Epidemiology最新文献

Immortal time arises when individuals in the analysis are either selected based on post-assignment eligibility criteria or assigned to treatment strategies based on post-eligibility information. Explicit target trial emulation prevents the introduction of immortal time in survival analyses of observational data because it synchronizes eligibility and treatment assignment at the start of follow-up. Describing the structure of the biases that generate immortal time is facilitated by specifying the target trial so that the procedures to determine eligibility and assignment can be appropriately evaluated. Selection based on eligibility criteria applied after treatment assignment at the start of follow-up results in immortal time when the analysis starts the follow-up at assignment. Misclassification of assignment to treatment strategies based on treatment received after the start of follow-up results in immortal time when the treatment strategies are not distinguishable at the start of follow-up. The above selection and misclassification can be represented using causal diagrams. We summarize analytic approaches that prevent immortal time when longitudinal data are available from the time of treatment assignment. The term "immortal time bias" suggests that the source of the bias is the immortal time, but it is selection or misclassification that generates the immortal time, leading to bias.

The relevance of the epidemiologic concept of preventable fraction to the study of the population-level impact of preventive exposures is unequivocal. Here, we discuss how the preventable fraction can be usefully understood for the class of outcomes that relate to disease progression (e.g., clinical severity given diagnosis), and, under the principal stratification framework, derive an expression for this quantity for this type of outcome. In particular, we show that, in the context of disease progression, the preventable fraction is a function of the effect on the postdiagnosis outcome in the principal stratum in the unexposed group who would have disease regardless of exposure status. This work will facilitate an understanding of the contribution of principal effects to the impact of preventive exposures at the population level.

Most prior work in quantitative approaches to generalizability and transportability emphasizes extending causal effect estimates from randomized trials to target populations. Extending findings from observational studies is also of scientific interest, and identifiability assumptions and estimation methods differ from randomized settings when there is selection on both the exposure and exposure-outcome mediators in combination with exposure-outcome confounders (and both confounders and mediators can modify exposure-outcome effects). We argue that this causal structure is common in observational studies, particularly in the field of life course epidemiology, for example, when extending estimates of the effect of an early-life exposure on a later-life outcome from a cohort enrolled in midlife or late life. We describe identifiability assumptions and identification using observed data in such settings, highlighting differences from work extending findings from randomized trials. We describe statistical methods, including weighting, outcome modeling, and doubly robust approaches, to estimate potential outcome means and average treatment effects in the target population and illustrate performance of the methods in a simulation study. We show that in the presence of selection into the study sample on both exposure and confounders, estimators must be able to address confounding in the target population. When there is also selection on mediators of the exposure-outcome relationship, estimators need to be able to use different sets of variables to account for selection (including the mediator), and confounding. We discuss conceptual implications of our results as well as highlight unresolved practical questions for applied work to extend findings from observational studies to target populations.

Background: It is known that cesarean birth affects maternal outcomes in subsequent pregnancies, but specific effect estimates are lacking. We sought to quantify the effect of cesarean birth reduction among nulliparous, term, singleton, vertex (NTSV) births (i.e., preventable cesarean births) on severe maternal morbidity (SMM) in the second birth.

Methods: We examined birth certificates linked with maternal hospitalization data (2007-2019) from California for NTSV births with a second birth (N = 779,382). The exposure was cesarean delivery in the first birth and the outcome was SMM in the second birth. We used adjusted Poisson regression models to calculate risk ratios and population attributable fraction for SMM in the second birth and conducted a counterfactual impact analysis to estimate how lowering NTSV cesarean births could reduce SMM in the second birth.

Results: The adjusted risk ratio for SMM in the second birth given a prior cesarean birth was 1.7 (95% confidence interval: 1.5, 1.9); 15.5% (95% confidence interval: 15.3%, 15.7%) of this SMM may be attributable to prior cesarean birth. In a counterfactual analysis where 12% of the California population was least likely to get a cesarean birth instead delivered vaginally, we observed 174 fewer SMM events in a population of individuals with a low-risk first birth and subsequent birth.

Conclusion: In our counterfactual analysis, lowering primary cesarean birth among an NTSV population was associated with fewer downstream SMM events in subsequent births and overall. Additionally, our findings reflect the importance of considering the cumulative accrual of risks across the reproductive life course.

