Epidemiology [JOURNAL]
Sun
200 papers
Target trial emulation without matching: a more efficient approach for evaluating vaccine effectiveness using observational data.
Wu E, Rogawski McQuade E, Stensrud M … +2 more , Nabi R, Benkeser D
Epidemiology · 2026 Mar · PMID 41911278 · Publisher ↗
Real-world vaccine effectiveness has increasingly been studied using matching-based approaches, particularly in observational cohort studies that follow the target trial emulation framework. Although matching is appealin...
Real-world vaccine effectiveness has increasingly been studied using matching-based approaches, particularly in observational cohort studies that follow the target trial emulation framework. Although matching is appealing in its simplicity, it has important limitations in terms of clarity of the target estimand and the precision with which it is estimated. Moreover, defining causal estimands of vaccine effectiveness requires care, because vaccine uptake often occurs over calendar time when infection dynamics may also be rapidly changing. We propose a causal estimand of vaccine effectiveness that summarizes vaccine effectiveness over calendar time, similar to how vaccine efficacy is summarized in a randomized controlled trial. We describe the identification of our estimand and propose simple-to-implement estimators based on two hazard regression models. We apply our proposed estimator in simulations and in a study assessing the effectiveness of the Pfizer-BioNTech COVID-19 vaccine to prevent SARS-CoV-2 infections in children 5-11 years old. In both settings, we find that our proposed estimator yields similar scientific inferences while providing significant efficiency gains over commonly used matching-based estimators.
Income Volatility During Early to Mid-adulthood and 10-year Memory Decline in a Longitudinal Synthetic Cohort.
Kezios KL, Zimmerman SC, Buto PT … +2 more , Glymour M, Zeki Al Hazzouri A
Epidemiology · 2026 Jul · PMID 41885325 · Full text
BACKGROUND: Income volatility may negatively affect midlife cognition, but its impact on cognitive decline is unknown. Lifecourse data to investigate this relationship are lacking; a carefully constructed synthetic cohor...
BACKGROUND: Income volatility may negatively affect midlife cognition, but its impact on cognitive decline is unknown. Lifecourse data to investigate this relationship are lacking; a carefully constructed synthetic cohort can fill this gap. METHODS: We linked baby boomers from two US cohorts-the National Longitudinal Survey of Youth 1979 (NLSY79, N = 2,871) and the Health and Retirement Study (HRS, N = 5,711)-to create a longitudinal synthetic cohort. In NLSY79, we computed income volatility between 1990 and 2010. In HRS, memory was assessed from 2010 to 2020 (sum of immediate and delayed 10-word recall scores). To link participants across cohorts, each HRS participant was matched to and assigned the income volatility history of their 20 most similar NLSY79 counterparts based on harmonized linking variables measured in 2010 (approximately age 50 in each cohort). In the synthetic cohort, we used confounder-adjusted linear mixed models to estimate the effect of income volatility on memory function and 10-year decline. RESULTS: Higher income volatility was associated with lower baseline memory function but not decline. Compared with experiencing no income drops ≥25% between surveys, memory scores were lower by 0.11 points (95% CI: -0.40, 0.18) for participants experiencing one drop, 0.32 (95% CI: -0.63, -0.00) for 2 drops, and 0.60 (95% CI: -0.93, -0.27) for ≥3 drops. Results attenuated with adjustment for earlier-life cognition (available only in NLSY79). CONCLUSION: Income volatility from early to mid-adulthood may impact midlife memory function, but no association with memory decline was observed. Results rely on strong cohort-linking assumptions and may be influenced by unmeasured earlier-life confounders.
Re: Generalizing and Transporting Causal Inferences from Randomized Trials in the Presence of Trial Engagement Effects.
Ross RK, Rudolph KE, Malinsky D
Epidemiology · 2026 Jul · PMID 41874467 · Publisher ↗
Abstract loading — click title to view on PubMed.
Temporal Variation in the Association Between Short-term Exposure to Fine Particulate Matter and Mortality Across Subpopulations in North Carolina and Michigan, U.S.
