Mental disorders and stress resilience in adolescence and long-term risk of early heart failure among Swedish men

    Mental Disorders and Stress Resilience in Adolescence and Long-Term Risk of Early Heart Failure among Swedish Men Josefina Robertson, Linus Schi¨oler, Kjell Tor´en, Mia S¨oderberg, Jesper ˚ L¨ove, Margda Waern, Annika Rosengren, Maria Aberg PII: DOI: Reference:

S0167-5273(16)34479-5 doi:10.1016/j.ijcard.2017.05.043 IJCA 25001

To appear in:

International Journal of Cardiology

Received date: Revised date: Accepted date:

30 December 2016 17 April 2017 9 May 2017

Please cite this article as: Robertson Josefina, Schi¨ oler Linus, Tor´en Kjell, S¨ oderberg Mia, L¨ ove Jesper, Waern Margda, Rosengren Annika, ˚ Aberg Maria, Mental Disorders and Stress Resilience in Adolescence and Long-Term Risk of Early Heart Failure among Swedish Men, International Journal of Cardiology (2017), doi:10.1016/j.ijcard.2017.05.043

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

1

SC RI

PT

Mental Disorders and Stress Resilience in Adolescence and Long-Term Risk of Early Heart Failure among Swedish Men

NU

Josefina Robertson, MD1, Linus Schiöler, PhD2, Kjell Torén, MD, PhD2, Mia Söderberg, PhD2, Jesper Löve, PhD3, Margda Waern, MD, PhD4, Annika Rosengren, MD, PhD5, and Maria Åberg, MD, PhD1.

1

MA

Department of Public Health and Community Medicine/Primary Health Care, Institute of Medicine, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden 2

ED

Section of Occupational and Environmental Medicine, Department of Public Health and Community Medicine, Institute of Medicine, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden 3

PT

Section of Social Medicine and Epidemiology, Institute of Medicine, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden 4

5

CE

Department of Psychiatry and Neurochemistry, Institute of Neuroscience and Physiology, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden

AC

Department of Molecular and Clinical Medicine, Institute of Medicine, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden

Corresponding author Dr Maria Åberg, Department of Public Health and Community Medicine/Primary Health Care, Institute of Medicine, Box 454, SE-405 30 Gothenburg, Sweden, e-mail: [email protected]; ph: 46-709668274, fax: + 46 31 7781704.

ACCEPTED MANUSCRIPT

2

Abstract Background Recent research suggests that the prevalence of early heart failure may be on the rise. Compromised mental

PT

health in adolescence may help to explain this phenomenon. We aimed to investigate whether nonpsychotic mental disorder and low stress resilience in late adolescence were associated with increased risk of early

SC RI

heart failure.

Methods

A prospective cohort study of 18-year-old men (n=1 784 450) who enlisted 1968-2005. At the conscription

NU

examination, 74 522 individuals were diagnosed with nonpsychotic mental disorders. Stress resilience was

MA

rated by psychologists; values were trichotomized. The risk of heart failure during the 46-year follow-up was calculated with Cox proportional hazards models. Baseline comorbidities, BMI, blood pressure, fitness, IQ,

ED

and parental education were included in the models.

Results

PT

Incident cases of heart failure (n= 9 962) were identified in the National Hospital Register. In fully adjusted models, increased risk of early heart failure was observed in males diagnosed with nonpsychotic mental

CE

disorders at conscription (hazard ratio (HR), 1.36; 95% confidence interval (CI), 1.25-1.47). The highest risk

AC

was seen among men with the risk factor alcohol/substance use (HR 1.90; 95% CI 1.59-2.28). Conscripts with the risk factor low stress resilience showed increased risk of heart failure compared to those with high scores (HR 1.41; 95% CI 1.30-1.53).

Conclusion Nonpsychotic mental disorder, as well as low stress resilience in late adolescence may be associated with increased risk of early heart failure. Adolescence is potentially an important time for mental health interventions that may reduce both short and long-term consequences.

Keywords Heart failure – mental disorders – stress resilience – adolescence – population

ACCEPTED MANUSCRIPT

3

Introduction Heart failure is the final outcome of several cardiac diseases and one of the greatest global health problems (1). It is a devastating condition, with a prognosis comparable to that of many common forms of cancer (2,

PT

3). According to a recent study, the incidence seems to be increasing among persons below the age of 45 (4),

SC RI

in contrast to an overall decrease in older persons (5, 6). At the same time, depression, anxiety, nervousness, and hospitalizations for alcohol and drug use are increasing among young Swedish citizens (7). Increased cardiovascular morbidity in schizophrenia and other psychoses is well known. There is also evidence of

NU

increased risk of cardiovascular illness in persons with nonpsychotic mental disorders (8, 9). The linkage is not completely understood, but individuals with mental disorders often present risk factors such as poor

MA

health behaviors, obesity, hypertension, and diabetes (10). In addition, some mental disorders and heart disease seem to have a shared genetic pathway (11). Several longitudinal cohort studies support an

ED

association between mental disorders and heart failure (12, 13). However, to our knowledge, there are no

PT

studies investigating whether mental disorder in youth is a risk factor for future heart failure.

Stress is associated with numerous unfavorable health outcomes, including cardiovascular diseases (14).

