Epidemiology最新文献

Background: The COVID-19 pandemic, and subsequent policy responses aimed at curbing disease spread and reducing economic fallout, had far-reaching consequences for maternal health. There has been little research to our knowledge on enduring disruptions to maternal health trends beyond the early pandemic, and limited understanding of how these impacted pre-existing disparities in maternal health.

Methods: We leveraged rigorous interrupted time-series methods and US National Center for Health Statistics Vital Statistics Birth Data Files of all live births for 2015-2021 (N = 24,653,848) and estimated whether changes in maternal health trends after the onset of the COVID-19 pandemic (March 2020) differed from predictions based on pre-existing temporal trends. Outcomes included gestational diabetes, hypertensive disorders of pregnancy, gestational weight gain, and adequacy of prenatal care.

Results: We found increased incidence of gestational diabetes (December 2020 peak:1.7 percentage points (pp); 95%CI: 1.3, 2.1), hypertensive disorders of pregnancy (January 2021 peak: 1.3 pp; 95%CI: 0.4, 2.1), and gestational weight gain (March 2021 peak: 0.1 standard deviation (SD); 95%CI: 0.03, 0.1), and declines in inadequate prenatal care (January 2021 nadir: -0.4pp; 95%CI: -0.7, -0.1). Key differences by subgroups included greater and more sustained increases in gestational diabetes among Black, Hispanic, and less educated individuals.

Conclusion: These patterns in maternal health likely reflect not only effects of COVID-19 infection, but also changes in healthcare access, health behaviors, remote work, economic security, and maternal stress. Further research about causal pathways and longer-term trends will inform public health and clinical interventions to address maternal disease burden and disparities.

When conducting database studies, researchers sometimes use an algorithm known as "case definition," "outcome definition," or "computable phenotype" to identify the outcome of interest. Generally, algorithms are created by combining multiple variables and codes, and we need to select the most appropriate one to apply to the database study. Validation studies compare algorithms with the gold standard and calculate indicators such as sensitivity and specificity to assess their validities. As the indicators are calculated for each algorithm, selecting an algorithm is equivalent to choosing a pair of sensitivity and specificity. Therefore, receiver operating characteristic curves can be utilized, and two intuitive criteria are commonly used. However, neither was conceived to reduce the biases of effect measures (e.g., risk difference and risk ratio), which are important in database studies. In this study, we evaluated two existing criteria from perspectives of the biases and found that one of them, called the Youden index always minimizes the bias of the risk difference regardless of the true incidence proportions under nondifferential outcome misclassifications. However, both criteria may lead to inaccurate estimates of absolute risks, and such property is undesirable in decision-making. Therefore, we propose a new criterion based on minimizing the sum of the squared biases of absolute risks to estimate them more accurately. Subsequently, we apply all criteria to the data from the actual validation study on postsurgical infections and present the results of a sensitivity analysis to examine the robustness of the assumption our proposed criterion requires.

Difference-in-differences (DiD) is a powerful, quasi-experimental research design widely used in longitudinal policy evaluations with health outcomes. However, DiD designs face several challenges to ensuring reliable causal inference, such as when policy settings are more complex. Recent economics literature has revealed that DiD estimators may exhibit bias when heterogeneous treatment effects, a common consequence of staggered policy implementation, are present. To deepen our understanding of these advancements in epidemiology, in this methodologic primer, we start by presenting an overview of DiD methods. We then summarize fundamental problems associated with DiD designs with heterogeneous treatment effects and provide guidance on recently proposed heterogeneity-robust DiD estimators, which are increasingly being implemented by epidemiologists. We also extend the discussion to violations of the parallel trends assumption, which has received less attention. Last, we present results from a simulation study that compares the performance of several DiD estimators under different scenarios to enhance understanding and application of these methods.

Background: Sibling studies of maternal smoking during pregnancy and subsequent risk of depression have produced mixed results. A recent study identified not considering the amount of maternal smoking and age of onset as potentially masking a true association. We examine these issues and also the amount of maternal smoking during pregnancy as a determinant of the severity of depressive symptoms.