Stewart RK, Kim H, Song Y … +4 more , Choi HM, Chen C, Choi Y, Bell ML
Epidemiology · 2026 Jul · PMID 41849246 · Full text
Growing evidence suggests that the health impacts of a given increment of fine particulate matter (PM 2.5 ) are changing over time. While these trends could influence environmental health disparities, how temporal trends...
Growing evidence suggests that the health impacts of a given increment of fine particulate matter (PM 2.5 ) are changing over time. While these trends could influence environmental health disparities, how temporal trends vary across locations and subpopulations remains unknown. Using case-crossover analyses, we analyzed how PM 2.5 -mortality associations varied over time (2001-2016) in two U.S. states (North Carolina and Michigan) among persons 65+ years of age. We further characterized temporal variation in the PM 2.5 -mortality association by sociodemographic variables including age, sex, race/ethnicity, educational attainment, urbanicity, and substate region. From the 2001-2008 study period to the 2009-2016 study period, the odds of mortality per 10 μg/m 3 PM 2.5 decreased by 0.77% in North Carolina but increased by 0.28% in Michigan. Using a nonlinear temporal model, the associations between PM 2.5 concentration and mortality were variable over time in North Carolina and steadily increased over time in Michigan. We also found suggestive evidence of widening disparities in PM 2.5 -mortality odds ratios over time. For example, PM 2.5 -mortality odds in Michigan increased more dramatically for non-Hispanic black (1.71% [-1.17, 4.67%]) than for non-Hispanic white (0.14% [95% confidence interval (CI) = -0.93, 1.22%]) subpopulations. In North Carolina, these groups experienced a 0.40% (-3.56, 2.85%) and 0.96% (-2.52, 0.62%) reduction in PM 2.5 -mortality, respectively. Our results suggest that PM 2.5 -related mortality impacts may be changing over time, with different trends by location and subpopulation, potentially exacerbating environmental injustice.
Informed Presence in Electronic Health Record Data: Illustrating Bias and Bias Reduction Approaches in Longitudinal Analyses.
Vader DT, Shu D, Hubbard RA … +4 more , Boge CLK, Sharova A, Downes K, Li Y
Epidemiology · 2026 Jul · PMID 41849238 · Publisher ↗
Electronic health record (EHR) systems capture patient information inconsistently, with patients generally contributing more data when they are sick than when they are healthy. This creates "informed presence," systemati...
Electronic health record (EHR) systems capture patient information inconsistently, with patients generally contributing more data when they are sick than when they are healthy. This creates "informed presence," systematic differences between captured and noncaptured data, potentially biasing association estimates. There is growing interest in methods that account for informed presence, but practical approaches for conceptualizing, identifying, and addressing this bias in applied EHR research have received limited attention. Focusing on longitudinal settings, we present a conceptual framework for informed presence bias, which arises when data capture depends on exposure and outcome, and thus the visit process acts as a collider. We then illustrate methods that aim to reduce bias by reweighting or resampling observed data to approximate conditional independence between the visit process and the outcome. We illustrate these methods using longitudinal EHR data from pediatric solid organ transplant recipients (N = 271) to examine the association between steroids and cytomegalovirus viremia, where the frequency of cytomegalovirus testing varies across patients and over time. Incidence rate ratios decreased from 1.83 (95% CI = 1.02, 3.28) in a naïve analysis to 1.37 (0.73, 2.57) when accounting for informed presence using inverse intensity weighting. Incidence rate ratio estimates from bootstrapped inverse intensity weighting were 1.37 (0.71, 2.27) and 1.40 (0.73, 2.68) from multiple outputation. These results show the anticipated attenuation of effect estimates after accounting for informed presence bias. When analyzing irregularly measured EHR data, we recommend (1) identifying the expected observation process using conceptual diagrams, (2) assessing dependence in the observation process, and (3) accounting for outcome dependence in statistical analysis.
Validation of Body Mass Index-For-Age Percentile Curves in Older Adults Using Data From the Canadian Longitudinal Study on Aging.
