American journal of epidemiology最新文献

Evidence on the association between education and cognitive decline is mixed. Recent systematic reviews suggest no association, but most evidence comes from high-income contexts. We used data from the Longitudinal Aging Study in India - Diagnostic Assessment of Dementia (LASI-DAD; N=3,673; 2017-2024) to estimate the association between educational attainment and cognitive decline using linear mixed effects models. We evaluated findings across a range of models and explored the roles of practice effects, selective survival, and assumptions about the underlying causal structure using supplementary analyses. Despite large differences in cognitive level (those with higher education had higher cognition in wave 1), compared to those with no education, those with less than primary school (difference of -0.03 SD units/year; 95% CI -0.04 to -0.01), primary school (-0.04; -0.06 to -0.03), middle-secondary school (-0.06; -0.07 to -0.04), and higher secondary school or above (-0.05; -0.07 to -0.03) had steeper cognitive decline after adjustment for demographic and early-life socioeconomic factors. Neither practice effects nor selective survival could explain findings, though findings were consistent with one of the two simulated causal structures. Analyses illustrate challenges in analyzing longitudinal cognitive data and showcase potential approaches to rule out methodological explanations and probe results.

Migraine is a neurological disease associated with adverse perinatal outcomes. We examined the separate and combined impacts of migraine and comorbidity on risks of severe maternal morbidity/mortality (SMM-M) and severe neonatal morbidity/mortality (SNM-M). This population-based cohort study of pregnancies (n=2,643,335) in Ontario, Canada, 2007-2022, compared females with (1) pre-pregnancy migraine and ≥1 other chronic conditions, (2) migraine alone, (3) other chronic conditions alone, and (4) neither migraine/other chronic conditions (referent) using modified Poisson regression. Attributable proportion due to interaction (aAP) reflected additive interaction between migraine and comorbidity. In the cohort, 6.8% had migraine and other chronic conditions, 3.2% migraine alone, 45.7% other chronic conditions alone, and 44.3% neither. The incidence of SMM-M was 1.6%, while SNM-M affected 7.1% of neonates. Risks of SMM-M and SNM-M were greatest in those doubly exposed (aRRSMM-M 1.60, 95% CI 1.54-1.66; aRRSNM-M 1.43, 1.39-1.46), followed by other chronic conditions alone (aRRSMM-M 1.34, 1.32-1.37; aRRSNM-M 1.30, 1.28-1.32), and migraine alone (aRRSMM-M 1.13, 1.07-1.20; aRRSNM-M 1.07, 1.04-1.11). Additive interaction was small for SMM-M (aAP 7.4%, 2.2-12.5) and SNM-M (aAP 3.7%, 0.3-6.9). Although synergistic effects were small, findings suggest individuals with migraine and comorbidity could benefit from preconception and perinatal supports to reduce their risks of perinatal complications.

The National Institutes of Health's All of Us Research Program (All of Us) aims to enhance precision medicine by collecting multimodal data from one million or more participants. Because All of Us prioritizes enrollment from populations for which there is limited data on health outcomes using nonprobability sampling methods, prevalence estimates may not reflect those of the general U.S. population. This study examines the challenges of estimating electronic health record-based disease prevalence from All of Us and offers a framework and novel R package (waou) to help researchers consider these complex issues. We investigated the application of three weighting techniques to improve generalizability for dementia, type 2 diabetes, and depression prevalence estimates. Using data from All of Us alongside the National Health Interview Survey as a benchmark, we found that weighting approaches yielded more representative estimates for dementia and type 2 diabetes, yet amplified bias for depression. Waou is presented as a tool to facilitate the application of these methodologies, empowering researchers to critically evaluate the generalizability of their estimates. This work underscores the need for careful consideration of bias in epidemiological research when using the All of Us dataset for population-level inferences.