CE

Stress resilience is a significant determinant of the consequences of exposure to stressing events and environments. It is commonly defined as an individual’s ability to successfully adapt to stressing tasks or

AC

highly adverse conditions, while maintaining normal psychological and physical functioning (15). Low stress resilience might end up in an extended physiological response to stressful incidents, which increases the eventual undesirable long-term consequences of chronic stress. Previous studies have shown that low stress resilience is associated with an increased risk of both stroke and coronary heart disease (16, 17). So far, to our knowledge, no studies have investigated stress resilience in relation to risk of heart failure.

Our first aim was to investigate whether nonpsychotic mental disorder in late adolescence is associated with an increased risk of future heart failure. Our second aim was to determine whether low stress resilience in late adolescence increases the risk of future heart failure. We therefore performed a prospective cohort study of all Swedish men who underwent psychiatric and physical examinations when enlisting for compulsory military service at the age of 18 years. These men were followed up to 46 years. All incident cases can be

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA

NU

SC RI

PT

considered early heart failure, as the oldest participants were ≤65 years at the time of diagnosis.

4

ACCEPTED MANUSCRIPT

5

Methods Participants A cohort of young individuals who enlisted for military service between 1968 and 2005 (ie, born between

PT

1950 and 1987) was extracted from the Swedish Military Service Conscription Register (n=1 886 542). At

SC RI

that time, all Swedish males had to enlist according to the Swedish law. Exceptions were made regarding those who were imprisoned or had severe chronic somatic or mental conditions or functional disabilities documented by a medical certificate (approximately 2–3% annually). Exclusion criteria were: late allocated

NU

personal identity number (n=1 154), women (n=10 228), men aged >25 years (n=54 651), heart failure diagnosis before or at conscription (n=176), men with psychotic disorders before or at conscription (n=1

MA

566) and men with psychotic disorders during the follow-up (hospital admissions for schizophrenia, other non-affective psychoses, and bipolar disorder) (n=34 317). The remaining 1 784 450 subjects were included

ED

in the present study (Fig. 1).

PT

Conscription Register Data The men were examined by psychologists and physicians at any of six conscription centers during two days

CE

according to a standardized procedure. This included measurement of weight, height and blood pressure.

AC

Psychiatric diagnosis A psychologist evaluated psychiatric symptoms during a structured interview. Conscripts with such symptoms were referred to a physician. Psychiatric disorders were diagnosed according to the International Classification of Diseases (ICD). For the purpose of this study, the diagnostic codes were divided into three subgroups; a) depressive disorders and neurotic/adjustment disorders, b) personality disorders, c) alcoholrelated disorders and other substance use disorders. To avoid misclassification of conscripts with prodromal episodes of psychotic disorders, individuals with hospital admissions for those disorders were excluded (Supplementary Table 1). Stress Resilience Stress resilience was rated in 1 602 053 conscripts (89.8% of the study population). The purpose of this assessment was to evaluate the ability to cope with wartime stress. The psychological examination and semistructured interview included questions about predisposition to anxiety, ability to control and channel

ACCEPTED MANUSCRIPT

6

nervousness, and stress tolerance. Factors relevant to everyday life, such as interests, recreational activities, psychological motivation, social maturity, and emotional stability, were also registered. The answers were summarized in a stress resilience score (1–9), which was grouped into low (1-3), medium (4-7), and high (8-

PT

9). Higher values indicated greater resilience. To ensure consistent evaluation over time and between

SC RI

different test centers, a written manual was used. Stress resilience data have been published in prior research (16).

NU

Cardiovascular Fitness Test Cardiovascular fitness was evaluated by a cycle ergometric test. During exercise the work rate was successively increased until limited by exhaustion. The final work rate (Wmax) was recorded and divided by

MA

body weight. The resulting values (Wmax/kg) were converted into stanine scores that served as a measure of fitness. The test has previously been described in detail (18-20).

ED

Intelligence Quotient A measure of combined intelligence (IQ) was calculated from the results of four tests (logical performance,

PT

verbal test of synonyms and opposites, test of visuospatial/geometric perception, technical/mechanical skills

CE

including mathematical/physics problems) (21). Test results were standardized against data from previous years to give scores from 1-9 (low to high). The standardization to a stanine scale provided long-term

AC

stability of the data sets.

Parental education Information on parental education was derived from the longitudinal integration database for health insurance and labor market studies (Swedish acronym LISA, 80% coverage). It is divided into seven levels: <9 years, pre-high school education of 9 years, high school education, university (<2 years), university (≥2 years), postgraduate education, and postgraduate research training. The parent with the highest education level was chosen. Outcomes Data sources The universal healthcare system in Sweden offers low-cost outpatient and hospital care to all citizens. All Swedes have a unique personal identity number making linkage to other registers possible. Using this number, data from the Swedish Military Service Conscription Register was linked to the Swedish Hospital

ACCEPTED MANUSCRIPT

7

Discharge and Cause of Death registries. Swedish Hospital Discharge Register coverage gradually increased during the period 1968 to 1986, and is complete since 1987. Each patient is given a principal diagnosis and if necessary, one or more secondary diagnoses at discharge. Diagnoses in hospital outpatient care are recorded

PT

from 2001.

SC RI

Heart failure Heart failure (HF) incidents were identified by the national Hospital Discharge Register. The cases were classified according to the International Classification of Diseases (ICD). The eighth revision (ICD-8) was

NU

in use from 1968 to 1986, the ninth revision (ICD-9) from 1987 to 1996 and the tenth revision (ICD-10) from 1997. HF was defined by ICD codes 427.00 and 427.10 for ICD-8, 428 for ICD-9 and I50 for ICD-10.