Kim CD, Cook CE, Banack HR
Epidemiology · 2026 Jul · PMID 41849231 · Publisher ↗
BACKGROUND: The use of a standard body mass index (BMI) cut-point (e.g., BMI ≥30 kg/m 2 ) to define obesity across the lifespan does not reflect our current understanding of changes in adiposity that occur throughout the...
BACKGROUND: The use of a standard body mass index (BMI) cut-point (e.g., BMI ≥30 kg/m 2 ) to define obesity across the lifespan does not reflect our current understanding of changes in adiposity that occur throughout the aging process. The objective of the present study was to examine the validity of BMI-for-age percentiles as a measure of obesity in older adults, compared with an objective gold-standard measure of obesity status, total body fat percent (BF%) measured by dual energy X-ray absorptiometry scan. METHODS: Using data from the Canadian Longitudinal Study on Aging (n = 28,764), sensitivity, specificity, positive predictive value, and negative predictive value were calculated for BMI ≥85th percentile and other thresholds against BF% ≥35%, 38%, and 40%, respectively. RESULTS: When comparing obesity defined by BMI percentile values (≥85th percentile) to BF% (≥40% BF) as a measure of true obesity status, sensitivity, specificity, positive predictive value, and negative predictive value values were 0.32 (0.31, 0.33), 0.98 (0.98, 0.98), 0.94 (0.93, 0.95), and 0.61 (0.60, 0.62) in women and 0.85 (0.81, 0.89), 0.86 (0.85, 0.86), 0.14 (0.13, 0.16), and 1.00 (0.99, 1.00) in men, respectively. CONCLUSION: Incorporating age and sex information, BMI-for-age percentile is a highly specific and moderately sensitive measure of obesity in older adults, with potential as a valuable screening tool. Further research on BMI-for-age percentiles could help develop more accurate obesity screening tools than the current standard BMI cut-points.
Spatial Heterogeneity in Synergistic Effects of Extreme Heat and NO 2 Exposures on Cardiorespiratory Hospitalizations in California.
Ma Y, Chen C, Aguilera R … +5 more , Gershunov A, Jerrett M, Connolly R, Marlier ME, Benmarhnia T
Epidemiology · 2026 Jul · PMID 41790999 · Full text
BACKGROUND: Extreme heat and high air pollution levels may co-occur and act synergistically. The synergistic health effects of extreme heat and nitrogen dioxide (NO 2 ) , and the fine-scale spatial heterogeneity in such...
BACKGROUND: Extreme heat and high air pollution levels may co-occur and act synergistically. The synergistic health effects of extreme heat and nitrogen dioxide (NO 2 ) , and the fine-scale spatial heterogeneity in such joint effects, remain unclear. We aimed to investigate the synergistic effect of extreme heat and NO 2 exposures on cardiorespiratory hospitalizations in California and explore its spatial heterogeneity. METHODS: Daily ZIP Code Tabulation Area (ZCTA)-level hospitalization data, 2000-2019, were obtained from the California Department of Health Care Access and Information. Extreme heat and NO 2 exposures were defined as days when the heat index and NO 2 concentration exceeded a ZCTA-specific threshold, respectively. We first investigated the state-level synergistic effects with a case-crossover analysis and then estimated the ZCTA-specific synergistic effects using a within-community matched design combined with a spatial Bayesian hierarchical model. Finally, we explored how community characteristics modify these effects using meta-regressions. RESULTS: We found mild synergistic effects of extreme heat and NO 2 exposures on cardiorespiratory hospitalizations at the state level, with a relative excess risk due to interaction (RERI) of 0.005 (95% confidence interval = -0.002, 0.011). Yet, great spatial heterogeneity was observed, with ZCTA-specific RERIs ranging from -1.08 to 2.22. Higher RERIs were observed in communities with lower socioeconomic status, higher population density, reduced green space, greater proportions of racial and ethnic minority residents, and higher historical temperatures. CONCLUSIONS: Our findings underscore the need for adaptation policies that integrate compound exposures to heat and air pollution and inform the development of targeted intervention strategies to protect vulnerable communities.
The Association Between Outdoor Temperature During the Past Weeks and Current Fluid Homeostasis.
