International journal of epidemiology最新文献

Background: Examining trajectories of undernutrition and overnutrition separately limits understanding of the double burden of malnutrition. We investigated transitions between normal, stunting, overweight and concurrent stunting and overweight (CSO) and associations with sociodemographic factors in children and adolescents.

Methods: We used data from the Young Lives cohort in India, Peru and Vietnam, which follow children 1-15 (N = 5413) and 8-22 years (N = 2225) over five rounds between 2002 and 2016. We estimated transitions between nutritional states using a Markov chain model and estimated sociodemographic associations employing a logit parametrization.

Results: Transitions into stunting peaked in ages 1-5 years (India: 22.9%, Peru: 17.6%, Vietnam: 14.8%), while stunting reversal was highest during adolescence across all countries. Transitions into overweight peaked in ages 19-22, while overweight reversal increased in ages 1-5 and 12-15 years. Transitions away from stunting to overweight were rare; more commonly, stunted individuals developed overweight while remaining stunted, leading to a CSO state. In Peru, 20.2% of 19-year-olds who were stunted reached CSO by age 22, with 4% shifting from stunted to overweight. Reversion to a normal state is least likely for those in a CSO state. Household wealth gradually reduced the likelihood of transitioning into stunting [odds ratios (ORs) for wealthiest quartile in Peru: 0.29, 95% confidence interval (CI) 0.20-0.41; India: 0.43, 95% CI 0.32-0.57; Vietnam: 0.36, 95% CI 0.26-0.50), with stunting reversal only being more likely in the two wealthiest quartiles across all countries (ORs for wealthiest quartile in Peru: 2.39, 95% CI 1.57-3.65; India: 1.28, 95% CI 1.05-1.54; Vietnam: 1.89, 95% CI 1.23-2.91). In Vietnam, only the richest quartile was at higher risk of transitioning into overweight (OR 1.87, 95% CI 1.28-2.72), while in Peru and India, the risk gradually rose across all wealth quartiles (ORs for wealthiest quartile in Peru: 2.84, 95% CI 2.14-3.77; India: 2.99, 95% CI 1.61-5.54).

Conclusions: Childhood and adolescence represent critical periods for prevention and reversal of stunting and overweight, thereby averting the development of CSO later in life. Context-specific interventions are crucial for preventing disparate transitions towards the double burden of malnutrition across socioeconomic groups.

Background: Youth psychiatric hospitalizations have been associated with negative outcomes, including premature death and post-discharge self-harm. Identifying risk factors for youth psychiatric hospitalization is crucial for informing prevention strategies. We aimed to evaluate the risk factors for psychiatric hospitalizations among low-income youth in Brazil.

Methods: This cohort study used interpersonal violence and psychiatric hospitalization data linked to the 100 Million Brazilian Cohort baseline. We considered 9 985 917 youths aged 5-24 years who enrolled at the baseline, between 2011 and 2018. We estimated the incidence rate (IR) with 95% confidence interval (CI) for psychiatric hospitalization by calculating the number of hospitalizations per person-year in 100 000 individuals at risk. The multilevel, multivariate Cox proportional hazards regression estimated the hazard risks (HR) with 95% CI for psychiatric hospitalization.

Results: The IR of psychiatric hospitalization was 12.28 per 100 000 person-years (95% CI, 11.96-12.6). Interpersonal violence victimization was the main risk factor for youth psychiatric hospitalization (HR, 5.24; 95% CI, 4.61-5.96). Other risk factors for psychiatric hospitalization included living with the oldest family member who had low education (HR, 2.51; 95% CI, 2.16-2.91) or was unemployed (HR, 1.49; 95% CI, 1.36-1.62), living with seven or more family members (HR, 1.84; 95% CI, 1.49-2.26) and being male (HR, 1.28; 95% CI, 1.21-1.36).

Conclusions: Urgent action is needed to prevent youth from suffering violence. Addressing this may alleviate the mental health burden in developmental ages, benefiting youth, families and the government through reduced costs in preventable psychiatric hospitalizations.

We outline a geometric perspective on causal inference in cohort studies that can help epidemiologists understand the role of standardization in controlling for confounding. For simplicity, we focus on a binary exposure X, a binary outcome D, and a binary confounder C that is not causally affected by X. Rothman diagrams plot the risk of disease in the unexposed on the x-axis and the risk in the exposed on the y-axis. The crude risks define a point in the unit square, and the stratum-specific risks at each level of C define two other points in the unit square. Standardization produces points along the line segment connecting the stratum-specific points. When there is confounding by C, the crude point is off this line segment. The set of all possible crude points is a rectangle with corners at the stratum-specific points and sides parallel to the axes. When there are more than two strata, standardization produces points in the convex hull of the stratum-specific points, and there is confounding if the crude point is outside this convex hull. We illustrate these ideas using data from a study in Newcastle, United Kingdom, in which the causal effect of smoking on 20-year mortality was confounded by age.