MA

Because a large proportion of patients with HF had other primary diagnoses (for example cardiomyopathy), a first time HF diagnosis code in any position was assessed as HF. The following diagnoses were included as

ED

concomitant or pre-existing comorbidities at baseline: diabetes 250 (ICD-8 and 9), E10-E14 (ICD-10); hypertension 401-405 (ICD-8 and 9), I10-I15 (ICD-10); congenital heart disease 746-747 (ICD-8), 745-747

PT

(ICD-9), Q20-Q28, Q87, Q89 (ICD-10).

CE

Statistical Analysis The statistical calculations were performed with SAS version 9.4 (SAS Institute, Cary, NC). The follow-up

AC

period started at the date of conscription (baseline) and subjects were censored at the time of: 1) first heart failure hospitalization, 2) death, 3) emigration, or 4) at the end of follow-up, December 31, 2014 (minimum 9 years, maximum 46 years). Incidence rates and corresponding 95% confidence intervals were calculated using Poisson regression. Cox proportional hazards models were used to assess the influences of nonpsychotic mental disorder and stress resilience at the age of 18 years and potential confounders on a first hospitalization for heart failure. Since differences among regions, test centers, and examination years could result in bias, these were considered potential confounders and adjusted for. Further, as early obesity (22) and hypertension (23) are risk factors for heart failure, adjustments for the continuous variables body mass index and systolic and diastolic blood pressures were performed. Also alcohol use disorders are known risk factors for heart failure (24) and these conditions may co-exist with depressive, neurotic, and personality disorders. Therefore

ACCEPTED MANUSCRIPT

8

models including alcohol/substance use disorder as a potential confounder were constructed. To examine how cardiovascular fitness, IQ, and parental education would impact the association between

PT

nonpsychotic mental disorder in late adolescence and risk of heart failure in adulthood, we included these variables as covariates. They were trichotomized as low, medium, and high. Because of the large number of

SC RI

observations, the P-values were very small and not reported (P-values 0.0001 in all analyses when the 95% confidence interval was separated from 1).

NU

To study effect modification of fitness and IQ in late adolescence on the associations between nonpsychotic mental disorder and risk of heart failure, we included fitness and IQ as interaction terms. The interaction

MA

terms (fitness×nonpsychotic disorders and IQ×nonpsychotic disorders) were introduced together in model 3. Population-attributable risk, that is the association of a specific risk factor with a specific disease as a

ED

proportion of all risk factors for that disease, was calculated by the method of Natarajan et al using the

CE

PT

hazard ratios (HRs) from the Cox proportional hazard regression (25).

The Ethics Committee of the University of Gothenburg and Confidentiality Clearance at Statistics Sweden

AC

approved the study. The investigation conforms to the principles outlined in the Declaration of Helsinki.

ACCEPTED MANUSCRIPT

9

Results Study Population Of the 1 784 450 conscripts in the study (mean age 18.3 years, SD 0.7), 74 522 were diagnosed with a

PT

nonpsychotic mental disorder (Fig 1). Baseline data for parental education, IQ, cardiovascular fitness, and stress resilience are shown in Table 1. The group with nonpsychotic mental disorders had higher proportions

SC RI

of individuals with poor fitness and poor IQ compared to those without mental disorders. Among men with nonpsychotic mental disorders, 70.7% had low stress resilience. Low parental education level was noted in 44% in the nonpsychotic mental disorder group, compared to 27% in the group without mental disorders.

NU

The group with high stress resilience had higher proportions of individuals with high fitness and high IQ

MA

compared to those with low and medium stress resilience (Table 1).

Follow-up

The relative frequency of nonpsychotic mental disorder diagnoses among the conscripts decreased over time.

ED

The time of conscription was therefore divided into decades (-1970 (9.1%, n=4 898), 1971-1980 (7.8%, n=

PT

37 805), 1981-1990 (4.0%, n= 21 638), 1991-2000 (1.8%, n= 8 637), 2001- (0.7%, n= 1 544)). A sensitivity analysis showed that the risk of heart failure was consistent over time, despite decreasing frequencies of

CE

nonpsychotic mental disorders. Hence, we decided to use the total period in the further analyses.

AC

During the 46-year follow-up (median 26 years, interquartile range 17-34; 45 823 381 person years), there were 9 962 unique cases of hospitalization for heart failure (principal or secondary discharge diagnosis) (Fig. 1). The mean age at first hospitalization for heart failure diagnosis was 49 years (SD 8.9) (Table 1). The proportion with heart failure in the nonpsychotic mental disorder group was 2.3 times greater than the proportion in the rest of the population (50.4 versus 21.7 cases per 100 000 observation years, P<0.0001). The cumulative incidence (%) of heart failure during the follow-up was greater in the nonpsychotic mental disorder group compared to the group without mental disorders. The risk of heart failure seemed to increase with increasing follow-up time (Fig. 2). Risk factors for heart failure All nonpsychotic mental disorders were considered together in the initial analyses. In agreement with our first aim, the presence of a nonpsychotic mental disorder diagnosis in late adolescence was associated with

ACCEPTED MANUSCRIPT

10

increased risk of future heart failure, HR 1.60 (95% CI 1.50-1.71) (Table 2). The increased risk was independent of the cause of heart failure (congenital heart disease/valvulopathies, coronary heart disease/diabetes/hypertension, cardiomyopathy or other causes). The strength of the association slightly

PT

changed in models adjusted for comorbidities at baseline, BMI, blood pressure, fitness, IQ, and parental

SC RI

education. Separate analyses for the different diagnostic categories (depressive/neurotic, personality disorders, and alcohol/substance use disorders) showed that all these groups were associated with future risk of heart failure. The strongest association was found with alcohol/substance use disorders, HR 1.90 (95% CI

NU

1.59-2.28) in the fully adjusted model (Table 2). When pooling depressive/neurotic disorders together with personality disorders, and adjusting for concomitant alcohol/substance use, there was a marginal attenuation

MA

of the HR (1.49 (95% CI 1.39-1.60) compared to 1.60 (95% CI 1.50-1.71) in model 3.

