossRef Search ADS PubMed 27. El Marroun H , Schmidt MN , Franken IH et al. Prenatal tobacco exposure and brain morphology: A prospective study in young children . Neuropsychopharmacology . 2014 ; 39 : 792 – 800 . Google Scholar CrossRef Search ADS PubMed 28. Ross CE , Bird CE . Sex stratification and health lifestyle: Consequences for men’s and women’s perceived health . J Health Soc Behav . 1994 ; 35 : 161 – 178 . Google Scholar CrossRef Search ADS PubMed 29. Singh GK , Kenney MK , Ghandour RM , Kogan MD , Lu MC . Mental health outcomes in US children and adolescents born prematurely or with low birthweight . Depress Res treat . 2013 ; 2013 : 570743 . Google Scholar PubMed 30. Power C , Elliott J . Cohort profile: 1958 British birth cohort (National Child Development Study) . Int J Epidemiol . 2006 ; 35 : 34 – 41 . Google Scholar CrossRef Search ADS PubMed 31. Stott DH. The social-adjustment of children: Manual to the bristol social-adjustment guides . London, UK: University of London Press ; 1963 . 32. Rutter M . A children’s behaviour questionnaire for completion by teachers: Preliminary findings . J Child Psychol Psychiatry Allied Disciplines . 1967 ; 8 : 1 – 11 . Google Scholar CrossRef Search ADS 33. Shepherd P. Bristol social adjustment guides at 7 and 11 years . London, UK : Centre for Longitudinal Studies, Institute of Education, University of London , 2013 . 34. Tuddenham RD . Review of Bristol social adjustment guides . J Consult Psychol . 1960 ; 24 : 99 . Google Scholar CrossRef Search ADS 35. Cable N , Bartley M , McMunn A , Kelly Y . Gender differences in the effect of breastfeeding on adult psychological well-being . Eur J Public Health . 2012 ; 22 : 653 – 658 . Google Scholar CrossRef Search ADS PubMed 36. Clark C , Rodgers B , Caldwell T , Power C , Stansfeld S . Childhood and adulthood psychological ill health as predictors of midlife affective and anxiety disorders: The 1958 British Birth Cohort . Arch Gen Psychiatry 2007 ; 64 : 668 – 678 . Google Scholar CrossRef Search ADS PubMed 37. Sourander A , Jensen P , Davies M et al. Who is at greatest risk of adverse long-term outcomes? The Finnish From a Boy to a Man study . J Am Acad Child Adolesc Psychiatry . 2007 ; 46 : 1148 – 1161 . Google Scholar CrossRef Search ADS PubMed 38. Jokela M , Ferrie J , Kivimaki M . Childhood problem behaviors and death by midlife: The British National Child Development Study . J Am Acad Child Adolesc Psychiatry . 2009 ; 48 : 19 – 24 . Google Scholar CrossRef Search ADS PubMed 39. Lilienfeld SO . Comorbidity between and within childhood externalizing and internalizing disorders: Reflections and directions . J Abnorm Child Psychol . 2003 ; 31 : 285 – 291 . Google Scholar CrossRef Search ADS PubMed 40. Kitchens JM . Does this patient have an alcohol problem ? JAMA . 1994 ; 272 : 1782 – 1787 . Google Scholar CrossRef Search ADS PubMed 41. Mukamal KJ , Chiuve SE , Rimm EB . Alcohol consumption and risk for coronary heart disease in men with healthy lifestyles . Arch Intern Med . 2006 ; 166 : 2145 – 2150 . Google Scholar CrossRef Search ADS PubMed 42. Parsons TJ , Power C , Manor O . Longitudinal physical activity and diet patterns in the 1958 British Birth Cohort . Med Sci Sports Exerc . 2006 ; 38 : 547 – 554 . Google Scholar CrossRef Search ADS PubMed 43. Parsons TJ , Manor O , Power C . Changes in diet and physical activity in the 1990s in a large British sample (1958 birth cohort) . Eur J Clin Nutr . 2005 ; 59 : 49 – 56 . Google Scholar CrossRef Search ADS PubMed 44. Pringle M , Butler N , Davie R. 11,000 Seven-year olds: First report of the National Child Development Study (1958 Cohort) . London : Longman , 1967 . 45. Southgate V. Southgate group reading tests . London : University of London , 1962 . 46. Atherton K , Fuller E , Shepherd P , Strachan DP , Power C . Loss and representativeness in a biomedical survey at age 45 years: 1958 British birth cohort . J Epidemiol Community Health . 2008 ; 62 : 216 – 223 . Google Scholar CrossRef Search ADS PubMed 47. Cole TJ , Bellizzi MC , Flegal KM , Dietz WH . Establishing a standard definition for child overweight and obesity worldwide: International survey . BMJ . 2000 ; 320 : 1240 – 1243 . Google Scholar CrossRef Search ADS PubMed 48. Rutter M , Tizard J , Whitmore K. Education, health and behaviour . London : Longmans ; 1970 . 49. Gilman SE , Rende R , Boergers J et al. Parental smoking and adolescent smoking initiation: An intergenerational perspective on tobacco control . Pediatrics . 2009 ; 123 : e274 – e281 . Google Scholar CrossRef Search ADS PubMed 50. Hanson MD , Chen E . Socioeconomic status and health behaviors in adolescence: A review of the literature . J Behav Med . 2007 ; 30 : 263 – 285 . Google Scholar CrossRef Search ADS PubMed 51. Roshanaei-Moghaddam B , Katon WJ , Russo J . The longitudinal effects of depression on physical activity . Gen Hosp Psychiatry . 2009 ; 31 : 306 – 315 . Google Scholar CrossRef Search ADS PubMed 52. Gapin JI , Labban JD , Etnier JL . The effects of physical activity on attention deficit hyperactivity disorder symptoms: The evidence . Prev Med . 2011 ; 52 ( Suppl ): S70 – S74 . Google Scholar CrossRef Search ADS PubMed 53. Trudel-Fitzgerald C , Tworoger SS , Poole EM , Williams DR , Kubzansky LD . Prospective changes in healthy lifestyle among midlife women: When psychological symptoms get in the way . Am J Prev Med . 2016 ; 51 : 327 – 335 . Google Scholar CrossRef Search ADS PubMed 54. Gibson EL . The psychobiology of comfort eating: Implications for neuropharmacological interventions . Behav Pharmacol . 2012 ; 23 : 442 – 460 . Google Scholar CrossRef Search ADS PubMed 55. Gehricke JG , Loughlin SE , Whalen CK et al. Smoking to self-medicate attentional and emotional dysfunctions . Nicotine Tob Res . 2007 ; 9 ( Suppl 4 ): S523 – S536 . Google Scholar CrossRef Search ADS PubMed 56. Bolton JM , Robinson J , Sareen J . Self-medication of mood disorders with alcohol and drugs in the National Epidemiologic Survey on Alcohol and Related Conditions . J Affect Disord . 2009 ; 115 : 367 – 375 . Google Scholar CrossRef Search ADS PubMed 57. Colman I , Murray J , Abbott RA et al. Outcomes of conduct problems in adolescence: 40 year follow-up of national cohort . BMJ . 2009 ; 338 : a2981 . Google Scholar CrossRef Search ADS PubMed 58. Hagger MS , Wood C , Stiff C , Chatzisarantis NLD . The strength model of self-regulation failure and health-related behaviour . Health Psychol Rev . 2009 ; 3 : 208 – 238 . Google Scholar CrossRef Search ADS 59. McGue M , Bouchard TJ , Jr . Genetic and environmental influences on human behavioral differences . Ann Rev Neurosci . 1998 ; 21 : 1 – 24 . Google Scholar CrossRef Search ADS PubMed 60. Clark C , Caldwell T , Power C , Stansfeld SA . Does the influence of childhood adversity on psychopathology persist across the lifecourse? A 45-year prospective epidemiologic study . Ann Epidemiol . 2010 ; 20 : 385 – 394 . Google Scholar CrossRef Search ADS PubMed 61. Grant KE , Compas BE , Thurm AE , McMahon SD , Gipson PY . Stressors and child and adolescent psychopathology: Measurement issues and prospective effects . J Clin Child and Adolesc Psychol . 2004 ; 33 : 412 – 425 . Google Scholar CrossRef Search ADS 62. Reiss F . Socioeconomic inequalities and mental health problems in children and adolescents: A systematic review . Soc Sci Med . 2013 ; 90 : 24 – 31 . Google Scholar CrossRef Search ADS PubMed 63. O’Leary-Barrett M , Topper L , Al-Khudhairy N et al. Two-year impact of personality-targeted, teacher-delivered interventions on youth internalizing and externalizing problems: A cluster-randomized trial . J Am Acad Child Adolesc Psychiatry . 2013 ; 52 : 911 – 920 . Google Scholar CrossRef Search ADS PubMed 64. Waddell C , Hua JM , Garland OM , Peters RD , McEwan K . Preventing mental disorders in children: A systematic review to inform policy-making . Can J Public Health . 2007 ; 98 : 166 – 173 . Google Scholar PubMed 65. Weare K , Nind M . Mental health promotion and problem prevention in schools: What does the evidence say ? Health Promot Int . 2011 ; 26 ( Suppl 1 ): i29 – i69 . Google Scholar CrossRef Search ADS PubMed 66. Stewart-Brown SL , Schrader-McMillan A . Parenting for mental health: What does the evidence say we need to do? Report of Workpackage 2 of the DataPrev project . Health Promot Int . 2011 ; 26 ( Suppl 1 ): i10 – i28 . Google Scholar CrossRef Search ADS PubMed 67. Durlak JA , Wells AM . Primary prevention mental health programs for children and adolescents: A meta-analytic review . Am J Community Psychol . 1997 ; 25 : 115 – 152 . Google Scholar CrossRef Search ADS PubMed 68. Chaiton MO , Cohen JE , O’Loughlin J , Rehm J . A systematic review of longitudinal studies on the association between depression and smoking in adolescents . BMC Public Health . 2009 ; 9 : 356 . Google Scholar CrossRef Search ADS PubMed 69. Wang JC , Hinrichs AL , Stock H et al. Evidence of common and specific genetic effects: Association of the muscarinic acetylcholine receptor M2 (CHRM2) gene with alcohol dependence and major depressive syndrome . Hum Mol Gen . 2004 ; 13 : 1903 – 1911 . Google Scholar CrossRef Search ADS PubMed 70. Stavrakakis N , de Jonge P , Ormel J , Oldehinkel AJ . Bidirectional prospective associations between physical activity and depressive symptoms. The TRAILS Study . J Adolesc Health . 2012 ; 50 : 503 – 508 . Google Scholar CrossRef Search ADS PubMed 71. Humphreys KL , Mehta N , Lee SS . Association of parental ADHD and depression with externalizing and internalizing dimensions of child psychopathology . J Atten Disord . 2012 ; 16 : 267 – 275 . Google Scholar CrossRef Search ADS PubMed 72. Shonkoff JP , Garner AS . The lifelong effects of early childhood adversity and toxic stress . Pediatrics . 2012 ; 129 : e232 – e246 . Google Scholar CrossRef Search ADS PubMed © Society of Behavioral Medicine 2018. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Annals of Behavioral Medicine Oxford University Press

Childhood Psychological Distress and Healthy Cardiovascular Lifestyle 17–35 Years Later: The Potential Role of Mental Health in Primordial Prevention

Loading next page...
