Determinants in Adolescence of Stroke-Related Hospital Stay Duration in Men
A National Cohort Study
Background and Purpose—Physical and psychological characteristics in adolescence are associated with subsequent stroke risk. Our aim is to investigate their relevance to length of hospital stay and risk of second stroke.
Methods—Swedish men born between 1952 and 1956 (n=237 879) were followed from 1987 to 2010 using information from population-based national registers. Stress resilience, body mass index, cognitive function, physical fitness, and blood pressure were measured at compulsory military conscription examinations in late adolescence. Joint Cox proportional hazards models estimated the associations of these characteristics with long compared with short duration of stroke-related hospital stay and with second stroke compared with first.
Results—Some 3000 men were diagnosed with nonfatal stroke between ages 31 and 58 years. Low stress resilience, underweight, and higher systolic blood pressure (per 1-mm Hg increase) during adolescence were associated with longer hospital stay (compared with shorter) in ischemic stroke, with adjusted relative hazard ratios (and 95% confidence intervals) of 1.46 (1.08–1.89), 1.41 (1.04–1.91), and 1.01 (1.00–1.02), respectively. Elevated systolic and diastolic blood pressures during adolescence were associated with longer hospital stay in men with intracerebral hemorrhage: 1.01 (1.00–1.03) and 1.02 (1.00–1.04), respectively. Among both stroke types, obesity in adolescence conferred an increased risk of second stroke: 2.06 (1.21–3.45).
Conclusions—Some characteristics relevant to length of stroke-related hospital stay and risk of second stroke are already present in adolescence. Early lifestyle influences are of importance not only to stroke risk by middle age but also to recurrence and use of healthcare resources among stroke survivors.
Characteristics in adolescence, including low stress resilience (limited ability to cope with stress), unhealthy body mass index, low cognitive function, poor physical fitness, and elevated blood pressure, are relevant to subsequent stroke risk in middle age,1,2 with a higher magnitude association for fatal than nonfatal stroke.1 It is uncertain whether these factors are relevant to length of hospital admission and risk of second stroke, which are indicators of patient suffering and societal costs among stroke survivors. We assessed this using a population-based cohort of men in Sweden.
The study population, described elsewhere,1 was drawn from a cohort comprising all male Swedish residents born between 1952 and 1956, who underwent compulsory assessments for military conscription and were recorded in the Swedish Military Conscription Register. The cohort originally consisted of 284 198 men, with complete data for 237 879. Psychologists systematically assessed potential conscripts for their predicted ability to cope with wartime stress. Using semistructured interviews, a stress resilience score (1–9) was derived. Other conscription assessment measures used in the analysis included body mass index, blood pressure, physical fitness, and cognitive function, as previously described.1 Using the Swedish National Patient Register and the Cause of Death Register, we identified nonfatal strokes and defined them as patients who survived at least 28 days. Fatal strokes were excluded as they would confound assessment of hospital stay duration. On the basis of the median duration of hospital stay, we defined long stay as 1 week or more and short stay as less than a week. Second stroke was defined as occurring at least 28 days after the first stroke, regardless of duration of hospital admission. Joint Cox models3 were used to assess the associations of characteristics in adolescence with long versus short duration of hospital admission after first stroke, as well as the risk of second stroke compared with first-ever stroke without recurrence. Results are presented as relative hazard ratios with 95% confidence intervals. The study was approved by the Regional Ethics Committee in Uppsala, Sweden (Dnr 2014/324). Further details are available in Methods and Tables I and II in the online-only Data Supplement.
Characteristics of men with a first nonfatal or second stroke and the stroke-free population are shown in Table 1. Results from joint Cox models for duration of hospital admission and second stroke are presented in Table 2. Low stress resilience, underweight, and higher systolic blood pressure are statistically significantly associated with a raised risk for long compared with short hospital stay among patients with ischemic stroke. Among those with intracerebral hemorrhage, raised systolic and diastolic blood pressures are statistically significantly associated with long compared with short stay. Only obesity is statistically significantly associated with a raised risk of second stroke, compared with first stroke without recurrence. We noted only a small reduction in the magnitude of associations after adjustment for a diagnosis of diabetes mellitus.
In this cohort of men who were ostensibly healthy during adolescence and subsequently followed up until middle age, we found that some previously identified early (adolescent) risk factors for stroke1,2 were associated with longer hospital admission and risk of second stroke. Predictors in adolescent men of future stroke-related hospital stay duration and recurrent stroke are largely unknown. Psychosocial stress has been associated with a risk of stroke onset and poststroke outcomes.1,4 Stress resilience may be an important determinant of the physiological and psychological consequences of exposures to stress.5 We have demonstrated that low resilience to psychosocial stress in adolescence can affect health across the life course as indicated by associations with raised risks of depression and anxiety in middle age.6 Markers of childhood stress (that may impair stress resilience) have been associated with higher blood pressure in old age.7 Thus, low stress resilience may influence stroke risk and outcome in adulthood in part through elevated blood pressure. We did not find an association with length of hospital stay or second stroke for all characteristics in adolescence linked with incident stroke risk.1,2 Possibly because factors associated with fatal stroke might be less notably associated with nonfatal stroke outcomes due to higher fatality among cohort members with these characteristics. We speculate that diabetes mellitus with onset after adolescence is intermediate in the causal pathway.
Potential limitations of the study include the risk of stroke misclassification, especially of intracerebral hemorrhagic strokes, when using International Classification of Diseases codes. However, the validity of stroke diagnoses recorded in the Swedish Patient Register is high.8 Because no women were included in the study (conscription information was limited to Swedish men), the results cannot be readily generalized to women or other ethnic groups. The length of stay is recorded accurately but might be affected by other factors, such as availability of beds and rehabilitation facilities, discharge possibilities, and local practice.9 Such nondifferential misclassification would only reduce the precision of our results, rendering conservative estimates. Stress resilience and other prestroke characteristics were measured only once, and their stability over time could be questioned, but associations with outcomes later in life suggest persistence.6 The proportion of variance explained for long compared with short hospital stay for ischemic stroke (8%) and hemorrhagic stroke (17%) and second compared with first stroke (7%) indicates potential residual confounding and thus the potential importance of unmeasured factors. In conclusion, some risks influencing length of hospital stay and second stroke in stroke survivors may have their origins in childhood and adolescence, including low stress resilience, unhealthy body mass index, and high blood pressure: they are thus relevant not only to stroke risk but also to recurrence and use of health service resources among stroke survivors.
Sources of Funding
This study received support from the UK Economic and Social Research Council as grants to the International Centre for Life Course Studies (grants RES-596-28-0001 and ES/JO19119/1) and was funded by grants from Stiftelsen Olle Engqvist Byggmästare, Folksam, and Strategic funding from Örebro University.
The online-only Data Supplement is available with this article at http://stroke.ahajournals.org/lookup/suppl/doi:10.1161/STROKEAHA.116.014265/-/DC1.
- Received June 11, 2016.
- Revision received June 30, 2016.
- Accepted July 1, 2016.
- © 2016 American Heart Association, Inc.
- Bergh C,
- Udumyan R,
- Fall K,
- Nilsagård Y,
- Appelros P,
- Montgomery S
- Xue X,
- Kim MY,
- Gaudet MM,
- Park Y,
- Heo M,
- Hollenbeck AR,
- et al
- Montgomery SM,
- Berney LR,
- Blane D