The Causal Roadmap outlines a systematic approach to asking and answering questions of cause and effect: define the quantity of interest, evaluate needed assumptions, conduct statistical estimation, and carefully interpret results. To protect research integrity, it is essential that the algorithm for statistical estimation and inference be prespecified prior to conducting any effectiveness analyses. However, it is often unclear which algorithm will perform optimally for the real-data application. Instead, there is a temptation to simply implement one's favorite algorithm, recycling prior code or relying on the default settings of a computing package. Here, we call for the use of simulations that realistically reflect the application, including key characteristics such as strong confounding and dependent or missing outcomes, to objectively compare candidate estimators and facilitate full specification of the statistical analysis plan. Such simulations are informed by the Causal Roadmap and conducted after data collection but prior to effect estimation. We illustrate with two worked examples. First, in an observational longitudinal study, we use outcome-blind simulations to inform nuisance parameter estimation and variance estimation for longitudinal targeted minimum loss-based estimation. Second, in a cluster randomized trial with missing outcomes, we use treatment-blind simulations to examine type-I error control in two-stage targeted minimum loss-based estimation. In both examples, realistic simulations empower us to prespecify an estimation approach with strong expected finite sample performance, and also produce quality-controlled computing code for the actual analysis. Together, this process helps to improve the rigor and reproducibility of our research.

Background: Studies suggest biologic mechanisms for gestational exposure to radiation and impaired fetal development. We explored associations between gestational radon exposure and term low birthweight, for which evidence is limited.

Methods: We examined data for 68,159 singleton full-term births in Connecticut, United States, 2016-2018. Using a radon spatiotemporal model, we estimated ZIP code-level basement and ground-level exposures during pregnancy and trimesters for each participant's address at birth or delivery. We used logistic regression models, including confounders, to estimate odds ratios (ORs) for term low birth weight in four exposure quartiles (Q1-Q4) with the lowest exposure group (Q1) as the reference.

Results: Exposure levels to basement radon throughout pregnancy (0.27-3.02 pCi/L) were below the guideline level set by the US Environmental Protection Agency (4 pCi/L). The ORs for term low birth weight in the second-highest (Q3; 1.01-1.33 pCi/L) exposure group compared with the reference (<0.79 pCi/L) group for basement radon during the first trimester was 1.22 (95% confidence interval [CI] = 1.02, 1.45). The OR in the highest (Q4; 1.34-4.43 pCi/L) quartile group compared with the reference group during the first trimester was 1.26 (95% CI = 1.05, 1.50). Risks from basement radon were higher for participants with lower income, lower maternal education levels, or living in urban regions.

Conclusion: This study found increased term low birth weight risks for increases in basement radon. Results have implications for infants' health for exposure to radon at levels below the current national guideline for indoor radon concentrations and building remediations.

Machine learning techniques for causal effect estimation can enhance the reliability of epidemiologic analyses, reducing their dependence on correct model specifications. However, the stochastic nature of many machine learning algorithms implies that the results derived from such approaches may be influenced by the random seed that is set before model fitting. In this work, we highlight the substantial influence of random seeds on a popular approach for machine learning-based causal effect estimation, namely doubly robust estimators. We illustrate that varying seeds can yield divergent scientific interpretations of doubly robust estimates produced from the same dataset. We propose techniques for stabilizing results across random seeds and, through an extensive simulation study, demonstrate that these techniques effectively neutralize seed-related variability without compromising the statistical efficiency of the estimators. Based on these findings, we offer practical guidelines to minimize the influence of random seeds in real-world applications, and we encourage researchers to explore the variability due to random seeds when implementing any method that involves random steps.

Background: The use of machine learning to estimate exposure effects introduces a dependence between the results of an empirical study and the value of the seed used to fix the pseudo-random number generator.

Methods: We used data from 10,038 pregnant women and a 10% subsample (N = 1004) to examine the extent to which the risk difference for the relation between fruit and vegetable consumption and preeclampsia risk changes under different seed values. We fit an augmented inverse probability weighted estimator with two Super Learner algorithms: a simple algorithm including random forests and single-layer neural networks and a more complex algorithm with a mix of tree-based, regression-based, penalized, and simple algorithms. We evaluated the distributions of risk differences, standard errors, and P values that result from 5000 different seed value selections.

Results: Our findings suggest important variability in the risk difference estimates, as well as an important effect of the stacking algorithm used. The interquartile range width of the risk differences in the full sample with the simple algorithm was 13 per 1000. However, all other interquartile ranges were roughly an order of magnitude lower. The medians of the distributions of risk differences differed according to the sample size and the algorithm used.

Conclusions: Our findings add another dimension of concern regarding the potential for "p-hacking," and further warrant the need to move away from simplistic evidentiary thresholds in empirical research. When empirical results depend on pseudo-random number generator seed values, caution is warranted in interpreting these results.