 
/lp/ou_press/childhood-psychological-distress-and-healthy-cardiovascular-lifestyle-oOU9wXGK0Y
Copyright
© Society of Behavioral Medicine 2018. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
ISSN
0883-6612
eISSN
1532-4796
D.O.I.
10.1093/abm/kax001
Publisher site
See Article on Publisher Site

Abstract

Abstract Background Maintaining a healthy lifestyle in adulthood has been shown to significantly reduce cardiovascular disease risk. Increasing evidence suggests that behavioral risk factors for cardiovascular disease are established in childhood; however, limited research has evaluated whether childhood psychological factors play a role. Purpose To evaluate the association between childhood psychological distress and young to mid adulthood healthy lifestyle. Methods Using prospective data from the 1958 British Birth Cohort, we assessed whether psychological distress in childhood (captured by internalizing and externalizing symptoms at ages 7, 11, and 16 years) predicted healthy lifestyle at ages 33 (N = 10,748) and 42 (N = 9,581) years. Healthy lifestyle was measured using an index previously demonstrated to predict cardiovascular disease, consisting of five components: absence of smoking, moderate alcohol consumption, regular physical activity, healthy diet, and ideal body weight. Results Few participants (3.8% at age 33 years and 2.8% at age 42 years) endorsed all five healthy lifestyle components. Linear regression models, adjusting for potential child- and family-level confounders, revealed that higher distress levels in childhood were negatively associated with healthy lifestyle at age 33 years (β = −0.11, SE = 0.01, p < .001) and 42 years (β = −0.13, SE = 0.01, p < .001). Higher distress was also associated with significantly lower odds of endorsing each lifestyle component, except physical activity, at both ages. Additional analyses indicated that childhood distress levels were highest among those whose lifestyle scores were low at age 33 and further declined between ages 33 and 42. Conclusions Psychological distress in childhood may indicate children at risk of less healthy lifestyle practices later in life. Although our findings are preliminary, psychological distress may also provide an important target for public health interventions aimed at preventing cardiovascular disease. Psychological distress, Lifestyle, Prospective cohort study, Epidemiology, Cardiovascular disease prevention Introduction Maintaining a healthy lifestyle reliably reduces risk of chronic disease, including coronary heart disease (CHD) [1, 2], stroke [3], diabetes [4], and cancer [5, 6], as well as all-cause mortality [7]. Clusters of lifestyle factors have a multiplicative impact on mortality compared with individual behaviors [8], and growing evidence suggests greater potential for public health impact of multiple- versus single-behavior interventions [9]. Thus, researchers are increasingly clustering multiple components into an overall “healthy lifestyle” index, which typically includes five components: absence of smoking, moderate alcohol consumption, regular physical activity, healthy diet, and body mass index (BMI) under 25 kg/m2 [1, 2, 10]. Particular attention has been paid to the impact of a healthy lifestyle on cardiovascular disease (CVD) outcomes. For example, researchers followed 84,129 initially disease-free women from the Nurses’ Health Study for 14 years and found that women with three, four, or five components of a healthy lifestyle (defined as above), respectively, had 57%, 66%, and 83% lower risk for incident CHD compared with all other women [1]. A similar graded relationship was found among 42,847 men in the Health Professionals Follow-up Study [2]. Furthermore, this study showed that men who adopted at least two additional healthy lifestyle factors during follow-up, compared to those who did not make any lifestyle changes, had a 27% reduction in CHD risk [2]. A strong reduction in risk has also been demonstrated for stroke, with one study reporting that over 50% of the ischemic stroke cases in each of two prospective cohorts were attributable to lack of adherence to a healthy lifestyle (less than all five health behaviors) [3]. While most cardiovascular and biomedical research to date has considered healthy lifestyle as a predictor of reduced risk [1–3, 7, 10, 11], less has considered the early-life determinants of healthy lifestyle as an outcome. With recognition that childhood factors set the stage for adulthood cardiovascular health [12], CVD primordial prevention—that is, prevention of disease risk factors before they begin to develop—may benefit greatly from identifying factors in childhood that offer some ability to predict who will adopt and maintain healthier lifestyle practices and therefore have better cardiovascular health as adults [13, 14]. Identification of such factors will enable both targeted surveillance and strategic intervention well before disease processes are initiated. One potential early-life contributor to a healthy adult lifestyle is childhood psychological distress. Psychological distress in childhood has been linked with a number of important cardiovascular health- related outcomes in adulthood, including increased inflammation [15], obesity [16], and cardiometabolic risk [17]. Although several lines of evidence suggest that better childhood mental health may lead to increased likelihood of engaging in healthier behaviors, most research looks at individual unhealthy behaviors as the end point, as opposed to an overall healthy lifestyle. Furthermore, nearly all the research linking mental health to behaviors has been conducted with adult populations and is cross-sectional in nature, so directions of effect cannot be established. Some prospective research has found that specific childhood disorders (e.g., anxiety, conduct disorder) are associated with certain cardiovascular health-related factors, such as cigarette smoking, problem drinking, and obesity [18–20]. However, to the best of our knowledge, no previous research has investigated the prospective relationship between a spectrum of child psychological distress (including levels that would not met criteria for any disorder) and an overall healthy lifestyle (as captured by multiple health components). Child psychological distress may provide an early indicator of risk for less healthy lifestyle and subsequently poorer health in adulthood. Using data from the 1958 British Birth Cohort Study, we evaluated whether psychological distress in childhood was associated with an overall healthy cardiovascular lifestyle 17–35 years later in young to mid adulthood. We hypothesized that children with higher levels of distress would have be less likely to engage in healthy behaviors in adulthood, even when adjusting for potential child- and family-level confounders of the relationship. We selected covariates based on previous evidence of their association with psychological distress and (at least one component of) a healthy lifestyle. For example, maternal education and marital circumstances have been shown to predict earlier onset of smoking in offspring [21]. Additionally, children of mothers who smoked during pregnancy are more likely, as adults, to have alcohol use disorders [22] and be obese [23]. Lower social class has also been found to predict less healthy lifestyles [24, 25]. These family factors may also be associated with child psychological distress [26, 27]. In addition, both childhood distress and adult health behaviors have been linked with characteristics of the child, including sex, low birth weight (LBW), academic performance, BMI, and physical health status [28, 29]. Thus, the present study included consideration of these relevant characteristics of the child and the child’s family. Methods Sample Population Data come from the 1958 British Birth Cohort Study (also known as the NCDS) [30]—an ongoing longitudinal study of people born in Great Britain during a single week in March 1958 (N = 18,558). Information on various aspects of the cohort members’ physical and mental health was collected at birth and during six follow-up waves between 1964 and 2000 (at ages 7, 11, 16, 23, 33, and 42 years). Informed consent was obtained from all individual participants included in the study. The cohort is described in detail elsewhere [30]. Outcomes for this study were derived from measurements at Wave 5 (age 33) and Wave 6 (age 42). At Wave 5, 11,469 of the original cohort members participated in the study. Participants missing data on any of the five healthy lifestyle components at age 33 years were excluded from the age 33 analyses (N = 588). We excluded an additional 133 people who did not have any measures of psychological distress in childhood, leaving an analytic sample of 10,748 at age 33 (Wave 5). At Wave 6 (age 42), 11,419 of the original cohort members participated in the study. Most (N = 9,890) of the Wave 6 participants also participated in Wave 5. Participants missing data on any of the five healthy lifestyle components at age 42 years were excluded from the age 42 analyses (N = 1,723). The majority (N = 1,641) of those excluded were missing data on only one behavior (predominantly alcohol consumption, N = 1,531). We excluded an additional 115 people who did not have any measures of psychological distress in childhood, leaving an analytic sample of 9,581 at age 42 (Wave 6). (See Fig. 1 for a flow chart of sample selection and exclusions.) Most participants (N = 7,998) were in the analytic samples at both waves. Generally, as expected, those in the analytic samples tended to be healthier and better off than those excluded from the original birth cohort sample (Supplementary Table A). Fig. 1. View largeDownload slide Flow chart of sample selection and exclusions. Fig. 1. View largeDownload slide Flow chart of sample selection and exclusions. Measures Childhood psychological distress Childhood