ED

Effect modification of fitness and IQ To examine whether fitness and IQ in late adolescence modified the associations between nonpsychotic mental disorder and risk of heart failure, fitness and IQ were included as interaction terms in model 3. For

PT

fitness the interaction term was statistically nonsignificant (P=0.13) and therefore no further analyses were

CE

performed. However, for IQ the interaction term was statistically significant (P=0.03). We estimated hazard ratios for all combinations of IQ and npm, using no mental disorder and high IQ as reference group (HR

AC

1.00). This resulted in, for the combination npm and low IQ, HR 2.80 (CI 2.52-3.11); medium IQ, HR 2.01 (CI 1.81-2.24); and high IQ, HR 1.75 (CI 1.46-2.10). The hazard ratios for the combination of no npm and low IQ was 2.10 (CI 1.96-2.24), and for no npm and medium IQ 1.37 (CI 1.29-1.45). The model was adjusted for age at conscription, conscription year, test center, comorbidities at baseline, BMI, and systolic and diastolic blood pressure. Stress resilience Of the study population, 19.4 % was classified as having high (7-9 p) stress resilience, 52.4% medium (4-6 p), and 18.0% low (1-3 p). There was an increased risk of heart failure for individuals with low stress resilience scores compared to those with high scores (Table 2). For the low and medium groups, HRs were 2.09 (1.96-2.23) and 1.27 (1.20-1.35), respectively. This indicates a graded association. When additionally adjusting for concomitant nonpsychotic mental disorder, fitness, IQ, and parental education, the associations

ACCEPTED MANUSCRIPT

11

were slightly attenuated but still significant in the group with low stress resilience. The Pearson correlation coefficient regarding stress resilience and nonpsychotic mental disorders at conscription was -0.29. A graded association was also seen between stress resilience at age 18 and future risk of nonpsychotic mental

AC

CE

PT

ED

MA

NU

SC RI

PT

disorders with HRs 2.85 (2.80-2.90) for the low group and 1.39 (1.36-1.41) for the medium group.

ACCEPTED MANUSCRIPT

12

Discussion Nonpsychotic mental disorder in late adolescence was associated with increased risk of early heart failure in this large cohort of young men. Relationships were consistent for all three subgroups of nonpsychotic mental

PT

disorders, although strongest for alcohol/substance use disorders. Associations remained significant in all of

SC RI

the adjusted models. Moreover, we also found that low stress resilience in adolescence was associated with increased risk of early heart failure.

NU

Previous studies have shown that mental disorders are associated with an increased risk of cardiovascular diseases (8, 9). A weak association between some mental disorders and incidence of heart failure, has also

MA

been reported (12, 13). The current study adds to the literature by showing an association with nonpsychotic mental disorders that are present already in youth. It parallels recent findings from the Swedish Military

ED

Service Conscription Register, with similar associations with nonpsychotic mental disorder and both future risk of early stroke (26), and increased mortality starting at the age of 30 years (27). The linkage between

PT

mental disorders and heart failure is not well understood but individuals with mental disorders often present

CE

risk factors such as adverse lifestyle factors, obesity, hypertension, and diabetes (10). In our study, the strongest association was found with alcohol/substance use disorders at the age of 18 and risk of heart

AC

failure. This was anticipated, because of the known presence of alcohol-associated cardiomyopathy, which can lead to heart failure (24). When we adjusted for alcohol/substance use disorders as potential mediating factors in the etiology of heart failure in the other mental disorder categories, the association between nonpsychotic mental disorders and heart failure remained. This indicates that also other etiologies may have been at play. Possible biological mediating pathways could be inflammation, endothelial dysfunction, neurohormonal and autonomic nervous system dysfunction, and low levels of brain-derived neurotrophic factor (BDNF) (28). As reported in a twin study, depression and cardiac disease also seemed to share common genetic factors in addition to life-style factors (11). In this study, effect modifications showed that adolescent IQ, but not cardiovascular fitness, might modify the association between nonpsychotic mental disorders and early heart failure. The analyses indicated a graded association with the highest risk of heart failure in the low IQ group compared to high IQ. Similar

ACCEPTED MANUSCRIPT

13

negative association between childhood IQ and cardiovascular risk factors and events later in life, has previously been presented in a prospective study (29). But this issue needs to be further explored. Stress is known to be a risk factor for coronary heart disease (14). A weak relationship was observed

PT

between stress and risk of heart failure in a study of Ogilvie et al, with a 2-fold higher incidence of heart

SC RI

failure among individuals with chronic stress and poor health. However, no association was seen in individuals with good health despite chronic stress (30). In the present study, we showed that low stress resilience in youth increased the risk of future heart failure. This could possibly be explained by assuming

NU

that stress resilience in adolescence designates ability to cope with adversities later in life. Lack of resilience may therefore result in higher levels of stress during adulthood. The association between stress resilience

MA

and heart failure remained even when adjusting for concomitant nonpsychotic mental disorders. While the correlation coefficient between stress resilience and nonpsychotic mental disorders at baseline was not more

ED

than 0.29, stress resilience seemed to be a strong predictor for future nonpsychotic mental disorder. The latter could have mediated the association between stress resilience and future heart failure. Although stress

PT

resilience score at conscription is based on interview questions and no validation test has been done as far as

CE

we know. Thus, our findings regarding adolescent stress resilience must be seen as exploratory.