Methods: We analyzed data from the community-based National Longitudinal Survey of Youth (US, 1994-2016). Mothers reported smoking during pregnancy (none, <1 pack/day, ≥1 pack/day). We assessed offspring's lifetime depression (i.e., ≥8 symptoms) and symptom counts with the Centers for Epidemiologic Studies Depression scale. We estimated the risk of these two outcomes in the full sample (n = 7172) and among siblings (n = 6145) using generalized linear mixed-effects models with random intercepts by family and family-averaged means for sibling analyses.

Results: Among siblings, we observed dose-dependent elevations for both risk of depression (smoking during pregnancy <1 pack/day adjusted risk ratio [aRR] = 1.18; 95% confidence interval [CI] = 1.07, 1.30; smoking ≥1 aRR = 1.36; 95% CI = 1.19, 1.56) and severity of depressive symptoms (smoking <1 pack/day aRR = 1.12; 95% CI = 1.08, 1.16); smoking ≥1 pack/day aRR = 1.25; 95% CI = 1.18, 1.31). Among both samples, the P for trend was <0.01. In analysis limited to offspring diagnosed before age 18, results for severity were attenuated.

Conclusions: This evidence supports the existence of an independent association between maternal smoking during pregnancy and both the risk of depression and the severity of depressive symptoms. The results highlight the utility of considering the amount of smoking, severity of symptoms, and age of onset.

Background: Testing etiologic heterogeneity, whether a disorder subtype is more or less impacted by a risk factor, is important for understanding causal pathways and optimizing statistical power. The study of mental health disorders especially benefits from strategic subcategorization because these disorders are heterogeneous and frequently co-occur. Existing methods to quantify etiologic heterogeneity are not appropriate for noncompeting events in an open cohort of variable-length follow-up. Thus, we developed a new method.

Methods: We estimated risks from urban residence, maternal smoking during pregnancy, and parental psychiatric history, with subtypes defined by the presence or absence of a codiagnosis: autism alone, attention deficit hyperactivity disorder (ADHD) alone, and joint diagnoses of autism + ADHD. To calculate the risk of a single diagnosis (e.g., autism alone), we subtracted the risk for autism + ADHD from the risk for autism overall. We tested the equivalency of average risk ratios over time, using a Wald-type test and bootstrapped standard errors.

Results: Urban residence was most strongly linked with autism + ADHD and least with ADHD only; maternal smoking was associated with ADHD only but not autism only; and parental psychiatric history exhibited similar associations with all subgroups.

Conclusion: Our method allowed the calculation of appropriate P values to test the strength of association, informing etiologic heterogeneity wherein two of these three risk factors exhibited different impacts across diagnostic subtypes. The method used all available data, avoided neurodevelopmental outcome misclassification, exhibited robust statistical precision, and is applicable to similar heterogeneous complex conditions using common diagnostic data with variable follow-up.

Background: Children of immigrants often have excess mortality rates, in contrast to the low mortality typically exhibited by their parents' generation. However, prior research has studied children of immigrants who were selected for migration, thereby rendering it difficult to isolate the intergenerational impact of migration on adult mortality.

Methods: We use semiparametric survival analysis to carry out a total population cohort study estimating all-cause and cause-specific mortality among all adult men and women from age of 17 years among all men and women born in 1953-1972 and resident in Finland in 1970-2020. We compare children of forced migrants from ceded Karelia, an area of Finland that was ceded to Russia during the Second World War, with the children of parents born in present-day Finland.

Results: Children with two parents who were forced migrants have higher mortality than children with two parents born in Northern, Southern, and Western Finland, but similar or lower mortality than the subpopulation of children whose parents were born in the more comparable areas of Eastern Finland. For women and men, a mortality advantage is largest for external causes and persists after controlling for socioeconomic factors.

Conclusion: Our findings suggest that forced migration can have a beneficial impact on the mortality of later generations, at least in the case where forced migrants are able to move to contextually similar locations that offer opportunities for rapid integration and social mobility. The findings also highlight the importance of making appropriate comparisons when evaluating the impact of forced migration.

Background: Gestational diabetes is associated with adverse outcomes such as preterm birth (<37 weeks). However, there is no international consensus on screening criteria or diagnostic levels for gestational diabetes, and it is unknown whether body mass index (BMI) or obesity modifies the relation between glucose level and preterm birth.