Enhörning S, Melander O, Elmståhl S … +3 more , Engström G, Pihlsgård M, Timpka S
Epidemiology · 2026 Jul · PMID 41790995 · Full text
BACKGROUND: It remains unclear why cool temperatures cause more persistent adverse health effects compared with hot weather. Fluid homeostasis may constitute a causal link between past temperatures and adverse health eff...
BACKGROUND: It remains unclear why cool temperatures cause more persistent adverse health effects compared with hot weather. Fluid homeostasis may constitute a causal link between past temperatures and adverse health effects. In this study, we investigated the association between past outdoor temperatures and current fluid homeostasis. METHODS: We studied participants from five cohorts during three decades in Sweden (total n = 29,755, age 18-86 years, 50.4% women). We quantified fluid homeostasis through indicators of hormonal regulation (vasopressin biomarker plasma copeptin), urine concentration (urine osmolality), and replenishment of fluid loss (total water intake) and related these parameters to past outdoor temperatures (21 days) using distributed nonlinear lag models. RESULTS: Past temperatures were nonlinearly associated with current copeptin and urine osmolality. Cool temperatures during days and weeks prior contributed to distinct patterns of high copeptin with concomitant influence on urine osmolality. Overall, a scenario of temperatures of 0 °C for 21 days showed 14.9% (95% confidence interval = 11.5%, 18.3%) higher copeptin levels compared with reference temperatures of 14.3 °C for 21 days. Associations between copeptin and warm temperatures were less complex and of shorter duration, linking elevated temperatures within 24 hours with higher copeptin compared with the reference temperature. CONCLUSIONS: Alterations in human fluid homeostasis may partly explain the observed link between moderately cool outdoor temperatures and adverse health effects weeks later. If so, avoiding altered water balance through moderately increased water intake might mitigate the adverse health effects of cool weather.
Improving Inference in Air Pollution Epidemiology: The Case for Rethinking Multipollutant Adjustment.
Chen H, Quick M, Kaufman JS … +9 more , Chen C, Kwong JC, Rana J, Kim J, van Donkelaar A, Martin RV, Tjepkema M, Benmarhnia T, Burnett RT
Epidemiology · 2026 Jul · PMID 41790994 · Full text
Air quality regulations and programs are vital for protecting the public from harms caused by air pollution. To support these actions, numerous epidemiological studies have sought to identify the pollutants most responsi...
Air quality regulations and programs are vital for protecting the public from harms caused by air pollution. To support these actions, numerous epidemiological studies have sought to identify the pollutants most responsible for adverse outcomes. These studies often used statistical adjustments for copollutants in outcome regression models, a practice also commonly applied to assess interactions between copollutants. Here, we highlight possible pitfalls of multipollutant analyses. Indiscriminate copollutant adjustment can induce noncausal associations through collider adjustment, distorting effect estimates for individual air pollutants. We describe the underlying mechanisms and provide empirical evidence on how such bias may realistically influence the relationships between air pollution and health outcomes from a well-characterized Canadian national cohort alongside a simulation study. Additionally, we discuss strategies to mitigate the impact of this bias. Given the widespread interest in multipollutant approaches among the scientific and policy communities, greater caution is needed when conducting and interpreting research on multiple pollutants.
What Makes Something Worth Knowing? Epistemology and Public Health Impact.
Keyes KM
Epidemiology · 2026 May · PMID 41734343 · Publisher ↗
Abstract loading — click title to view on PubMed.
Considerations for the Analysis of Urinary Environmental Chemical Concentrations During Pregnancy.
Stevens DR, Hinton K, O'Brien KM … +10 more , Buckley JP, Welch BM, Watts JA, Calafat AM, Botelho JC, Sinkovskaya E, Przybylska A, Saade G, Abuhamad A, Ferguson KK
Epidemiology · 2026 May · PMID 41698219 · Full text
BACKGROUND: Pregnancy physiology may impact indicators of hydration, affecting exposure assessment in gestational studies with urinary chemical biomarkers. METHODS: We aimed to characterize hydration and demonstrate the...