psychological distress was ascertained from validated measures of psychopathology and emotional/behavioral functioning: the Bristol Social Adjustment Guide (BSAG) [31] at ages 7 and 11 and the Rutter Behaviour Scale (RBS) [32] at age 16; all measures were completed by participants’ teachers. Items from the BSAG and the RBS were designed to capture levels of emotional disturbance and social maladjustment [33]. The overall score derived from the 146-item teacher-rated BSAG has been found to have good inter-rater reliability (r = .76) [34]. Similarly, the overall score derived from the 26-item teacher-rated version of the RBS has been shown to have good inter-rater reliability (r = .72) and test-retest reliability (r = .89) [32]. Within the National Child Development Study (NCDS) population, the BSAG scores at ages 7 and 11 and the RBS score at age 16 were all significantly correlated (Rs = 0.26–0.39; p-values all <.001). The two measures derived from the overall scores appear to be fairly comparable [35] and have each been shown to predict increased risk of adulthood psychiatric disorders [36]. Prior research has more specifically demonstrated two primary forms of childhood distress—externalizing and internalizing—associated with increased risk of adverse developmental and health outcomes [37, 38]. These two types of distress share some common characteristics, but they can also be distinguished from one another on the basis of distinctive specific features [39]. To ensure that each form of distress was adequately captured, and to be consistent with previous work with these measures in this cohort [36, 38], we first created separate scores for each type of distress at each age by summing across symptoms related to externalizing (e.g., restlessness, aggression) and those related to internalizing (e.g., worry, depression). If participants were missing at least half of either the externalizing or the internalizing items, we set their corresponding subscale score to missing; otherwise, for approximately 1% of the sample, we substituted missing values (typically only one scale item) with the mean of their completed items. We then created a continuous total distress score at each age in childhood (7, 11, and 16 years), by computing the mean of the standardized externalizing and internalizing scores. Internal consistency reliabilities of the total distress measures were high: Cronbach’s alphas = 0.79 at age 7, 0.76 at age 11, and 0.86 at age 16. Finally, we computed the mean score across the three time points to create the overall childhood psychological distress measure, where higher scores indicate greater psychological distress. For ease of interpretation, we standardized this summary score to have a mean of zero and a SD of 1. Healthy lifestyle index We created separate healthy lifestyle index scores at ages 33 and 42 years, based on criteria used in previous work demonstrated to predict disease outcomes [1–3, 10]. The healthy lifestyle index comprised five components, all self-reported: (i) absence of current cigarette smoking; (ii) moderate alcohol consumption (≤7 drinks/week for women, ≤14 for men; excluding abstention); (iii) regular physical activity (≥4–5 times/week); (iv) healthy diet (top 40% of diet score); and (v) ideal body weight (BMI <25 kg/m2). Following prior work that developed this measure, we created a dichotomous measure of each healthy lifestyle component (0/1, with 1 indicating presence of healthy levels of the component), and then, summed the components to create a continuous score. Thus, the healthy lifestyle index ranged from 0 to 5, with higher scores indicating healthier lifestyles. Smoking status was self-reported at ages 33 and 42. We assigned all current nonsmokers, including both never smokers and former smokers, a score of 1 on the absence of smoking component. We derived alcohol consumption from a self-reported measure of overall frequency of drinking (with response options ranging from never to most days) and from a series of self-report questions on the number of specific alcoholic beverages (e.g., beer, wine, spirits) consumed in the last 7 days. In addition, we assigned a problem drinking score based on endorsement of two or more out of four items on the CAGE questionnaire, a widely used and validated method for identifying alcoholism [40]. Women who reported drinking more than seven drinks in the past 7 days, men who reported drinking more than 14 drinks in the past 7 days, and all those who reported a past-year drinking problem, were given a score of 0. Consistent with several measures of healthy lifestyle (e.g., 1, 2, 3, 7, 11), those who reported never drinking were also given a score of 0, as abstention is not considered optimal with regard to cardiovascular health [41]. All others were considered to be moderate drinkers and, thus, were given a score of 1 on the moderate alcohol consumption component. If participants reported regularly engaging in any sport of leisure activity that involved physical exercise (e.g., sports, aerobics classes, weight training running, swimming cycling, walking, dancing) [42], they were asked a follow-up question about how frequently they took part in the activities. Six response options ranged from less than twice a month to every day. Anyone who reported regularly exercising at least 4 to 5 days a week was characterized as engaging in regular physical activity (score = 1). To capture quality of diet, we followed an approach previously used in this cohort [43]. Participants self-reported their average weekly frequency of consumption of a variety of food types, with response options ranging from never to more than once a day. We coded food consumption frequency on a scale from zero to five, giving higher scores to frequent “healthy” food consumption (fresh fruit and salad/raw vegetables) and lower scores to frequent “unhealthy” food consumption (chips, fried food, sweets/chocolates, and cakes/biscuits). We then summed the scores over the six food types to create a continuous measure of overall diet quality, with higher scores indicating better diet quality. Following previous literature on a healthy lifestyle [2, 3], we dichotomized the continuous diet score to define healthy diet as the top 40% of the healthy diet scores. We calculated BMI from weight and height measurements, which were self-reported at age 33 and 42 and obtained from a medical examination at age 45. If self-reported measurements were not available at age 42 (N = 951), we substituted weight and height measurements from the medical examination at age 45. BMI measures at ages 42 and 45 were highly correlated (r = .87, p<.001). Following prior work, we defined ideal body weight as BMI <25 kg/m2 (i.e., not overweight or obese) [3, 13]. While this categorization of ideal body weight includes underweight (BMI <18.5 kg/m2), being underweight has relatively few implications for overall cardiovascular health [13]. Less than 2.5% of our analytic samples were underweight. Covariates Child covariates consisted of sex (male, female); LBW (<5.5 pounds vs. ≥ = 5.5 pounds); cognition at age 7 (assessed using a using a 10-problem arithmetic test [44] and the Southgate Reading Test [45], respectively, with the lowest 10% of scores in both tests counting as low performance) [46]; overweight at age 7 (medical-examiner-recorded BMI reached or exceeded cut points laid out by the International Obesity Task Force: 17.92 kg/m2 for boys and 17.75 kg/m2 for girls [47]); and presence of any physical health problems by age 7 (any asthma, chronic illness, or signs of heart problems; yes/no). Family covariates consisted of father’s occupation at child’s birth (or age 7 if missing; manual vs. nonmanual), mother’s education level (left school at/before vs. after minimum leaving age), mother’s cigarette smoking during pregnancy (yes/no), and mother’s marital status at child’s birth (single/divorced/widowed vs. married). For all covariates, we included a “missing” category, to maintain the full available sample size at each time point in adjusted models. For regression analyses, covariates were dummy coded, with the typically more favorable category counting as the reference category (e.g., normal birth weight, normal math and reading scores, not overweight, no physical health problems, nonmanual paternal occupation, beyond minimal maternal education, maternal nonsmoking, married). Statistical Analysis Because healthy lifestyle scores were normally distributed, we used linear regression models to estimate the relationship between childhood psychological distress and adulthood healthy lifestyle, running separate models for the age 33 and 42 outcomes. (Note: we came to the same conclusions using Poisson regression models.) In a minimally adjusted model (Model 1), we adjusted for sex, and in a second model (Model 2), we adjusted for sex as well as all child and family covariates, to assess potential confounding. As a sensitivity analysis, we further adjusted for concurrent adult distress, to assess the independent effect of childhood distress. Adult distress was measured at ages 33 and 42 years, using the self-report psychological distress subscale of the Malaise Inventory [48]. The Inventory includes 15 items measuring psychological distress (e.g., “do you often feel miserable or depressed?”) with yes/no response options. We dealt with missing items in the same manner as in childhood. In additional analyses, we examined change in healthy lifestyle scores over time, among those who had healthy lifestyle measures at both time points (N = 7,998). We categorized lifestyle as improving (i.e., higher score at age 42 compared with 33), declining, or staying the same, in conjunction with whether people started with unhealthy (0 to 2 healthy lifestyle components at age 33) or healthy (3 to 5 components) lifestyle. This yielded six categories: unhealthy and declined, unhealthy constant, unhealthy