There is increasing interest in the clustering of risk behaviours in adolescence. However, few studies have examined what clusters of risk behaviours exist among adolescents, their earlylife predictors, and their associations with later health. MethodsWe analysed data derived from 8754 participants (women 53.3%) in the 1970 British Cohort Study. Latent class analysis was used to identify clusters of risk behaviours at age 16.Regression modelling was then used to examine predictors of clusters and their consequences of risk behaviours and health outcomes at age 42. ResultsWe identified two latent classes: a risky-behaviour (men: 20.0%, women: 23.6%) and less-riskybehaviour class. Among men, those in the risky-behaviour class were more likely to report smoking, multiple binge drinking, sexual debut before 16, involvement in fights and delinquency than were women. Membership in risky-behaviour class was mainly predicted by sociodemographic and parental risk behaviours and monitoring. The risky-behaviour class at age 16 was associated with the following outcome age 42: smoking status (more strongly among women), excessive alcohol consumption (more strongly among men), worse self-rated health (more strongly among men), and psychological distress (only among women). ConclusionsEngagement in multiple risk behaviours in adolescence is an important driver of health inequalities later in life. Early life intervention, for example via school-based interventions, may be warranted for favourable lifelong health.

BackgroundAttrition, the loss of participants as a study progresses, is a considerable challenge in longitudinal studies. This study examined whether two forms of attrition, ‘withdrawal’ (formal discontinued participation) and ‘non-response’ (non-response among participants continuing in the study), have different associations with mortality and whether these associations differed across time in a multi-wave longitudinal study.MethodsParticipants were 10 012 civil servants who participated at the baseline of the Whitehall II cohort study with 11 data waves over an average follow-up of 28 years. We performed competing-risks analyses to estimate sub-distribution HRs and 95% CIs, and likelihood ratio tests to examine whether hazards differed between the two forms of attrition. We then applied linear regression to examine any trend of hazards against time.ResultsAttrition rate at data collections ranged between 13% and 34%. There were 495 deaths recorded from cardiovascular disease and 1367 deaths from other causes. Study participants lost due to attrition had 1.55 (95% CI 1.26 to 1.89) and 1.56 (1.39 to 1.76) times higher hazard of cardiovascular and non-cardiovascular mortality than responders, respectively. Hazards for withdrawal and non-response did not differ for either cardiovascular (p value =0.28) or non-cardiovascular mortality (p value =0.38). There was no linear trend in hazards over the 11 waves (cardiovascular mortality p value =0.11, non-cardiovascular mortality p value =0.61).ConclusionAttrition can be a problem in longitudinal studies resulting in selection bias. Researchers should examine the possibility of selection bias and consider applying statistical approaches that minimise this bias.

Objectives Adverse childhood experience is thought to be associated with risk of coronary heart disease, but it is not clear which experiences are cardiotoxic, and whether risk increases with the accumulation of adverse childhood experiences. Methods Participants were 5149 adults (72.6% men) in the Whitehall II cohort study. Parental death was recorded at phase 1 (median age in years 44.3), and 13 other adverse childhood experiences at phase 5 (55.3). We applied Cox proportional hazards regression with person-time from phase 5 to examine associations of adverse childhood experiences with incident coronary heart disease. We predicted hazard ratios according to count of the experiences, and examined dose-response effect. We finally estimated reduction of coronary heart disease in a hypothetical scenario, the absence of adverse childhood experiences. Results Among study participants, 62.9% had at least one adversity, with “financial problems” having the highest prevalence (26.1%). There were 509 first episodes of coronary heart disease during an average 12.9 years follow-up. Among 14 adverse childhood experiences in a multiply adjusted model, “parental unemployment” showed the highest hazard of coronary heart disease incidence (hazard ratio; 95% confidence interval: 1.53; 1.16 to 2.02). No dose-response effect was observed (constant for proportionality in hazard ratio: 1.05, 0.99 to 1.11). Based on the estimates of final model, in the absence of childhood adversities, we estimated a 6.0% reduction in coronary heart disease (0.94; 0.87 to 1.01), but the confidence interval includes one. Conclusion Although individual adverse childhood experiences show some association with coronary heart disease, there is no clear relationship with the number of adverse experiences. Further research is required to quantify effects of multiple and combinations of adverse childhood experiences considering timing, duration, and severity.

Background Dysregulation of diurnal cortisol patterns may mediate between chronic stress and health problems. Findings on the association of childhood adversities with adult cortisol, however, are equivocal, mainly due to low statistical power, sample selection, and unstandardised indices of cortisol. Methods Participants were from Whitehall II study (n = 3434), and National Child Development Study (NCDS) (n = 2072). In Whitehall II, multilevel models were used to examine associations of retrospectively measured childhood adversities with two indices of diurnal cortisol patterns (awakening response and diurnal slope) at a median age of 65. In NCDS, regression was used to investigate associations of prospectively measured adversities with two measures of morning cortisol at the age of 44/45. We then regressed predicted values of cortisol from these models against count of adversities to examine dose-response effects. Results The prevalence of reporting at least one adversity was 67.8% in Whitehall II, and 47.8% in NCDS. None of the individual adversities were associated with overall cortisol levels in either study. However, for each additional adversity, there was a 1% elevated awakening response (95%CI: 0.8% to 1.19%), and, among men only, a 1.2% lower cortisol level at awakening (-1.98 to -0.40) with flatter diurnal slope (0.1 to 0.1) in Whitehall II; and a 1.3% lower cortisol level (-1.78 to -0.70) at 3.75 hrs after awakening in NCDS. Conclusions Experience of multiple childhood adversities may be associated with dysregulated salivary cortisol patterns later in life. Key messages Accumulation of childhood adversities might be associated with dysregulated diurnal cortisol patterns in adulthood.

reactive protein [CRP]), physician-diagnosed chronic conditions (diabetes, hypertension, stroke and Coronary heart disease) and depressive symptoms ascertained with the Center for Epidemiologic Studies Depression Scale range 0 to 8. We controlled for age, sex and Apolipoprotein E (APOE). Results Our findings suggest that experiencing financial hardship in childhood (20% of the sample) was associated with a higher risk of cognitive impairment (OR=1.55, 95% CI 1.15 -2.09) in later life. A similar pattern was observed for having parents unemployed (7% of the sample) (OR=1.63, 95% CI 1.09 -2.46) or physically abusive parents (3%) (OR=3.21, 95% CI 1.82-5.66). We also found that increased depressive symptoms were interlinked with higher cognitive impairment, while APOEe4 and inflammatory markers were not directly associated. However, inflammation was indirectly associated with cognitive impairment, via depressive symptoms (b=0.08, SE=0.03, p=0.020) and chronic conditions (b=0.39, SE=0.19, p=0.042). Conclusion These findings support the psychosocial paradigm. They suggest that those from disadvantaged family backgrounds are more likely to have lower levels of education and be less wealthy, which in turn lead to poorer health, such as higher overall inflammation and increased depressive symptoms. These findings provide a plausible explanation for inequalities in late-life cognitive health.