BACKGROUND: Pregnancy physiology may impact indicators of hydration, affecting exposure assessment in gestational studies with urinary chemical biomarkers. METHODS: We aimed to characterize hydration and demonstrate the impact of different methods for standardizing urinary chemical biomarker concentrations on exposure descriptives in the Human Placenta and Phthalates Study (n = 303, 2017-2020), a prospective pregnancy cohort with eight study visits between 12 and 38 weeks of gestation. We assessed trajectories and predictors of hydration using urine flow rate (UFR) and specific gravity (SG). Likewise, we examined trajectories and predictors of mono-n-butyl phthalate (MBP), for which concentrations were unstandardized or standardized via UFR (analyte excretion rates), SG alone (Boeniger method), or covariate-adjusted SG (O'Brien method). We used generalized additive mixed effects models to examine trajectories. We used linear mixed effects models to investigate participant demographic and pregnancy characteristics influencing UFR and SG, and agreement between unstandardized and standardized MBP concentrations. RESULTS: As pregnancy progressed, SG declined linearly, whereas UFR varied in a nonlinear manner. Several demographic and pregnancy characteristics, notably race and ethnicity, were associated with UFR and SG. Unstandardized and standardized MBP concentrations showed good agreement, with lower agreement observed between concentrations standardized using UFR (analyte excretion rates) relative to SG (Boeniger and O'Brien methods). Nevertheless, trajectories and predictors of MBP concentrations were similar across all standardization approaches. CONCLUSIONS: Despite systematic variations in hydration across gestation and by demographic and pregnancy characteristics, methods commonly used for standardizing urinary concentrations of chemical biomarkers were robust to differences in the hydration indicator as well as the standardization method employed.
State Minimum Wages and Food Insecurity Among Households Receiving Government Food Assistance.
Neumann K, Laraia BA, Riddell CA
Epidemiology · 2026 May · PMID 41666385 · Publisher ↗
BACKGROUND: While the Supplemental Nutrition Assistance Program (SNAP) aims to reduce food insecurity among low-income households, nearly half of recipients remain food insecure. Increasing state minimum wages could help...
BACKGROUND: While the Supplemental Nutrition Assistance Program (SNAP) aims to reduce food insecurity among low-income households, nearly half of recipients remain food insecure. Increasing state minimum wages could help improve food security, but because SNAP benefits are income-dependent, net effects are unclear. METHODS: Using the US Current Population Survey Food Security Supplement (2002-2019), we linked households interviewed in two consecutive Decembers to create a 2-year panel. The primary sample included SNAP recipient households with at least one adult working in year 1. The exposure was the average effective state minimum wage (2019 $) for each state and year. We estimated prevalence differences in food insecurity per $1 increase in minimum wage using a within-household linear fixed-effects model adjusting for time-varying economic confounders and concurrent safety-net policies. We investigated variation in the effect by household and family structure, race and ethnicity, and educational attainment using stratified models. RESULTS: Overall estimates were most compatible with protective effects (prevalence difference per 10,000 households: -298, 95% confidence interval [CI]: -673, 77). The strongest protective estimates were for senior-headed (-1,472, 95% CI: -2,869, -76), Hispanic (-865, 95% CI: -1,638, -92), and some college households (-988, 95% CI: -1,664, -312). Estimates for Indigenous households were imprecise and possibly harmful (900, 95% CI: -736, 2,537). Most other subgroup estimates were near zero. CONCLUSIONS: Increased minimum wages may modestly support food security for working SNAP households. As SNAP benefit rules evolve, these findings suggest that minimum-wage policies can complement and reinforce the program's goals to protect low-income households from food hardship.
Development and Validation of Gestational Age Estimation Algorithms for Nonlive Births in Administrative Healthcare Databases.
Cho Y, Choi EY, Lee H … +5 more , Noh Y, Han JY, Choe SA, Kim H, Shin JY
Epidemiology · 2026 May · PMID 41666378 · Publisher ↗
BACKGROUND: Algorithms to estimate gestational age (GA) for nonlive births have been developed in other administrative healthcare databases, but their applicability in Korea remains unknown. We adapted algorithms develop...