but improved, healthy but declined, healthy constant, and healthy and improved. We used generalized linear models to assess whether childhood psychological distress was associated with the nature of the change in healthy lifestyle between age 33 and 42 years. To see if the child psychological functioning had a particularly strong impact on any one of the healthy lifestyle components, we explored associations with each component separately (at both ages) using logistic regression models, adjusting for sex (Model 1) plus all covariates (Model 2). While ideal body weight (BMI < 25 kg/m2) is typically included in healthy lifestyle indices [1, 3, 10, 11], unlike the other index components, it is not, strictly speaking, a health behavior. In sensitivity analyses, we reran the main models excluding BMI from the index. To minimize potential bias due to participant attrition, we applied inverse probability weights to all analyses, estimating weights based on key factors associated with attrition: sex, LBW, math and reading scores, internalizing and externalizing symptoms, overweight status, physical health problems, father’s occupation, mother’s education, and mother’s marital status. All analyses were performed using SAS version 9.3 software (SAS Institute, Inc., Cary, NC). Access to the data, including a special license agreement for use of biomedical data, was granted by the Economic and Social Data service council, the UK Data Archive, and the Center for Longitudinal Studies. Ethical approval of this study was given by the Harvard T.H. Chan School of Public Health Institutional Review Board. Results Baseline Characteristics and Healthy Lifestyle in Adulthood In the age 33 sample, half of the participants were female (50.7%, Table 1). At birth, 4.7% of these participants were low birth weight, the majority of their fathers had a manual occupation (70.0%), and the majority of their mothers were married (92.3%), had not smoked cigarettes during pregnancy (63.4%), and had left school at or before the minimum leaving age (70.2%). By age 7, 6.0% of respondents had experienced physical health problems. At age 7, 5.4% had low math scores, 7.1% had low reading scores, and 8.6% were overweight. The distribution of characteristics was almost identical in the sample at age 42 years. All covariates were significantly associated with childhood distress. Table 1 Sample Characteristics at Wave 5, and Mean Childhood Psychological Distress Z-score by Covariates Child variables Wave 5 (age 33)a Child psychological distress Z-scoreb N = 10,748 N (%) Mean (SD)  Sex   Female 5,446 (50.67) −0.15 (0.92)   Male 5,302 (49.33) 0.14 (1.05)  Low birth weight   No 9,692 (90.17) −0.02 (0.99)   Yes 500 (4.65) 0.17 (1.04)   Missing 556 (5.17) 0.08 (1.12)  Math performance   Normal 9,061 (84.30) −0.07 (0.93)   Low 584 (5.43) 0.76 (1.07)   Missing 1,103 (10.26) 0.12 (1.23)  Reading performance   Normal 8,905 (82.85) −0.11 (0.90)   Low 766 (7.13) 0.88 (1.06)   Missing 1,077 (10.02) 0.12 (1.25)  Overweight   No 7,886 (73.37) −0.04 (0.96)   Yes 925 (8.61) −0.06 (0.91)   Missing 1,937 (18.02) 0.15 (1.16)  Physical health problems   No 8,880 (82.62) −0.03 (0.96)   Yes 641 (5.96) 0.04 (1.02)   Missing 1,227 (11.42) 0.13 (1.19) Family variables  Father’s occupation   Nonmanual 2,939 (27.34) −0.25 (0.86)   Manual 7,524 (70.00) 0.08 (1.02)   Missing 285 (2.65) 0.16 (1.24)  Mother’s education level   Stayed after min leaving age 2,657 (24.72) −0.23 (0.88)   Left at/before min leaving age 7,549 (70.24) 0.07 (1.01)   Missing 542 (5.04) 0.09 (1.13)  Mother’s smoking during pregnancy   No 6,814 (63.40) −0.08 (0.95)   Yes 3,298 (30.68) 0.13 (1.04)   Missing 636 (5.92) 0.07 (1.13)  Mother’s marital status   Married 9,919 (92.29) −0.02 (0.99)   Unmarried/ Divorced/Widowed 310 (2.88) 0.29 (1.06)   Missing 519 (4.83) 0.09 (1.13) Child variables Wave 5 (age 33)a Child psychological distress Z-scoreb N = 10,748 N (%) Mean (SD)  Sex   Female 5,446 (50.67) −0.15 (0.92)   Male 5,302 (49.33) 0.14 (1.05)  Low birth weight   No 9,692 (90.17) −0.02 (0.99)   Yes 500 (4.65) 0.17 (1.04)   Missing 556 (5.17) 0.08 (1.12)  Math performance   Normal 9,061 (84.30) −0.07 (0.93)   Low 584 (5.43) 0.76 (1.07)   Missing 1,103 (10.26) 0.12 (1.23)  Reading performance   Normal 8,905 (82.85) −0.11 (0.90)   Low 766 (7.13) 0.88 (1.06)   Missing 1,077 (10.02) 0.12 (1.25)  Overweight   No 7,886 (73.37) −0.04 (0.96)   Yes 925 (8.61) −0.06 (0.91)   Missing 1,937 (18.02) 0.15 (1.16)  Physical health problems   No 8,880 (82.62) −0.03 (0.96)   Yes 641 (5.96) 0.04 (1.02)   Missing 1,227 (11.42) 0.13 (1.19) Family variables  Father’s occupation   Nonmanual 2,939 (27.34) −0.25 (0.86)   Manual 7,524 (70.00) 0.08 (1.02)   Missing 285 (2.65) 0.16 (1.24)  Mother’s education level   Stayed after min leaving age 2,657 (24.72) −0.23 (0.88)   Left at/before min leaving age 7,549 (70.24) 0.07 (1.01)   Missing 542 (5.04) 0.09 (1.13)  Mother’s smoking during pregnancy   No 6,814 (63.40) −0.08 (0.95)   Yes 3,298 (30.68) 0.13 (1.04)   Missing 636 (5.92) 0.07 (1.13)  Mother’s marital status   Married 9,919 (92.29) −0.02 (0.99)   Unmarried/ Divorced/Widowed 310 (2.88) 0.29 (1.06)   Missing 519 (4.83) 0.09 (1.13) aDistributions are shown for sample at Wave 5, which were almost identical to distributions at Wave 6. bHigher (positive) scores indicate more distress. All p-values <.001 for tests of between group differences in child distress Z-scores. View Large Table 1 Sample Characteristics at Wave 5, and Mean Childhood Psychological Distress Z-score by Covariates Child variables Wave 5 (age 33)a Child psychological distress Z-scoreb N = 10,748 N (%) Mean (SD)  Sex   Female 5,446 (50.67) −0.15 (0.92)   Male 5,302 (49.33) 0.14 (1.05)  Low birth weight   No 9,692 (90.17) −0.02 (0.99)   Yes 500 (4.65) 0.17 (1.04)   Missing 556 (5.17) 0.08 (1.12)  Math performance   Normal 9,061 (84.30) −0.07 (0.93)   Low 584 (5.43) 0.76 (1.07)   Missing 1,103 (10.26) 0.12 (1.23)  Reading performance   Normal 8,905 (82.85) −0.11 (0.90)   Low 766 (7.13) 0.88 (1.06)   Missing 1,077 (10.02) 0.12 (1.25)  Overweight   No 7,886 (73.37) −0.04 (0.96)   Yes 925 (8.61) −0.06 (0.91)   Missing 1,937 (18.02) 0.15 (1.16)  Physical health problems   No 8,880 (82.62) −0.03 (0.96)   Yes 641 (5.96) 0.04 (1.02)   Missing 1,227 (11.42) 0.13 (1.19) Family variables  Father’s occupation   Nonmanual 2,939 (27.34) −0.25 (0.86)   Manual 7,524 (70.00) 0.08 (1.02)   Missing 285 (2.65) 0.16 (1.24)  Mother’s education level   Stayed after min leaving age 2,657 (24.72) −0.23 (0.88)   Left at/before min leaving age 7,549 (70.24) 0.07 (1.01)   Missing 542 (5.04) 0.09 (1.13)  Mother’s smoking during pregnancy   No 6,814 (63.40) −0.08 (0.95)   Yes 3,298 (30.68) 0.13 (1.04)   Missing 636 (5.92) 0.07 (1.13)  Mother’s marital status   Married 9,919 (92.29) −0.02 (0.99)   Unmarried/ Divorced/Widowed 310 (2.88) 0.29 (1.06)   Missing 519 (4.83) 0.09 (1.13) Child variables Wave 5 (age 33)a Child psychological distress Z-scoreb N = 10,748 N (%) Mean (SD)  Sex   Female 5,446 (50.67) −0.15 (0.92)   Male 5,302 (49.33) 0.14 (1.05)  Low birth weight   No 9,692 (90.17) −0.02 (0.99)   Yes 500 (4.65) 0.17 (1.04)   Missing 556 (5.17) 0.08 (1.12)  Math performance   Normal 9,061 (84.30) −0.07 (0.93)   Low 584 (5.43) 0.76 (1.07)   Missing 1,103 (10.26) 0.12 (1.23)  Reading performance   Normal 8,905 (82.85) −0.11 (0.90)   Low 766 (7.13) 0.88 (1.06)   Missing 1,077 (10.02) 0.12 (1.25)  Overweight   No 7,886 (73.37) −0.04 (0.96)   Yes 925 (8.61) −0.06 (0.91)   Missing 1,937 (18.02) 0.15 (1.16)  Physical health problems   No 8,880 (82.62) −0.03 (0.96)   Yes 641 (5.96) 0.04 (1.02)   Missing 1,227 (11.42) 0.13 (1.19) Family variables  Father’s occupation   Nonmanual 2,939 (27.34) −0.25 (0.86)   Manual 7,524 (70.00) 0.08 (1.02)   Missing 285 (2.65) 0.16 (1.24)  Mother’s education level   Stayed after min leaving age 2,657 (24.72) −0.23 (0.88)   Left at/before min leaving age 7,549 (70.24) 0.07 (1.01)   Missing 542 (5.04) 0.09 (1.13)  Mother’s smoking during pregnancy   No 6,814 (63.40) −0.08 (0.95)   Yes 3,298 (30.68) 0.13 (1.04)   Missing 636 (5.92) 0.07 (1.13)  Mother’s marital status   Married 9,919 (92.29) −0.02 (0.99)   Unmarried/ Divorced/Widowed 310 (2.88) 0.29 (1.06)   Missing 519 (4.83) 0.09 (1.13) aDistributions are shown for sample at Wave 5, which were almost identical to distributions at Wave 6. bHigher (positive) scores indicate more distress. All p-values <.001 for tests of between group differences in child distress Z-scores. View Large Distress Z-scores at each age in childhood (7, 11, and 16 years) were positively correlated with one another, with correlations ranging from 0.26 to 0.38 (all p-values were <.001). Child distress was also modestly correlated with adult distress at age 33 (0.18, p < .001) and 42 (0.15, p < .001). The healthy lifestyle index scores at age 33 were positively correlated with healthy lifestyle index at age 42 (polychoric correlation = 0.55, asymptotic standard error [ASE] = 0.009). Correlations between childhood psychological distress, overall healthy lifestyle score, and each dichotomous component of the healthy lifestyle index are presented in Supplementary Table B. The proportion of nonsmokers increased slightly between ages 33 and 42, while the proportion of participants reporting moderate alcohol consumption decreased over time (Table 2). The proportions of the other three lifestyle factors remained fairly stable between the two time points. Only 3.8% and 2.8% of 33 and 42 year olds, respectively, endorsed all five components of a healthy lifestyle. The majority of the people reported having two or three healthy lifestyle components at each age. Table 2 Number (and Percent) of People, and Mean (and SD) of Childhood Psychological Distress Z-score, by Healthy Lifestyle Component and Score Age 33 (Wave 5) N = 10,748 Age 42 (Wave 6) N = 9,581 People Child distress Z-scorea People Child distress Z-scorea N (%) M (SD) N (%) M (SD) Healthy lifestyle component Nonsmoker  No 3,561 (33.13) 0.13 (1.03) 2,809 (29.32) 0.11 (1.00)  Yes 7,187 (66.87) −0.21 (0.84) 6,772 (70.68) −0.23 (0.83) Moderate alcohol consumption  No 2,994 (27.86) −0.06 (0.96) 4,211 (43.95) −0.09 (0.94)  Yes 7,754 (72.14) −0.11 (0.91) 5,370 (56.05) −0.16 (0.86) Regular physical activity  No 7,988 (74.32) −0.11 (0.92) 7,112 (74.23) −0.14 (0.89)  Yes 2,760 (25.68) −0.07 (0.94) 2,469 (25.77) −0.11 (0.93) Healthy diet  No 6,759 (62.89) −0.04 (0.95) 6,126 (63.94) −0.07 (0.93)  Yes 3,989 (37.11) −0.20 (0.87) 3,455 (36.06) −0.24 (0.83) Ideal body weight (body mass index <25)  No 4,664 (43.39) −0.03 (0.95) 4,323 (45.12) −0.04 (0.93)  Yes 6,084 (56.61) −0.15 (0.90) 5,258 (54.88) −0.20 (0.86) Total number of healthy lifestyle components  0 295 (2.74) 0.17 (1.08) 318 (3.32) 0.24 (1.03)  1 1,450 (13.49) 0.12 (1.01) 1,666 (17.39) 0.06 (0.99)  2 3,262 (30.35) −0.02 (0.96) 3,029 (31.61) −0.08 (0.90)  3 3,569 (33.21) −0.16 (0.88) 2,944 (30.73) −0.21 (0.85)  4 1,767 (16.44) −0.29 (0.80) 1,352 (14.11) −0.33 (0.75)  5 405 (3.77) −0.28 (0.77) 272 (2.84) −0.38 (0.71) Age 33 (Wave 5) N = 10,748 Age 42 (Wave 6) N = 9,581 People Child distress Z-scorea People Child distress Z-scorea N (%) M (SD) N (%) M (SD) Healthy lifestyle component Nonsmoker  No 3,561 (33.13) 0.13 (1.03) 2,809 (29.32) 0.11 (1.00)  Yes 7,187 (66.87) −0.21 (0.84) 6,772 (70.68) −0.23 (0.83) Moderate alcohol consumption  No 2,994 (27.86) −0.06 (0.96) 4,211 (43.95) −0.09 (0.94)  Yes 7,754 (72.14) −0.11 (0.91) 5,370 (56.05) −0.16 (0.86) Regular physical activity  No 7,988 (74.32) −0.11 (0.92) 7,112 (74.23) −0.14 (0.89)  Yes 2,760 (25.68) −0.07 (0.94) 2,469 (25.77) −0.11 (0.93) Healthy diet  No 6,759 (62.89) −0.04 (0.95) 6,126 (63.94) −0.07 (0.93)  Yes 3,989 (37.11) −0.20 (0.87) 3,455 (36.06) −0.24 (0.83) Ideal body weight (body mass index <25)  No 4,664 (43.39) −0.03 (0.95) 4,323 (45.12) −0.04 (0.93)  Yes 6,084 (56.61) −0.15 (0.90) 5,258 (54.88) −0.20 (0.86) Total number of healthy lifestyle components  0 295 (2.74) 0.17 (1.08) 318 (3.32) 0.24 (1.03)  1 1,450 (13.49) 0.12 (1.01) 1,666 (17.39) 0.06 (0.99)  2 3,262 (30.35) −0.02 (0.96) 3,029 (31.61) −0.08 (0.90)  3 3,569 (33.21) −0.16 (0.88) 2,944 (30.73) −0.21 (0.85)  4 1,767 (16.44) −0.29 (0.80) 1,352 (14.11) −0.33 (0.75)  5 405 (3.77) −0.28 (0.77) 272 (2.84) −0.38 (0.71) aHigher (positive) scores indicate more distress. View Large Table 2 Number (and Percent) of People, and Mean (and SD) of Childhood Psychological Distress Z-score, by Healthy Lifestyle Component and Score Age 33 (Wave 5) N = 10,748 Age 42 (Wave 6) N = 9,581 People Child distress Z-scorea People Child distress Z-scorea N (%) M (SD) N (%) M (SD) Healthy lifestyle component Nonsmoker  No 3,561 (33.13) 0.13 (1.03) 2,809 (29.32) 0.11 (1.00)  Yes 7,187 (66.87) −0.21 (0.84) 6,772 (70.68) −0.23 (0.83) Moderate alcohol consumption  No 2,994 (27.86) −0.06 (0.96) 4,211 (43.95) −0.09 (0.94)  Yes 7,754 (72.14) −0.11 (0.91) 5,370 (56.05) −0.16 (0.86) Regular physical activity  No 7,988 (74.32) −0.11 (0.92) 7,112 (74.23) −0.14 (0.89)  Yes 2,760 (25.68) −0.07 (0.94) 2,469 (25.77) −0.11 (0.93) Healthy diet  No 6,759 (62.89) −0.04 (0.95) 6,126 (63.94) −0.07 (0.93)  Yes 3,989 (37.11) −0.20 (0.87) 3,455 (36.06) −0.24 (0.83) Ideal body weight (body mass index <25)  No 4,664 (43.39) −0.03 (0.95) 4,323 (45.12) −0.04 (0.93)  Yes 6,084 (56.61) −0.15 (0.90) 5,258 (54.88) −0.20 (0.86) Total number of healthy lifestyle components  0 295 (2.74) 0.17 (1.08) 318 (3.32) 0.24 (1.03)  1 1,450 (13.49) 0.12 (1.01) 1,666 (17.39) 0.06 (0.99)  2 3,262 (30.35) −0.02 (0.96) 3,029 (31.61) −0.08 (0.90)  3 3,569 (33.21) −0.16 (0.88) 2,944 (30.73) −0.21 (0.85)  4 1,767 (16.44) −0.29 (0.80) 1,352 (14.11) −0.33 (0.75)  5 405 (3.77) −0.28 (0.77) 272 (2.84) −0.38 (0.71) Age 33 (Wave 5) N = 10,748 Age 42 (Wave 6) N = 9,581 People Child distress Z-scorea People Child distress Z-scorea N (%) M (SD) N (%) M (SD) Healthy lifestyle component Nonsmoker  No 3,561 (33.13) 0.13 (1.03) 2,809 (29.32) 0.11 (1.00)  Yes 7,187 (66.87) −0.21 (0.84) 6,772 (70.68) −0.23 (0.83) Moderate alcohol consumption  No 2,994 (27.86) −0.06 (0.96) 4,211 (43.95) −0.09 (0.94)  Yes 7,754 (72.14) −0.11 (0.91) 5,370 (56.05) −0.16 (0.86) Regular physical activity  No 7,988 (74.32) −0.11 (0.92) 7,112 (74.23) −0.14 (0.89)  Yes 2,760 (25.68) −0.07 (0.94) 2,469 (25.77) −0.11 (0.93) Healthy diet  No 6,759 (62.89) −0.04 (0.95) 6,126 (63.94) −0.07 (0.93)  Yes 3,989 (37.11) −0.20 (0.87) 3,455 (36.06) −0.24 (0.83) Ideal body weight (body mass index <25)  No 4,664 (43.39) −0.03 (0.95) 4,323 (45.12) −0.04 (0.93)  Yes 6,084 (56.61) −0.15 (0.90) 5,258 (54.88) −0.20 (0.86) Total number of healthy lifestyle components  0 295 (2.74) 0.17 (1.08) 318 (3.32) 0.24 (1.03)  1 1,450 (13.49) 0.12 (1.01) 1,666 (17.39) 0.06 (0.99)  2 3,262 (30.35) −0.02 (0.96) 3,029 (31.61) −0.08 (0.90)  3 3,569 (33.21) −0.16 (0.88) 2,944 (30.73) −0.21 (0.85)  4 1,767 (16.44) −0.29 (0.80) 1,352 (14.11) −0.33 (0.75)  5 405 (3.77) −0.28 (0.77) 272 (2.84) −0.38 (0.71) aHigher (positive) scores indicate more distress. View Large Childhood Psychological Distress and Healthy Lifestyle Adjusting for sex, childhood distress was negatively associated with healthy lifestyle at age 33 (β = −0.14, SE = 0.01, p<.001) and 42 (β = −0.15, SE = 0.01, p<.001) (Table 3). The effect of distress on healthy lifestyle attenuated slightly in fully adjusted models: Holding all other variables constant, a 1 SD increase in distress was associated, on average, with a 0.11 lower healthy lifestyle index score at age 33 (SE = 0.01, p<.001) and a 0.13 lower score at age 42 (SE = 0.01, p<.001). These estimates are similar in magnitude, for example, to the significant effect on age 42 healthy lifestyle of maternal smoking versus nonsmoking during pregnancy (β = −0.10) and manual versus nonmanual paternal occupation (β = −0.12), both established predictors of health behaviors [49, 50]. When comparing mean healthy lifestyle levels at age 42 between the groups with lower (bottom quartile, mean healthy lifestyle score = 2.6) versus higher (top quartile, mean healthy lifestyle score = 2.1) psychological distress, the mean difference in lifestyle score was approximately 0.5 on a five-point scale, equivalent to a 10% decrease in level of healthy lifestyle (or half a health behavior). Table 3 Linear Regression of the Association Between Childhood Psychological Distress and Healthy Lifestylea at Ages 33 and 42 Years Model 1 Model 2 β (SE) p-value β (SE) p-value Age 33 years (N = 10,748) −0.14 (0.01) <.001 −0.11 (0.01) <.001 Age 42 years (N = 9,581) −0.15 (0.01) <.001 −0.13 (0.01) <.001 Model 1 Model 2 β (SE) p-value β (SE) p-value Age 33 years (N = 10,748) −0.14 (0.01) <.001 −0.11 (0.01) <.001 Age 42 years (N = 9,581) −0.15 (0.01) <.001 −0.13 (0.01) <.001 a“Healthy lifestyle” consists of 5 components: absence of smoking, moderate alcohol consumption, regular physical activity, healthy diet, and ideal body weight. We modeled childhood psychological distress, where higher scores indicate more distress. Model 1 adjusts for sex; Model 2 adjusts for sex plus child covariates (LBW, math performance, reading performance, overweight, and physical health problems) and family covariates (father’s occupation, and mother’s education, smoking during pregnancy, and marital status). View Large Table 3 Linear Regression of the Association Between Childhood Psychological Distress and Healthy Lifestylea at Ages 33 and 42 Years Model 1 Model 2 β (SE) p-value β (SE) p-value Age 33 years (N = 10,748) −0.14 (0.01) <.001 −0.11 (0.01) <.001 Age 42 years (N = 9,581) −0.15 (0.01) <.001 −0.13 (0.01) <.001 Model 1 Model 2 β (SE) p-value β (SE) p-value Age 33 years (N = 10,748) −0.14 (0.01) <.001 −0.11 (0.01) <.001 Age 42 years (N = 9,581) −0.15 (0.01) <.001 −0.13 (0.01) <.001 a“Healthy lifestyle” consists of 5 components: absence of smoking, moderate alcohol consumption, regular physical activity, healthy diet, and ideal body weight. We modeled childhood psychological distress, where higher scores indicate more distress. Model 1 adjusts for sex; Model 2 adjusts for sex plus child covariates (LBW, math performance, reading performance, overweight, and physical health problems) and family covariates (father’s occupation, and mother’s education, smoking during pregnancy, and marital status). View Large Excluding BMI from the healthy lifestyle index yielded similar results (Supplementary Table C). When simultaneously adjusting for concurrent adult distress, in addition to all other covariates, the effect of child distress on healthy lifestyle remained significant (age 33: β = −0.09, SE = 0.01, p<.001; age 42: β = −0.10, SE = 0.01, p<.001). Concurrent adult distress was also independently associated with healthy lifestyle, in fully adjusted models, at both age 33 (β = −0.06, SE = 0.005, p<.001) and 42 (β = −0.06, SE = 0.01, p<.001). Analysis of Change in Lifestyle Of those who had outcome measures at both time points (N = 7998), 8.6% started unhealthy and declined, 18.8% started unhealthy and remained constant, 18.2% started unhealthy and improved, 27.1% started healthy and declined, 20.2% started healthy and remained constant, and 7.2% started healthy and improved. Fully adjusted generalized linear models revealed significant between group differences in psychological distress in childhood (F = 12.31, p<.001). Distress Z-scores were highest among those who started unhealthy and declined (mean = 0.12) and lowest among those who started healthy and improved (mean = −0.28). Those who started unhealthy and declined had significantly higher distress than those in all other groups, except for those who started unhealthy and remained unhealthy (Fig. 2). Fig. 2. View largeDownload slide Childhood psychological distress Z-scores by category of lifestyle change between ages 33 and 42: unhealthy and declined (UD), unhealthy constant (UC), unhealthy but improved (UI), healthy but declined (HD), healthy