Methods: We studied a pregnancy cohort restricted to two Danish regions from the linked Danish Medical Birth Register to study associations between glucose measurements from the 2-hour postload 75-g oral glucose tolerance test (one-step approach) and preterm birth from 2004 to 2018. In Denmark, gestational diabetes screening is a targeted strategy for mothers with identified risk factors. We used Poisson regression to estimate rate ratios (RR) of preterm birth with z-standardized glucose measurements. We assessed effect measure modification by stratifying analyses and testing for heterogeneity.

Results: Among 11,337 pregnancies (6.2% delivered preterm), we observed an adjusted preterm birth RR of 1.2 (95% confidence interval [CI] = 1.1, 1.3) for a one-standard deviation glucose increase of 1.4 mmol/l from the mean of 6.7 mmol/l. There was evidence for effect measure modification by obesity, for example, adjusted RR for nonobese (BMI, <30): 1.2 (95% CI = 1.1, 1.3) versus obese (BMI, ≥30): 1.3 (95% CI = 1.2-1.5), P = 0.05 for heterogeneity.

Conclusion: Among mothers screened for gestational diabetes, increased glucose levels, even those below the diagnostic level for gestational diabetes in Denmark, were associated with increased preterm birth risk. Obesity (BMI, ≥30) may be an effect measure modifier, not just a confounder, of the relation between blood glucose and preterm birth risk.

Background: Causal graphs are an important tool for covariate selection but there is limited applied research on how best to create them. Here, we used data from the Coronary Drug Project trial to assess a range of approaches to directed acyclic graph (DAG) creation. We focused on the effect of adherence on mortality in the placebo arm, since the true causal effect is believed with a high degree of certainty.

Methods: We created DAGs for the effect of placebo adherence on mortality using different approaches for identifying variables and links to include or exclude. For each DAG, we identified minimal adjustment sets of covariates for estimating our causal effect of interest and applied these to analyses of the Coronary Drug Project data.

Results: When we used only baseline covariate values to estimate the cumulative effect of placebo adherence on mortality, all adjustment sets performed similarly. The specific choice of covariates had minimal effect on these (biased) point estimates, but including nonconfounding prognostic factors resulted in smaller variance estimates. When we additionally adjusted for time-varying covariates of adherence using inverse probability weighting, covariates identified from the DAG created by focusing on prognostic factors performed best.

Conclusion: Theoretical advice on covariate selection suggests that including prognostic factors that are not exposure predictors can reduce variance without increasing bias. In contrast, for exposure predictors that are not prognostic factors, inclusion may result in less bias control. Our results empirically confirm this advice. We recommend that hand-creating DAGs begin with the identification of all potential outcome prognostic factors.

Background: Executive function, which develops rapidly in childhood, enables problem-solving, focused attention, and planning. Animal models describe executive function decrements associated with ambient air pollution exposure, but epidemiologic studies are limited.

Methods: We examined associations between early childhood air pollution exposure and school-aged executive function in 1235 children from three US pregnancy cohorts in the ECHO-PATHWAYS Consortium. We derived point-based residential exposures to ambient particulate matter ≤2.5 µm in aerodynamic diameter (PM 2.5 ), nitrogen dioxide (NO 2 ), and ozone (O 3 ) at ages 0-4 years from spatiotemporal models with a 2-week resolution. We assessed executive function across three domains, cognitive flexibility, working memory, and inhibitory control, using performance-based measures and calculated a composite score quantifying overall performance. We fitted linear regressions to assess air pollution and child executive function associations, adjusting for sociodemographic characteristics, maternal mental health, and health behaviors, and examined modification by child sex, maternal education, and neighborhood educational opportunity.

Results: In the overall sample, we found hypothesized inverse associations in crude but not adjusted models. Modified associations between NO 2 exposure and working memory by neighborhood education opportunity were present ( Pinteraction = 0.05), with inverse associations more pronounced in the "high" and "very high" categories. Associations of interest did not differ by child sex or maternal education.

Conclusion: This work contributes to the evolving science regarding early-life environmental exposures and child development. There remains a need for continued exploration in future research endeavors, to elucidate the complex interplay between natural environment and social determinants influencing child neurodevelopment.