BACKGROUND: Algorithms to estimate gestational age (GA) for nonlive births have been developed in other administrative healthcare databases, but their applicability in Korea remains unknown. We adapted algorithms developed in the United States and evaluated their validity in Korean healthcare claims data. METHODS: Using the National Health Information Database of South Korea, we linked GA information on influenza vaccination from the national vaccination registry to establish a reference standard. Nonlive births were stratified into spontaneous/induced abortions and stillbirths. Four algorithms were tested: (1) assigning outcome-specific GA, (2) adjusting GA based on ultrasound scan records, (3) a regression model using gestational markers as predictors, and (4) a random forest model. Algorithms were evaluated by the proportion of estimates within 1-4 weeks of the reference standard and the mean squared error (MSE). External validation was conducted using an independent dataset. RESULTS: Random forests performed best for both spontaneous/induced abortions (MSE 1.68 weeks 2 ) and stillbirths (MSE 0.97 weeks 2 ), with 92.6% (95% confidence interval 91.6, 93.4) and 97.4% (96.2, 98.3) of predictions falling within 2 weeks of the reference standard, respectively. However, in the external validation set, the ultrasound record-based adjustment approach performed similarly to the random forest approach for both spontaneous/induced abortions (MSE 8.37 vs. 8.15 weeks 2 ) and stillbirths (MSE 12.42 vs. 12.52 weeks 2 ). CONCLUSIONS: Deterministic approaches may be preferable for estimating GA of nonlive births in the National Health Information Database, as they are simpler to implement and perform comparably to model-based algorithms. These algorithms can support pregnancy research in the Korean population.
Mind the Gap: Addressing Missing Person Time When Estimating Outcome Incidence in Longitudinal Data.
Rudolph JE, Ross RK, Zalla LC … +5 more , Mehta SH, Kirk GD, Genberg BL, Lau B, Lesko CR
Epidemiology · 2026 May · PMID 41666377 · Full text
BACKGROUND: Longitudinal data often include gaps in observation when outcomes (and other variables) are unmeasured due to missed study visits or dropout. We explore the fundamentals of data gaps and use simulation to com...
BACKGROUND: Longitudinal data often include gaps in observation when outcomes (and other variables) are unmeasured due to missed study visits or dropout. We explore the fundamentals of data gaps and use simulation to compare approaches for handling data gaps when estimating outcome incidence. METHODS: We generated a simulation of 1000 individuals across 10 study visits. We used 4 data-generating mechanisms: (1) missingness was independent of the outcome; (2) there was a baseline common cause of missingness and the outcome; (3) there was a time-varying common cause; and (4) the outcome directly affected future missingness. We estimated the risk and rate of the first outcome occurrence (generated as a transient, repeated, and permanent outcome), using crude and adjusted approaches, across 1000 iterations, and compared bias and empirical standard error. RESULTS: Under Scenario 1, in crude analyses, results were unbiased when censoring before a data gap but not when allowing participants to return. Under scenarios 2-4, all crude approaches were biased. Inverse probability of censoring weights and multiple imputation were relatively unbiased across scenarios and outcome types; multiple imputation was more precise. Inverse probability of observation weights was biased when the outcome was permanent and was less precise than either of the other two approaches. CONCLUSION: Crude approaches allowing participants to return following a data gap are not recommended because they can be biased even when missingness and the outcome are independent. Instead, one should either censor or handle the data gap using multiple imputation.
Long-term Cardiovascular Outcomes Following Bariatric Surgery: Reconciling Seemingly Conflicting Evidence.
Haneuse S, Benz L, Smith VA … +2 more , Arterburn D, Maceijewski ML
Epidemiology · 2026 Jul · PMID 41570241 · Full text
Substantial observational evidence exists in support of bariatric surgery being associated with a reduction in risk for a wide range of outcomes, including cardiovascular disease (CVD) in patients with diabetes. Two rece...