constant (HC), and healthy and improved (HI). Fig. 2. View largeDownload slide Childhood psychological distress Z-scores by category of lifestyle change between ages 33 and 42: unhealthy and declined (UD), unhealthy constant (UC), unhealthy but improved (UI), healthy but declined (HD), healthy constant (HC), and healthy and improved (HI). Analysis of Individual Lifestyle Components The overall child distress score was associated with almost all the individual components of a healthy lifestyle, at both ages 33 and 42 (Table 4). Apart from regular physical activity, a standard deviation increase in distress was associated with lower odds of endorsing each of the healthy lifestyle components. The largest reduction in odds was observed for smoking. Each 1 SD increase in childhood distress score corresponded to about a 30% decrease in the odds of being a nonsmoker (at both ages). Child distress was associated with slightly higher odds of reporting regular physical activity at ages 33 and 42 years. Table 4 Fully Adjusted Logistic Regression of Individual Dichotomous Healthy Lifestyle Components on Child Psychological Distress Age 33 (Wave 5) Age 42 (Wave 6) N = 10,748 N = 9,581 OR (95% CI) p-value OR (95% CI) p-value Healthy lifestyle componentsa Nonsmoker 0.71 (0.69–0.74) <.001 0.70 (0.68–0.73) <.001 Moderate alcohol consumption 0.96 (0.92–0.99) .009 0.92 (0.89–0.95) <.001 Regular physical activity 1.04 (1.01–1.08) .03 1.05 (1.01–1.09) .007 Healthy diet 0.90 (0.87–0.93) <.001 0.88 (0.85–0.91) <.001 Ideal body weight (body mass index <25) 0.96 (0.93–0.99) .009 0.94 (0.91–0.96) <.001 Age 33 (Wave 5) Age 42 (Wave 6) N = 10,748 N = 9,581 OR (95% CI) p-value OR (95% CI) p-value Healthy lifestyle componentsa Nonsmoker 0.71 (0.69–0.74) <.001 0.70 (0.68–0.73) <.001 Moderate alcohol consumption 0.96 (0.92–0.99) .009 0.92 (0.89–0.95) <.001 Regular physical activity 1.04 (1.01–1.08) .03 1.05 (1.01–1.09) .007 Healthy diet 0.90 (0.87–0.93) <.001 0.88 (0.85–0.91) <.001 Ideal body weight (body mass index <25) 0.96 (0.93–0.99) .009 0.94 (0.91–0.96) <.001 aEach healthy lifestyle component is dichotomized (0/1, with 1 indicating presence of healthy levels of the component). Each healthy lifestyle component is modeled separately, at both ages 33 and 42 years. Cell entries are odds ratios (ORs) and 95% confidence intervals (CIs). ORs show the odds of endorsing each healthy lifestyle component associated with a 1-SD increase in distress score. ORs < 1 indicate lower odds of having the health component. All models adjust for all child and family covariates. View Large Table 4 Fully Adjusted Logistic Regression of Individual Dichotomous Healthy Lifestyle Components on Child Psychological Distress Age 33 (Wave 5) Age 42 (Wave 6) N = 10,748 N = 9,581 OR (95% CI) p-value OR (95% CI) p-value Healthy lifestyle componentsa Nonsmoker 0.71 (0.69–0.74) <.001 0.70 (0.68–0.73) <.001 Moderate alcohol consumption 0.96 (0.92–0.99) .009 0.92 (0.89–0.95) <.001 Regular physical activity 1.04 (1.01–1.08) .03 1.05 (1.01–1.09) .007 Healthy diet 0.90 (0.87–0.93) <.001 0.88 (0.85–0.91) <.001 Ideal body weight (body mass index <25) 0.96 (0.93–0.99) .009 0.94 (0.91–0.96) <.001 Age 33 (Wave 5) Age 42 (Wave 6) N = 10,748 N = 9,581 OR (95% CI) p-value OR (95% CI) p-value Healthy lifestyle componentsa Nonsmoker 0.71 (0.69–0.74) <.001 0.70 (0.68–0.73) <.001 Moderate alcohol consumption 0.96 (0.92–0.99) .009 0.92 (0.89–0.95) <.001 Regular physical activity 1.04 (1.01–1.08) .03 1.05 (1.01–1.09) .007 Healthy diet 0.90 (0.87–0.93) <.001 0.88 (0.85–0.91) <.001 Ideal body weight (body mass index <25) 0.96 (0.93–0.99) .009 0.94 (0.91–0.96) <.001 aEach healthy lifestyle component is dichotomized (0/1, with 1 indicating presence of healthy levels of the component). Each healthy lifestyle component is modeled separately, at both ages 33 and 42 years. Cell entries are odds ratios (ORs) and 95% confidence intervals (CIs). ORs show the odds of endorsing each healthy lifestyle component associated with a 1-SD increase in distress score. ORs < 1 indicate lower odds of having the health component. All models adjust for all child and family covariates. View Large Discussion Using prospective data from the 1958 British Birth Cohort Study, we found that individuals with higher psychological distress between ages 7 and 16 years were less likely to maintain a healthy lifestyle at both ages 33 and 42 years. Greater distress in childhood was associated with less healthy lifestyle, even when controlling for concurrent adult distress. Childhood distress was also associated with deterioration in lifestyle between ages 33 and 42, suggesting that the effects of childhood distress may continue to compound over the life course with potentially substantial impacts on CVD development. Despite evidence of correlations between psychological distress and individual health behaviors, surprisingly little research has investigated these relationships prospectively or gone beyond considering a single behavior to evaluate a set of factors that comprise an overall healthy lifestyle. This is the first study to our knowledge to explore the prospective association between childhood psychological distress and an overall healthy lifestyle in adulthood. Our findings are consistent with those from other community-based longitudinal studies, reporting that aspects of child distress predict unhealthy behaviors later in adolescence or adulthood, such as cigarette smoking and/or problematic alcohol use [18, 19]. In the current study, when exploring the individual components of a healthy lifestyle as separate outcomes we also found that, with the exception of regular physical activity, more childhood distress was associated with lower odds of being healthy on each lifestyle component. The physical activity finding is somewhat surprising given prior evidence, among adults, of a relationship between depression and greater likelihood of sedentary lifestyle [51]. However, the impact of childhood distress on adult physical activity may depend on the specific aspect and manifestation of childhood distress. For example, individuals experiencing attention-deficit hyperactivity symptoms may exercise as a way to alleviate those symptoms [52]. The size of associations in this study was relatively small. However, determinants of health behaviors are likely multifactorial, and any one variable will likely contribute only modestly to such outcomes. Given the robust link between overall healthy lifestyle and reduced risk of chronic disease [1–7, 10, 11], even small changes in healthy lifestyle, associated with childhood distress, may be clinically relevant. For example, results from analyses using a similar measure of healthy lifestyle among women in the Nurses Health Study indicate a 16.20% increased risk of stroke, on average, for each 1-point decrease in lifestyle ([3], [53]). Assuming a comparable sample, the 0.5-point decrease in healthy lifestyle score assuming a comparable sample, would reflect an approximate 8% increase in stroke risk. High levels of distress in childhood could plausibly be linked with less healthy lifestyle components in adulthood in a number of ways. For example, smoking, drinking, or “comfort eating” may be used to self-medicate or cope with psychological distress or affective symptoms [54–56]. Distress in childhood could lead to impaired social relationships or alienation from mainstream peers, fostering affiliation with more “deviant” peer groups, which may influence substance use behaviors [21]. Distress in childhood may impair educational or professional attainment [57], thereby limiting resources to access healthy food or safe exercise spaces. Distress may also contribute to poorer self-regulatory capacity, due to demands on self-control resources, which is associated with physical activity, smoking, alcohol consumption, and eating behaviors [58]. These and other potential mechanisms should be explored further in future research. Moreover, as the distribution of child psychological distress in the population is not random, attention should also be given to its determinants. Though some aspects of child psychological functioning (e.g., hyperactivity) may have a genetic component [59], a great deal of research has suggested that child and adolescent psychopathology is strongly shaped by early-life experiences, with adversity [60], social stress [61], and lower socioeconomic status [62] each contributing to greater likelihood of high levels of distress. Other research suggests that childhood mental health is modifiable, with prevention and intervention strategies targeting children, caregivers, and schools [63–67]. The current study has some limitations. Despite having measured distress prospectively, we cannot completely rule out the possibility of reverse causation, given many unhealthy behaviors have their origins in adolescence and some in childhood [13]. Indeed, prior research suggests that the relationships between mental health and health behaviors is likely bidirectional, although these effects may be more apparent as individuals enter into late adolescence and young adulthood [68]. Although we adjusted for many potential confounders, there may be others that were unmeasured, including genetic variation [69] and parental factors such as child-rearing style [70] and psychopathology [71]. Another limitation is that the available measures for several behaviors were somewhat constrained. For example, we had information only on frequency of consumption of various foods but not portion size. However, other work has suggested that the consumption frequency explains most of the variation in food intake [43]. While we may have lost some information by dichotomizing the healthy lifestyle components, this approach allowed us to condense information into a meaningful index and is consistent with measures of healthy lifestyle that have previously been shown to predict disease