AC

Strengths and limitations Strengths of this study include the large sample size (>1.7 million participants) and the prospective population-based design. The study population is largely representative of the male general population as only a small proportion of Swedish men were exempted from conscription during that time. A large number of cases of heart failure were recorded during an extended follow-up. The hospital diagnosis of heart failure in Sweden has been shown to have high validity, with a positive predictive value of 82-88% (31). However, there are also some limitations to our study. The study population is homogeneous, with a majority of Caucasian males at the age of 18, which confines the generalizability to other populations, and to women. It has previously been reported that 0.7% of conscripts had a history of psychiatric hospitalization before conscription. Only 18% of these were diagnosed with a mental disorder at conscription. The relatively low proportion may in part be explained by recovery or failure to report symptoms (9), which would result in

ACCEPTED MANUSCRIPT

14

more conservative estimates. Further, the prevalence of conscripts with nonpsychotic mental disorders decreased over time. A possible explanation for this is an increasing practice of excusing young men with some medical diagnoses from enlistment. That could result in a selection bias where young men with more

PT

serious effects on, e.g. psychosocial functioning, were excused from conscription in a higher proportion. If

SC RI

so, the current estimations of risk for future heart failure may be underestimated. Finally, we were not able to study development of risk factors such as smoking and obesity during the follow-up. Therefore, it is not possible to discuss the influence of these variables on the outcome.

NU

Conclusion

In conclusion, we found that nonpsychotic mental disorder, as well as low stress resilience in late

MA

adolescence may be associated with increased risk of early heart failure. Therefore, adolescence is potentially an important time for mental health interventions that may reduce both short and long-term

AC

CE

PT

ED

consequences.

ACCEPTED MANUSCRIPT

15

Funding This study was supported by grants from the Swedish state under an agreement concerning economic support of research and education of doctors [ALFGBG-427301, 433511], the Swedish Research Council

PT

2013-2699, 2013-5187, 2013-4236, the Swedish Heart and Lung Foundation [2015-0438]; and the Swedish

SC RI

Council for Health, Working Life and Welfare (FORTE) [2007-2280, 2013-0325].

Conflict of interest

AC

CE

PT

ED

MA

NU

None.

ACCEPTED MANUSCRIPT

16

References 1.

Kannel WB. Incidence and epidemiology of heart failure. Heart failure reviews. 2000 Jun;5(2):167-73. PubMed PMID: 16228142. Stewart S, Ekman I, Ekman T, Oden A, Rosengren A. Population impact of heart failure and the most

PT

2.

SC RI

common forms of cancer: a study of 1 162 309 hospital cases in Sweden (1988 to 2004). Circulation Cardiovascular quality and outcomes. 2010 Nov;3(6):573-80. PubMed PMID: 20923990. 3.

Agrinier N, Thilly N, Briancon S, et al. Prognostic factors associated with 15-year mortality in patients

NU

with hospitalized systolic HF: Results of the observational community-based EPICAL cohort study. International journal of cardiology. 2016 Nov 12;228:940-7. PubMed PMID: 27912203. Barasa A, Schaufelberger M, Lappas G, Swedberg K, Dellborg M, Rosengren A. Heart failure in

MA

4.

young adults: 20-year trends in hospitalization, aetiology, and case fatality in Sweden. European heart

5.

ED

journal. 2014 Jan;35(1):25-32. PubMed PMID: 23900697. Pubmed Central PMCID: PMC3877433. Paren P, Schaufelberger M, Bjorck L, Lappas G, Fu M, Rosengren A. Trends in prevalence from 1990

PT

to 2007 of patients hospitalized with heart failure in Sweden. European journal of heart failure. 2014

6.

CE

Jul;16(7):737-42. PubMed PMID: 24863749. Schaufelberger M, Swedberg K, Koster M, Rosen M, Rosengren A. Decreasing one-year mortality and

AC

hospitalization rates for heart failure in Sweden; Data from the Swedish Hospital Discharge Registry 1988 to 2000. European heart journal. 2004 Feb;25(4):300-7. PubMed PMID: 14984918. 7.

Lager A, Berlin M, Heimerson I, Danielsson M. Young people's health: Health in Sweden: The National Public Health Report 2012. Chapter 3. Scandinavian journal of public health. 2012 Dec;40(9 Suppl):42-71. PubMed PMID: 23238401.

8.

Van der Kooy K, van Hout H, Marwijk H, Marten H, Stehouwer C, Beekman A. Depression and the risk for cardiovascular diseases: systematic review and meta analysis. International journal of geriatric psychiatry. 2007 Jul;22(7):613-26. PubMed PMID: 17236251.

9.

Gale CR, Batty GD, Osborn DP, Tynelius P, Rasmussen F. Mental disorders across the adult life course and future coronary heart disease: evidence for general susceptibility. Circulation. 2014 Jan 14;129(2):186-93. PubMed PMID: 24190959. Pubmed Central PMCID: PMC4107269.