Substantial observational evidence exists in support of bariatric surgery being associated with a reduction in risk for a wide range of outcomes, including cardiovascular disease (CVD) in patients with diabetes. Two recent studies, however, argued that much of that prior work suffers from various sources of underappreciated bias as well as design decisions that compromise whether one can conceive of a corresponding target trial. Furthermore, results based on analyses of claims data from Optum and electronic health record data from the Veterans Administration are presented as providing evidence of no CVD benefit for bariatric surgery in patients with diabetes. In this paper, we use data from a prior Kaiser Permanente study to emulate a trial that mimics the methods employed in the Veterans Administration study. This new analysis finds a reduction in risk of CVD in patients with diabetes, consistent with preexisting evidence. We discuss possible mechanisms by which the discrepant results can be reconciled, including issues of statistical validity that arise from small samples, whether recent work on transportability indicates that we should not always expect results to always be concordant, and the role of conservatism associated with "clinical trial thinking." We conclude with a discussion of what standards should be used when considering the work of others in the literature and the role that evidence triangulation may be play in the future.
A Call for Randomization: Bariatric Surgery and Cardiovascular Disease.
Madenci AL, Wanis KN, Trinquart L … +3 more , Kurgansky KE, Gerlovin H, Hernán MA
Epidemiology · 2026 Jul · PMID 41570240 · Full text
Abstract loading — click title to view on PubMed.
Defining and Estimating Outcomes Directly Averted by a Vaccination Program when Rollout Occurs Over Time.
Jia KM, Boyer CB, Bilinski A … +1 more , Lipsitch M
Epidemiology · 2026 Jul · PMID 41570185 · Full text
During the COVID-19 pandemic, estimating the total deaths averted by vaccination was of great public health interest. Instead of estimating total deaths averted by vaccination among both vaccinated and unvaccinated indiv...
During the COVID-19 pandemic, estimating the total deaths averted by vaccination was of great public health interest. Instead of estimating total deaths averted by vaccination among both vaccinated and unvaccinated individuals, some studies empirically estimated only "directly averted" deaths among vaccinated individuals, typically suggesting that vaccines prevented more deaths among unvaccinated and vaccinated individuals than directly among vaccinated individuals only, due to the indirect effect. Here, we define the causal estimand to quantify outcomes "directly averted" by vaccination-that is, the impact of vaccination for vaccinated individuals, holding vaccination coverage fixed-for vaccination at multiple time points, which is a lower bound on the total outcomes averted when the indirect effect is non-negative. We develop an unbiased estimator for the causal estimand in a one-stage randomized controlled trial and explore the bias of a popular "hazard difference" estimator frequently used in empirical studies. We show that even in a randomized controlled trial, the hazard difference estimator is biased if vaccination has a non-null effect, as it fails to incorporate the greater depletion of susceptibles among the unvaccinated individuals. In simulations, the overestimation is small for averted deaths when the infection-fatality rate is low, as for many important pathogens. However, the overestimation can be large for averted infections given a high basic reproduction number and a high vaccine efficacy against infection. Additionally, we define and compare estimand and estimators for avertible outcomes (i.e., outcomes that could have been averted by vaccination, but were not due to failure to vaccinate). Future studies can explore the identifiability of the causal estimand in observational settings.
Differential Reporting of Severe Maternal Morbidity on US Birth Certificate and Claims Data by Race and Ethnicity.
Pineles BL, Harris AD, Pineles L … +5 more , Davis EM, Joseph KS, Schisterman E, Magder LS, Goodman KE
Epidemiology · 2026 May · PMID 41570184 · Publisher ↗
BACKGROUND: US birth certificates are the only complete, national source of severe maternal morbidity (SMM) data, but state and local studies have identified data quality concerns. The objective of this study was to comp...