risk. For other ways to operationalize health behavior composites, and methodological considerations regarding combining information across multiple health behaviors, see the study by Prochaska et al. [9]. Finally, like many longitudinal studies, the 1958 British Birth Cohort Study is limited by participant attrition; however, the remaining samples at ages 33 and 42 were fairly representative of the original cohort, and our use of inverse probability weights in analyses reduces biases due to attrition. This study also has a number of important strengths. It is one of the first studies to measure psychological distress in childhood and consider it in relation to a set of subsequent health-related behaviors over a 17- to 35-year follow-up period. Childhood distress was reported by teachers when children were ages 7, 11, and 16 years, alleviating concerns that the participants’ adult health status or behavior could bias their memory or reporting of early symptoms; a potential concern with retrospective designs. Furthermore, we could evaluate and account for a broad range of potential confounders. With increased appreciation for the childhood origins of adult CVD [72], and the growing interest in primordial prevention [14], it is essential to evaluate whether and what factors in childhood may provide insight into whether individuals will initiate and maintain key healthy lifestyle practices. The small percentage of participants with all five components of a healthy lifestyle in young and mid adulthood (3.8% at age 33 and 2.8% at age 42) highlights the importance of identifying opportunities to promote a healthy lifestyle; primordial prevention efforts may benefit from greater consideration of the role of childhood mental health in setting trajectories of healthy lifestyle across the life course. Our findings suggest that even at a relatively young age, psychological distress may signal increased risk for engaging in an array of less healthy lifestyle practices later in life. Critical future steps will be to evaluate whether improving child mental health indeed improves subsequent health risk factors in adulthood and correspondingly decreases risk for CVD. CVD primordial prevention efforts would benefit from consideration of the child psychological distress and the role it plays in relation to known/traditional risk factors for CVD. Elucidating these mechanisms of risk may provide important targets for prevention efforts in children, which may have long-lasting effects on adult cardiovascular health. Supplementary Material Supplementary material is available at Annals of Behavioral Medicine online. Acknowledgments Dr. Winning was supported by the Julius B. Richmond Fellowship at the Harvard Center on the Developing Child and by the Martha May Eliot Fund at the Harvard T. H. Chan School of Public Health, and Dr. Gilsanz was supported by the Yerby Postdoctoral Fellowship; however, no direct funding was received or set aside for the writing of this paper. Compliance with Ethical Standards Authors’ Statement of Conflict of Interest and Adherence to Ethical Standards The authors declare that they have no conflict of interest. Primary Data: We performed secondary data analysis of data from the 1958 British Birth Cohort Study. Authors' Contributions: All authors contributed to the ideas in this paper. AW led the writing and analyses, and all authors contributed to drafts and approved the final version. Ethical approval: All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Informed consent: Informed consent was obtained from all individual participants included in the 1958 British Birth Cohort study. References 1. Stampfer MJ , Hu FB , Manson JE , Rimm EB , Willett WC . Primary prevention of coronary heart disease in women through diet and lifestyle . N Engl J Med . 2000 ; 343 : 16 – 22 . Google Scholar CrossRef Search ADS PubMed 2. Chiuve SE , McCullough ML , Sacks FM , Rimm EB . Healthy lifestyle factors in the primary prevention of coronary heart disease among men: Benefits among users and nonusers of lipid-lowering and antihypertensive medications . Circulation . 2006 ; 114 : 160 – 167 . Google Scholar CrossRef Search ADS PubMed 3. Chiuve SE , Rexrode KM , Spiegelman D et al. Primary prevention of stroke by healthy lifestyle . Circulation . 2008 ; 118 : 947 – 954 . Google Scholar CrossRef Search ADS PubMed 4. Hu FB , Manson JE , Stampfer MJ et al. Diet, lifestyle, and the risk of type 2 diabetes mellitus in women . N Engl J Med . 2001 ; 345 : 790 – 797 . Google Scholar CrossRef Search ADS PubMed 5. Platz EA , Willett WC , Colditz GA et al. Proportion of colon cancer risk that might be preventable in a cohort of middle-aged US men . Cancer Causes Control . 2000 ; 11 : 579 – 588 . Google Scholar CrossRef Search ADS PubMed 6. Dartois L , Fagherazzi G , Boutron-Ruault MC , Mesrine S , Clavel-Chapelon F . Association between five lifestyle habits and cancer risk: Results from the E3N cohort . Cancer Prevent Res . 2014 ; 7 : 516 – 525 . Google Scholar CrossRef Search ADS 7. Knoops KT , de Groot LC , Kromhout D et al. Mediterranean diet, lifestyle factors, and 10-year mortality in elderly European men and women: The HALE project . JAMA . 2004 ; 292 : 1433 – 1439 . Google Scholar CrossRef Search ADS PubMed 8. Shaw BA , Agahi N . A prospective cohort study of health behavior profiles after age 50 and mortality risk . BMC Public Health . 2012 ; 12 : 803 . Google Scholar CrossRef Search ADS PubMed 9. Prochaska JJ , Spring B , Nigg CR . Multiple health behavior change research: An introduction and overview . Prev Med . 2008 ; 46 : 181 – 188 . Google Scholar CrossRef Search ADS PubMed 10. Liu K , Daviglus ML , Loria CM et al. Healthy lifestyle through young adulthood and the presence of low cardiovascular disease risk profile in middle age: The Coronary Artery Risk Development in (Young) Adults (CARDIA) study . Circulation . 2012 ; 125 : 996 – 1004 . Google Scholar CrossRef Search ADS PubMed 11. Chomistek AK , Chiuve SE , Eliassen AH et al. Healthy lifestyle in the primordial prevention of cardiovascular disease among young women . J Am Coll Cardiol . 2015 ; 65 : 43 – 51 . Google Scholar CrossRef Search ADS PubMed 12. Kuh D , Ben Shlomo Y. A life course approach to chronic disease epidemiology: Tracing the origins of ill-health from early to adult life . Oxford : Oxford University Press ; 1997 . 13. Lloyd-Jones DM , Hong Y , Labarthe D et al. Defining and setting national goals for cardiovascular health promotion and disease reduction: The American Heart Association’s strategic Impact Goal through 2020 and beyond . Circulation . 2010 ; 121 : 586 – 613 . Google Scholar CrossRef Search ADS PubMed 14. Lloyd-Jones DM . Cardiovascular health and protection against CVD. More than the sum of the parts ? Circulation . 2014 ; 130 : 1671 – 1673 . Google Scholar CrossRef Search ADS PubMed 15. Appleton AA , Buka SL , McCormick MC et al. Emotional functioning at age 7 years is associated with C-reactive protein in middle adulthood . Psychosom Med . 2011 ; 73 : 295 – 303 . Google Scholar CrossRef Search ADS PubMed 16. Goodwin RD , Sourander A , Duarte CS et al. Do mental health problems in childhood predict chronic physical conditions among males in early adulthood? Evidence from a community-based prospective study . Psychol Med . 2009 ; 39 : 301 – 311 . Google Scholar CrossRef Search ADS PubMed 17. Winning A , Glymour MM , McCormick MC , Gilsanz P , Kubzansky LD . Psychological distress across the life course and cardiometabolic risk: Findings from the 1958 British Birth Cohort Study . J Am Coll Cardiol . 2015 ; 66 : 1577 – 1586 . Google Scholar CrossRef Search ADS PubMed 18. Charach A , Yeung E , Climans T , Lillie E . Childhood attention-deficit/hyperactivity disorder and future substance use disorders: Comparative meta-analyses . J Am Acad Child Adolesc Psychiatry . 2011 ; 50 : 9 – 21 . Google Scholar CrossRef Search ADS PubMed 19. Marmorstein NR , White HR , Loeber R , Stouthamer-Loeber M . Anxiety as a predictor of age at first use of substances and progression to substance use problems among boys . J Abnorm Child Psychol . 2010 ; 38 : 211 – 224 . Google Scholar CrossRef Search ADS PubMed 20. von Stumm S , Deary IJ , Kivimaki M et al. Childhood behavior problems and health at midlife: 35-year follow-up of a Scottish birth cohort . J Child Psychol Psychiatry . 2011 ; 52 : 992 – 1001 . Google Scholar CrossRef Search ADS PubMed 21. Hayatbakhsh R , Mamun AA , Williams GM , O’Callaghan MJ , Najman JM . Early childhood predictors of early onset of smoking: A birth prospective study . Addictive Behaviors . 2013 ; 38 : 2513 – 2519 . Google Scholar CrossRef Search ADS PubMed 22. Nomura Y , Gilman SE , Buka SL . Maternal smoking during pregnancy and risk of alcohol use disorders among adult offspring . J Stud Alcohol Drugs . 2011 ; 72 : 199 – 209 . Google Scholar CrossRef Search ADS PubMed 23. Power C , Jefferis BJ . Fetal environment and subsequent obesity: A study of maternal smoking . Int J Epidemiol . 2002 ; 31 : 413 – 419 . Google Scholar CrossRef Search ADS PubMed 24. Grzywacz JG , Marks NF . Social inequalities and exercise during adulthood: Toward an ecological perspective . J Health Soc Behav . 2001 ; 42 : 202 – 220 . Google Scholar CrossRef Search ADS PubMed 25. Snead MC , Cockerham WC . Health lifestyles and social class in the deep South . In Kronenfeld JJ , (ed), Social inequalities, health and health care delivery . Vol. 20 . Bingley, UK: Emerald Group Publishing Limited ; 2002 : 107 – 122 . Google Scholar CrossRef Search ADS 26. Melchior M , Moffitt TE , Milne BJ , Poulton R , Caspi A . Why do children from socioeconomically disadvantaged families suffer from poor health when they reach adulthood? A life-course study . Am J Epidemiol . 2007 ; 166 : 966 – 974 . Google Scholar CrossRef Search ADS PubMed 27. El Marroun H , Schmidt MN , Franken IH et al. Prenatal tobacco exposure and brain morphology: A prospective study in young children . Neuropsychopharmacology . 