ACCEPTED MANUSCRIPT 10.

17

Perez-Pinar M, Mathur R, Foguet Q, Ayis S, Robson J, Ayerbe L. Cardiovascular risk factors among patients with schizophrenia, bipolar, depressive, anxiety, and personality disorders. European psychiatry : the journal of the Association of European Psychiatrists. 2016 May;35:8-15. PubMed

Scherrer JF, Xian H, Bucholz KK, et al. A twin study of depression symptoms, hypertension, and heart

SC RI

11.

PT

PMID: 27061372.

disease in middle-aged men. Psychosomatic medicine. 2003 Jul-Aug;65(4):548-57. PubMed PMID: 12883104.

Gustad LT, Laugsand LE, Janszky I, Dalen H, Bjerkeset O. Symptoms of anxiety and depression and

NU

12.

risk of heart failure: the HUNT Study. European journal of heart failure. 2014 Aug;16(8):861-70.

13.

MA

PubMed PMID: 25044493. Pubmed Central PMCID: PMC4255780. Garfield LD, Scherrer JF, Hauptman PJ, et al. Association of anxiety disorders and depression with

ED

incident heart failure. Psychosomatic medicine. 2014 Feb;76(2):128-36. PubMed PMID: 24434950. Pubmed Central PMCID: PMC3946309. Rosengren A, Hawken S, Ounpuu S, et al. Association of psychosocial risk factors with risk of acute

PT

14.

CE

myocardial infarction in 11119 cases and 13648 controls from 52 countries (the INTERHEART study): case-control study. Lancet (London, England). 2004 Sep 11-17;364(9438):953-62. PubMed

15.

AC

PMID: 15364186.

Wu G, Feder A, Cohen H, et al. Understanding resilience. Frontiers in behavioral neuroscience. 2013;7:10. PubMed PMID: 23422934. Pubmed Central PMCID: PMC3573269.

16.

Bergh C, Udumyan R, Fall K, Nilsagard Y, Appelros P, Montgomery S. Stress resilience in male adolescents and subsequent stroke risk: cohort study. Journal of neurology, neurosurgery, and psychiatry. 2014 Dec;85(12):1331-6. PubMed PMID: 24681701. Pubmed Central PMCID: PMC4251543.

17.

Bergh C, Udumyan R, Fall K, Almroth H, Montgomery S. Stress resilience and physical fitness in adolescence and risk of coronary heart disease in middle age. Heart (British Cardiac Society). 2015 Apr;101(8):623-9. PubMed PMID: 25740818. Pubmed Central PMCID: PMC4396533.

ACCEPTED MANUSCRIPT 18.

18

Nordesjö LS, R. Validity of an ergometer cycle test and measures of isometric muscle strength when prediction some aspects of military performance. Swedish J Defence Med. 1974;10:11-23.

19.

Aberg MA, Nyberg J, Toren K, Sorberg A, Kuhn HG, Waern M. Cardiovascular fitness in early

20.

SC RI

2014 Mar;44(4):779-88. PubMed PMID: 23739044.

PT

adulthood and future suicidal behaviour in men followed for up to 42 years. Psychological medicine.

Nyberg J, Aberg MA, Schioler L, et al. Cardiovascular and cognitive fitness at age 18 and risk of early-onset dementia. Brain : a journal of neurology. 2014 May;137(Pt 5):1514-23. PubMed PMID:

21.

NU

24604561.

Aberg MA, Pedersen NL, Toren K, et al. Cardiovascular fitness is associated with cognition in young

MA

adulthood. Proceedings of the National Academy of Sciences of the United States of America. 2009 Dec 8;106(49):20906-11. PubMed PMID: 19948959. Pubmed Central PMCID: PMC2785721. Rosengren A, Aberg M, Robertson J, et al. Body weight in adolescence and long-term risk of early

ED

22.

PMID: 27311731.

Levy D, Larson MG, Vasan RS, Kannel WB, Ho KK. The progression from hypertension to

CE

23.

PT

heart failure in adulthood among men in Sweden. European heart journal. 2016 Jun 16. PubMed

congestive heart failure. Jama. 1996 May 22-29;275(20):1557-62. PubMed PMID: 8622246. Laonigro I, Correale M, Di Biase M, Altomare E. Alcohol abuse and heart failure. European journal of

AC

24.

heart failure. 2009 May;11(5):453-62. PubMed PMID: 19336433. 25.

Natarajan S, Lipsitz SR, Rimm E. A simple method of determining confidence intervals for population attributable risk from complex surveys. Statistics in medicine. 2007 Jul 30;26(17):3229-39. PubMed PMID: 17309113.

26.

Aberg MA, Toren K, Nilsson M, et al. Nonpsychotic Mental Disorders in Teenage Males and Risk of Early Stroke: A Population-Based Study. Stroke; a journal of cerebral circulation. 2016 Mar;47(3):814-21. PubMed PMID: 26846861. Pubmed Central PMCID: PMC4760382.

27.

Love J, Hensing G, Soderberg M, Toren K, Waern M, Aberg M. Future marginalisation and mortality in young Swedish men with non-psychotic psychiatric disorders and the resilience effect of cognitive

ACCEPTED MANUSCRIPT

19

ability: a prospective, population-based study. BMJ open. 2016;6(8):e010769. PubMed PMID: 27515748. Pubmed Central PMCID: PMC4985865. 28.