BACKGROUND: US birth certificates are the only complete, national source of severe maternal morbidity (SMM) data, but state and local studies have identified data quality concerns. The objective of this study was to compare reporting of SMM in birth certificate versus hospital claims data and evaluate differences by patient race/ethnicity. METHODS: We compared incidence rates of blood transfusion, hysterectomy, intensive care unit admission, uterine rupture, and third/fourth degree perineal laceration between 2019 deliveries in the United States birth certificate and the Premier Healthcare Database, overall and stratifying by maternal race/ethnicity. We then computed incidence rate ratios (IRRs) computed between datasets, and fit logistic regression models of race/ethnicity on SMM. RESULTS: Comparing 3,467,934 birth certificate deliveries with 3,450,569 Premier deliveries (n = 905,766 preweighting for national representativeness), incidence rates of SMMs were lower in birth certificate compared with Premier data, and these rate differentials varied by maternal race/ethnicity. For example, among non-Hispanic white patients, the incidence rate of blood transfusions in birth certificate data was 50% that of the incidence rate in the Premier claims dataset (IRR: 0.50, 95% confidence interval [CI] = 0.47, 0.52). Among all other races/ethnicities, the incidence rate of blood transfusions was even lower relative to the claims data (IRR range: 0.29-0.39). Adjusted odds ratios for SMM in non-Hispanic Black and Hispanic patients versus non-Hispanic white patients were closer to the null in birth certificate than Premier data (e.g., compared with non-Hispanic white patients, non-Hispanic Black patients had a 16% higher adjusted odds in the birth certificate [95% CI = 1.10, 1.21] data versus an 84% higher adjusted odds of blood transfusion in Premier data [95% CI = 1.79, 1.89]). CONCLUSIONS: Birth certificates report substantially less SMM than claims data, with a greater differential in reporting for non-Hispanic Black and Hispanic patients that may bias birth certificate-based research findings.
Understanding Algorithmic Fairness for Clinical Prediction in Terms of Subgroup Net Benefit and Health Equity.
Benitez-Aurioles J, Joules A, Brusini I … +2 more , Peek N, Sperrin M
Epidemiology · 2026 May · PMID 41512215 · Publisher ↗
There are concerns about the fairness of clinical prediction models. "Fair" models are defined as those for which their performance or predictions are not inappropriately influenced by protected attributes such as ethnic...
There are concerns about the fairness of clinical prediction models. "Fair" models are defined as those for which their performance or predictions are not inappropriately influenced by protected attributes such as ethnicity, gender, or socioeconomic status. Researchers have raised concerns that current algorithmic fairness paradigms enforce strict egalitarianism in healthcare, leveling down the performance of models in higher-performing subgroups instead of improving it in lower-performing ones. We propose assessing the fairness of a prediction model by expanding the concept of net benefit, using it to quantify and compare the clinical impact of a model in different subgroups. We use this to explore how a model distributes benefits across a population, its impact on health inequalities, and its role in the achievement of health equity. We show how resource constraints might introduce necessary trade-offs between health equity and other objectives of healthcare systems. We showcase our proposed approach with the development of two clinical prediction models: (1) a prognostic type 2 diabetes model used by clinicians to enroll patients into a preventive care lifestyle intervention programme and (2) a lung cancer screening algorithm used to allocate diagnostic scans across the population. This approach helps modelers better understand if a model upholds health equity by considering its performance in a clinical and social context.
Considerations for Estimating Causal Effects of Informatively Timed Treatments.
Oganisian A
Epidemiology · 2026 Mar · PMID 41505726 · Full text
Epidemiologic studies are often concerned with estimating causal effects of a sequence of treatment decisions on survival outcomes. In many settings, treatment decisions do not occur at fixed, prespecified follow-up time...
Epidemiologic studies are often concerned with estimating causal effects of a sequence of treatment decisions on survival outcomes. In many settings, treatment decisions do not occur at fixed, prespecified follow-up times. Rather, timing varies across subjects in ways that may be informative of subsequent treatment decisions and potential outcomes. Awareness of the issue and potential solutions is lacking in the literature, which motivates this work. Here, we formalize the issue of informative timing, problems associated with ignoring it, and show how g-methods can be used to analyze sequential treatments that are informatively timed. As we describe, in such settings, the waiting times between successive treatment decisions may be properly viewed as time-varying confounders. Using synthetic examples, we illustrate how g-methods that do not adjust for these waiting times may be biased and how adjustment can be done in scenarios where patients may die or be censored in between treatments. Finally, we provide implementation guidance and examples using publicly available software. Our concluding message is that (1) considering timing is important for valid inference and (2) correcting for informative timing can be done with g-methods that adjust for waiting times between treatments as time-varying confounders.
About
- Frequency
- Sun
- Papers found
- 200
- RSS feed
- Subscribe