2014 ; 39 : 792 – 800 . Google Scholar CrossRef Search ADS PubMed 28. Ross CE , Bird CE . Sex stratification and health lifestyle: Consequences for men’s and women’s perceived health . J Health Soc Behav . 1994 ; 35 : 161 – 178 . Google Scholar CrossRef Search ADS PubMed 29. Singh GK , Kenney MK , Ghandour RM , Kogan MD , Lu MC . Mental health outcomes in US children and adolescents born prematurely or with low birthweight . Depress Res treat . 2013 ; 2013 : 570743 . Google Scholar PubMed 30. Power C , Elliott J . Cohort profile: 1958 British birth cohort (National Child Development Study) . Int J Epidemiol . 2006 ; 35 : 34 – 41 . Google Scholar CrossRef Search ADS PubMed 31. Stott DH. The social-adjustment of children: Manual to the bristol social-adjustment guides . London, UK: University of London Press ; 1963 . 32. Rutter M . A children’s behaviour questionnaire for completion by teachers: Preliminary findings . J Child Psychol Psychiatry Allied Disciplines . 1967 ; 8 : 1 – 11 . Google Scholar CrossRef Search ADS 33. Shepherd P. Bristol social adjustment guides at 7 and 11 years . London, UK : Centre for Longitudinal Studies, Institute of Education, University of London , 2013 . 34. Tuddenham RD . Review of Bristol social adjustment guides . J Consult Psychol . 1960 ; 24 : 99 . Google Scholar CrossRef Search ADS 35. Cable N , Bartley M , McMunn A , Kelly Y . Gender differences in the effect of breastfeeding on adult psychological well-being . Eur J Public Health . 2012 ; 22 : 653 – 658 . Google Scholar CrossRef Search ADS PubMed 36. Clark C , Rodgers B , Caldwell T , Power C , Stansfeld S . Childhood and adulthood psychological ill health as predictors of midlife affective and anxiety disorders: The 1958 British Birth Cohort . Arch Gen Psychiatry 2007 ; 64 : 668 – 678 . Google Scholar CrossRef Search ADS PubMed 37. Sourander A , Jensen P , Davies M et al. Who is at greatest risk of adverse long-term outcomes? The Finnish From a Boy to a Man study . J Am Acad Child Adolesc Psychiatry . 2007 ; 46 : 1148 – 1161 . Google Scholar CrossRef Search ADS PubMed 38. Jokela M , Ferrie J , Kivimaki M . Childhood problem behaviors and death by midlife: The British National Child Development Study . J Am Acad Child Adolesc Psychiatry . 2009 ; 48 : 19 – 24 . Google Scholar CrossRef Search ADS PubMed 39. Lilienfeld SO . Comorbidity between and within childhood externalizing and internalizing disorders: Reflections and directions . J Abnorm Child Psychol . 2003 ; 31 : 285 – 291 . Google Scholar CrossRef Search ADS PubMed 40. Kitchens JM . Does this patient have an alcohol problem ? JAMA . 1994 ; 272 : 1782 – 1787 . Google Scholar CrossRef Search ADS PubMed 41. Mukamal KJ , Chiuve SE , Rimm EB . Alcohol consumption and risk for coronary heart disease in men with healthy lifestyles . Arch Intern Med . 2006 ; 166 : 2145 – 2150 . Google Scholar CrossRef Search ADS PubMed 42. Parsons TJ , Power C , Manor O . Longitudinal physical activity and diet patterns in the 1958 British Birth Cohort . Med Sci Sports Exerc . 2006 ; 38 : 547 – 554 . Google Scholar CrossRef Search ADS PubMed 43. Parsons TJ , Manor O , Power C . Changes in diet and physical activity in the 1990s in a large British sample (1958 birth cohort) . Eur J Clin Nutr . 2005 ; 59 : 49 – 56 . Google Scholar CrossRef Search ADS PubMed 44. Pringle M , Butler N , Davie R. 11,000 Seven-year olds: First report of the National Child Development Study (1958 Cohort) . London : Longman , 1967 . 45. Southgate V. Southgate group reading tests . London : University of London , 1962 . 46. Atherton K , Fuller E , Shepherd P , Strachan DP , Power C . Loss and representativeness in a biomedical survey at age 45 years: 1958 British birth cohort . J Epidemiol Community Health . 2008 ; 62 : 216 – 223 . Google Scholar CrossRef Search ADS PubMed 47. Cole TJ , Bellizzi MC , Flegal KM , Dietz WH . Establishing a standard definition for child overweight and obesity worldwide: International survey . BMJ . 2000 ; 320 : 1240 – 1243 . Google Scholar CrossRef Search ADS PubMed 48. Rutter M , Tizard J , Whitmore K. Education, health and behaviour . London : Longmans ; 1970 . 49. Gilman SE , Rende R , Boergers J et al. Parental smoking and adolescent smoking initiation: An intergenerational perspective on tobacco control . Pediatrics . 2009 ; 123 : e274 – e281 . Google Scholar CrossRef Search ADS PubMed 50. Hanson MD , Chen E . Socioeconomic status and health behaviors in adolescence: A review of the literature . J Behav Med . 2007 ; 30 : 263 – 285 . Google Scholar CrossRef Search ADS PubMed 51. Roshanaei-Moghaddam B , Katon WJ , Russo J . The longitudinal effects of depression on physical activity . Gen Hosp Psychiatry . 2009 ; 31 : 306 – 315 . Google Scholar CrossRef Search ADS PubMed 52. Gapin JI , Labban JD , Etnier JL . The effects of physical activity on attention deficit hyperactivity disorder symptoms: The evidence . Prev Med . 2011 ; 52 ( Suppl ): S70 – S74 . Google Scholar CrossRef Search ADS PubMed 53. Trudel-Fitzgerald C , Tworoger SS , Poole EM , Williams DR , Kubzansky LD . Prospective changes in healthy lifestyle among midlife women: When psychological symptoms get in the way . Am J Prev Med . 2016 ; 51 : 327 – 335 . Google Scholar CrossRef Search ADS PubMed 54. Gibson EL . The psychobiology of comfort eating: Implications for neuropharmacological interventions . Behav Pharmacol . 2012 ; 23 : 442 – 460 . Google Scholar CrossRef Search ADS PubMed 55. Gehricke JG , Loughlin SE , Whalen CK et al. Smoking to self-medicate attentional and emotional dysfunctions . Nicotine Tob Res . 2007 ; 9 ( Suppl 4 ): S523 – S536 . Google Scholar CrossRef Search ADS PubMed 56. Bolton JM , Robinson J , Sareen J . Self-medication of mood disorders with alcohol and drugs in the National Epidemiologic Survey on Alcohol and Related Conditions . J Affect Disord . 2009 ; 115 : 367 – 375 . Google Scholar CrossRef Search ADS PubMed 57. Colman I , Murray J , Abbott RA et al. Outcomes of conduct problems in adolescence: 40 year follow-up of national cohort . BMJ . 2009 ; 338 : a2981 . Google Scholar CrossRef Search ADS PubMed 58. Hagger MS , Wood C , Stiff C , Chatzisarantis NLD . The strength model of self-regulation failure and health-related behaviour . Health Psychol Rev . 2009 ; 3 : 208 – 238 . Google Scholar CrossRef Search ADS 59. McGue M , Bouchard TJ , Jr . Genetic and environmental influences on human behavioral differences . Ann Rev Neurosci . 1998 ; 21 : 1 – 24 . Google Scholar CrossRef Search ADS PubMed 60. Clark C , Caldwell T , Power C , Stansfeld SA . Does the influence of childhood adversity on psychopathology persist across the lifecourse? A 45-year prospective epidemiologic study . Ann Epidemiol . 2010 ; 20 : 385 – 394 . Google Scholar CrossRef Search ADS PubMed 61. Grant KE , Compas BE , Thurm AE , McMahon SD , Gipson PY . Stressors and child and adolescent psychopathology: Measurement issues and prospective effects . J Clin Child and Adolesc Psychol . 2004 ; 33 : 412 – 425 . Google Scholar CrossRef Search ADS 62. Reiss F . Socioeconomic inequalities and mental health problems in children and adolescents: A systematic review . Soc Sci Med . 2013 ; 90 : 24 – 31 . Google Scholar CrossRef Search ADS PubMed 63. O’Leary-Barrett M , Topper L , Al-Khudhairy N et al. Two-year impact of personality-targeted, teacher-delivered interventions on youth internalizing and externalizing problems: A cluster-randomized trial . J Am Acad Child Adolesc Psychiatry . 2013 ; 52 : 911 – 920 . Google Scholar CrossRef Search ADS PubMed 64. Waddell C , Hua JM , Garland OM , Peters RD , McEwan K . Preventing mental disorders in children: A systematic review to inform policy-making . Can J Public Health . 2007 ; 98 : 166 – 173 . Google Scholar PubMed 65. Weare K , Nind M . Mental health promotion and problem prevention in schools: What does the evidence say ? Health Promot Int . 2011 ; 26 ( Suppl 1 ): i29 – i69 . Google Scholar CrossRef Search ADS PubMed 66. Stewart-Brown SL , Schrader-McMillan A . Parenting for mental health: What does the evidence say we need to do? Report of Workpackage 2 of the DataPrev project . Health Promot Int . 2011 ; 26 ( Suppl 1 ): i10 – i28 . Google Scholar CrossRef Search ADS PubMed 67. Durlak JA , Wells AM . Primary prevention mental health programs for children and adolescents: A meta-analytic review . Am J Community Psychol . 1997 ; 25 : 115 – 152 . Google Scholar CrossRef Search ADS PubMed 68. Chaiton MO , Cohen JE , O’Loughlin J , Rehm J . A systematic review of longitudinal studies on the association between depression and smoking in adolescents . BMC Public Health . 2009 ; 9 : 356 . Google Scholar CrossRef Search ADS PubMed 69. Wang JC , Hinrichs AL , Stock H et al. Evidence of common and specific genetic effects: Association of the muscarinic acetylcholine receptor M2 (CHRM2) gene with alcohol dependence and major depressive syndrome . Hum Mol Gen . 2004 ; 13 : 1903 – 1911 . Google Scholar CrossRef Search ADS PubMed 70. Stavrakakis N , de Jonge P , Ormel J , Oldehinkel AJ . Bidirectional prospective associations between physical activity and depressive symptoms. The TRAILS Study . J Adolesc Health . 2012 ; 50 : 503 – 508 . Google Scholar CrossRef Search ADS PubMed 71. Humphreys KL , Mehta N , Lee SS . Association of parental ADHD and depression with externalizing and internalizing dimensions of child psychopathology . J Atten Disord . 2012 ; 16 : 267 – 275 . Google Scholar CrossRef Search ADS PubMed 72. Shonkoff JP , Garner AS . The lifelong effects of early childhood adversity and toxic stress . Pediatrics . 2012 ; 129 : e232 – e246 . Google Scholar CrossRef Search ADS PubMed © Society of Behavioral Medicine 2018. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Journal

Annals of Behavioral MedicineOxford University Press

Published: Jan 24, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off