Huffman JC, Celano CM, Beach SR, Motiwala SR, Januzzi JL. Depression and cardiac disease:

PT

epidemiology, mechanisms, and diagnosis. Cardiovascular psychiatry and neurology.

29.

SC RI

2013;2013:695925. PubMed PMID: 23653854. Pubmed Central PMCID: PMC3638710. Hart CL, Taylor MD, Smith GD, et al. Childhood IQ and cardiovascular disease in adulthood: prospective observational study linking the Scottish Mental Survey 1932 and the Midspan studies.

30.

NU

Social science & medicine (1982). 2004 Nov;59(10):2131-8. PubMed PMID: 15351478. Ogilvie RP, Everson-Rose SA, Longstreth WT, Jr., Rodriguez CJ, Diez-Roux AV, Lutsey PL.

MA

Psychosocial Factors and Risk of Incident Heart Failure: The Multi-Ethnic Study of Atherosclerosis.

PMCID: PMC4692172. 31.

ED

Circulation Heart failure. 2016 Jan;9(1):e002243. PubMed PMID: 26699386. Pubmed Central

Ludvigsson JF, Andersson E, Ekbom A, et al. External review and validation of the Swedish national

PT

inpatient register. BMC public health. 2011;11:450. PubMed PMID: 21658213. Pubmed Central

Vandenbroucke JP, von Elm E, Altman DG, et al. Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaboration. International journal of surgery

AC

32.

CE

PMCID: PMC3142234.

(London, England). 2014 Dec;12(12):1500-24. PubMed PMID: 25046751.

ACCEPTED MANUSCRIPT Figure legends Figure 1. Flow chart of included and excluded conscripts, based on recommendations in Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) (32). The figure shows median years of

PT

observation (follow-up time), total person years of observation, and numbers of nonpsychotic mental

SC RI

disorders and hospitalized heart failure cases.

Figure 2. Cumulative incidence (%) of early heart failure in men with and without nonpsychotic mental

AC

CE

PT

ED

MA

NU

disorders at the age of 18 years.

20

ACCEPTED MANUSCRIPT Conscripts 1968-2005 (n=1 886 542)

Excluded (n=102 092) 1. Late allocated personal identity number (n=1 154) 2. Women (n=10 228) 3. Age >25 (n=54 651) 4. Heart failure before/at conscription (n=176) 5. Psychotic disorder before/at conscription (n=1 566) 6. Psychotic disorder after conscription (n=34 317)

First heart failure diagnosis (n=1 173)

First heart failure diagnosis (n=8 789)

ED AC

CE

PT

Fig 1.

Follow-up time until heart failure diagnosis: 0-46 years (median 26, IQR 17-34) (45 823 381 person years)

NU

No mental disorder (n=1 709 928)

MA

Nonpsychotic mental disorder (n=74 522)

SC RI

PT

Conscripts included (n=1 784 450)

21

AC

Fig 2.

CE

PT

ED

MA

NU

SC RI

PT

ACCEPTED MANUSCRIPT

22

ACCEPTED MANUSCRIPT

All

npm

Stress resilience (n=1 602 053) Medium (4-6) n=934 969

High (7-9) n=346 462

18.8 (1.2)

18.3 (0.7)

18.5 (0.9)

18.3 (0.6)

18.3 (0.6)

1.79 (0.07)

1.78 (0.07)

1.79 (0.07)

1.78 (0.07)

1.79 (0.07)

1.80 (0.06)

70.5 (11.0)

68.0 (11.2)

70.6 (11.0)

69.1 (12.2)

70.1 (10.6)

71.9 (9.1)

21.9 (3.0)

21.4 (3.1)

21.9 (3.0)

21.8 (3.5)

21.8 (2.9)

22.1 (2.4)

128.5 (10.9)

127.1 (11.0)

128.6 (10.9)

127.8 (11.0)

128.5 (10.9)

129.2 (10.8)

67.6 (9.8)

69.9 (9.4)

67.5 (9.8)

68.5 (9.8)

67.4 (9.8)

67.4 (9.7)

9962

1173

8789

2731

4791

1486

21.7

50.4

20.2

33.3

19.0

15.9

(21.3-22.2)

(47.6-53.3)

(19.8-20.6)

(32.0-34.5)

(18.5-19.6)

(15.1-16.8)

49.0 (8.9)

50.9 (7.8)

48.8 (9.0)

49.2 (8.5)

49.7 (8.2)

50.6 (8.0)

26.6 (440 940)

34.8 (106 223)

29.0 (263 259)

20.5 (68 738)

Weight, mean (SD) BMI, mean (SD)

Diastolic BP, mean (SD) Heart failure in any diagnostic position, n Cases per 100 000 observation years (95% CI)

ED

Age at HF diagnosis, years (SD) Parental education (highest achieved)

SC RI

Height, mean (SD)

MA

18.3 (0.7)

NU

Age, years (SD)

n=74 522

1-2, % (n)

27.3 (471 249)

3-4, % (n)

44.0 (757 836)

39.3 (26 762)

44.2 (731 074)

44.5 (135 816)

44.6 (403 961)

39.4 (132 366)

5-7, % (n)

28.7 (495 119)

16.3 (11 101)

29.2 (484 018)

20.8 (63 405)

26.4 (239 390)

40.2 (134 998)

21.2 (358 320)

39.2 (27 101)

20.4 (331 219)

41.8 (129 047)

17.4 (162 255)

5.7 (19 711)

CE

PT

44.5 (30 309)

PT

n=1 709 928

Low (1-3) n=320 622

n=1 784 450

Systolic BP, mean (SD)

No npm

23

IQ

AC

Low (1-3), % (n) Medium (4-6), % (n)

54.2 (917 389)

47.0 (32 459)

54.5 (884 930)

46.3 (143 094)

59.1 (552 293)

51.5 (178 388)

High (7-9), % (n)

24.7 (417 523)

13.8 (9 534)

25.1 (407 989)

12.0 (36 957)

23.5 (219 602)

42.8 (148 148)

Low (1-4), % (n)

13.5 (163 085)

30.6 (15 060)

12.7 (148 023)

28.9 (53 370)

13.2 (94 504)

3.8 (10 864)

Medium (5-7), % (n)

58.7 (711 520)

57.5 (28 278)

58.8 (683 242)

61.3 (113 421)

63.1 (452 999)

45.8 (132 111)

High (8-9), % (n)

27.8 (337 285)

11.9 (5 854)

28.5 (331 431)

9.9 (18 228)

23.7 (170 414)

50.4 (145 385)

Cardiovascular fitness

Table 1. Baseline characteristics in a national cohort of conscripts at the age of 18 years and incidence of heart failure (HF) during 46-year follow-up.

BMI indicates body mass index; BP, blood pressure; IQ, intelligence quotient; npm, nonpsychotic mental disorder; and SD, standard deviation.

ACCEPTED MANUSCRIPT

24

Table 2. Nonpsychotic mental disorders and stress resilience in late adolescence and risk of early heart failure (HF).

PAR Estimate (%)

All cases of HF,

Model 1

Model 2

Model 3

Model 4

Model 5

Model 6

Model 6

n=9 962

n=9 955

n=9 955

n=9 104

n=7 238

n=7 215

n=6 663

n=6 663

1.00

1.00

1.00

1.00

PT

Hazard Ratio (95% CI)

1.00

1.00

1.60 (1.50-1.71)

1.63 (1.53-1.74)

1.60 (1.50-1.71)

1.35 (1.25-1.46)

1.36 (1.25-1.47)

1.00

1.00

1.00

1.00

1.00

1.00

1.44 (1.33-1.55)

1.46 (1.36-1.58)

1.37 (1.25-1.50)

1.27 (1.16-1.39)

1.26 (1.15-1.39)

1.00

1.00

1.00

1.00

1.00

1.00

1.58 (1.38-1.81)

1.49 (1.27-1.75)

1.34 (1.15-1.58)

1.41 (1.19-1.66)

n=1 709 928

Nonpsychotic mental disorders, n=74 522

Alcohol/substance use, n=16 555

1.62 (1.42-1.84)

MA

2.70 (1.79-3.66)

1.47 (0.74-2.26)

0.63 (0.25-1.08)

CE

No alcohol/substance use, n=1 767 895

1.59 (1.40-1.81)

AC

n=10 566

PT

No personality disorders, n=1 773 884

Personality disorders,

1.47 (1.37-1.59)

ED

Depressive/neurotic disorders, n=57 706

1.48 (1.37-1.60)

NU

No depressive/neurotic disorders, n=1 726 744

SC RI

No mental disorder,

1.00

1.00

1.00

1.00

1.00

1.00

2.29 (2.01-2.61)

2.32 (2.04-2.64)

2.22 (1.92-2.56)

2.08 (1.76-2.47)

1.90 (1.60-2.25)

1.90 (1.59-2.28)

Model 1

Model 2

Model 3*

Model 4

Model 5

Model 6

Model 6

n=9 003

n=9 003

n=8 903

n=7 137

n=7 128

n= 6 588

n= 6 588

1.00

1.00

1.00

1.00

1.00

1.00

1.27 (1.20-1.35)

1.26 (1.19-1.34)

1.26 (1.19-1.34)

1.15 (1.08-1.23)

1.06 (0.99-1.13)

1.05 (0.98-1.12)

0.88 (0.53-1.29)

Stress resilience score 7-9 (high), n=346 462 (19.4%)

Stress resilience score 4-6 (medium), n=934 969 (52.4%)

Stress resilience score 1-3

2.52 (-1.66-6.58)

ACCEPTED MANUSCRIPT (low),

2.09 (1.96-2.23)

2.08 (1.95-2.22)

1.90 (1.78-2.04)

1.71 (1.59-1.84)

1.43 (1.33-1.55)

25 1.41 (1.30-1.53)

7.89 (5.75-10.03)

n=320 622 (18.0%)

SC RI

PT

Hazard ratios for heart failure in relation to nonpsychotic mental disorders and stress resilience in a national cohort of 18year-old male conscripts (n= 1 784 450) with adjusted and fully adjusted models. BMI indicates body mass index; CI, confidence interval; and PAR, population-attributable risk.

Model 1: Adjusted for age at conscription, conscription year, and test center.

Model 2: Additionally adjusted for comorbidities at baseline (hypertension, diabetes, congenital heart disease).

NU

Model 3: Additionally adjusted for BMI, and systolic and diastolic blood pressure.

Model 3*: Additionally adjusted for BMI, systolic and diastolic blood pressure, and concomitant nonpsychotic mental disorders.

MA

Model 4: Additionally adjusted for fitness. Model 5: Additionally adjusted for IQ.

AC

CE

PT

ED

Model 6: Additionally adjusted for parental education.