Wildfire-induced fine particulate matter (PM2.5) poses a growing public health risk. To examine public responses to wildfire smoke, we analyzed smoke PM2.5 and Google Trends data across 11 Designated Market Areas (DMAs) in California from 2016 to 2020. Scaled Relative Search Interest (RSI) for 'air pollution' and 'air purifier' was used as a proxy for public awareness and preparedness. Applying a Bayesian spatiotemporal distributed lag model, we found that a one-unit increase in smoke PM2.5 (6.6 μg/m3) was associated with 149% (95% Credible Interval (CrI): 107%, 486%) higher 'air pollution' searches and 34% (95% CrI: 3%, 76%) and 68% (95% CrI: 26%, 123%) higher 'air purifier' searches over the first two weeks, respectively. Among all DMAs, Palm Springs showed the strongest responses. Public search responses were amplified under higher temperature and pressure but decreased with higher wind speed and humidity. Communities with greater socioeconomic vulnerability exhibited lower search interest in both awareness- and preparedness-related terms. These findings highlight the need for timely, targeted, and equitable public communication strategies that align with real-time environmental conditions and improve access to protective resources among vulnerable populations.

The 32.5-year follow-up of the Lung Health Study (LHS) published in this issue highlights the long-term impact of a well-executed randomized clinical trials (RCTs) evaluating a smoking cessation intervention. Between 1986 and 1989, the LHS enrolled 5,887 smokers aged 35-59 with mild-to-moderate airway obstruction across ten North American sites and randomized them to a 10-day smoking cessation intervention with placebo inhaler, the same intervention with ipratropium bromide, or usual care. The latest 32.5-year analysis confirms a reduction in respiratory-related mortality, though the earlier observed all-cause mortality benefit observed at 14.5 years was not seen. The updated analysis excludes 608 participants from the one Canadian clinic and their baseline smoking characteristics differed from the US participants. Regardless, qualitative effect modification is unlikely. Shifts in leading causes of death over time, competing risks, and potential postrandomization selection bias are challenges inherent in extended follow-up, yet the findings of reduced respiratory mortality for participants assigned to smoking cessation stood the test of time. This publication highlights the importance of trial conduct, data preservation, and the value of long-term follow-up using the National Death Index. Congratulations to the authors for this fourth update and to all LHS researchers who contributed to this landmark clinical trial.

Obstructive sleep apnea (OSA) may cause functional hearing loss through ischemic damage to the cochlea or impaired auditory processing in the brain, but longitudinal evidence is lacking. Medicare beneficiaries completed the National Health and Aging Trends Study round 3 (2013) sleep module (n = 1433, 41% male, 21% Black) were followed until round 11 (2021). Elevated OSA risk was defined as a modified STOP-BANG (Snoring, Tiredness, Observed apneas, high blood Pressure, BMI, Age, Neck circumference, Gender) score ≥ 3. Functional hearing loss was defined as self-reported deafness, hearing aid use, or inability to hear well enough to use the telephone or have a conversation in a room with the television or radio on. Discrete-time complementary log-log models were used for incident functional hearing loss. Multinomial logistic regression was used for membership in hearing loss trajectory groups (stable normal hearing, transition to hearing loss, stable hearing loss) estimated using group-based trajectory modeling. Elevated OSA risk score was associated with 1.34 times the hazard of incident functional hearing loss (95% confidence interval [CI], 1.06-1.70) and with a higher likelihood of being in the stable hearing loss group (relative risk ratio = 1.79, 95% CI, 1.32-2.42). Screening and treating OSA may be important for hearing health.

Twin studies have been underutilized for examining female-specific etiology of hypertension (HT). The discordant co-twin design offers unique strengths not afforded by other approaches primarily because relatedness and shared origins naturally match twin pairs more closely on genetic and early environmental confounders. We aimed to confirm HT-associated factors across the lifecourse among 202 female twin pairs identified from the California Twin Program (CTP) registry. A baseline questionnaire queried current and past risk factors and self-reported prevalent HT. HT-discordant pairs were compared for differences in demographics, growth and body size, reproductive factors, and lifestyle factors. Data were analyzed using conditional logistic regression restricted to nonsmokers and stratified by zygosity. Weighing ≥150lbs. versus <130lbs. at 18 years was associated with 3-fold odds of HT, and weighing ≥180lbs. at the time of questionnaire was associated with 6-fold odds of HT. Exercising more than one's co-twin was associated with 0.54-fold odds of HT. Menopause was positively associated with HT, with post-menopausal DZ twins experiencing 4-fold odds of HT compared to pre-menopausal co-twins. This study with minimal bias using unaffected co-twins confirmed multiple factors associated with female HT including overweight at age 18 and in adulthood, lower level of exercise, and menopausal.

Uterine serous carcinoma (USC) is a rare diagnosis but is associated with high mortality. There are limited data to guide adjuvant treatment decisions in early-stage disease. The purpose of this study is to evaluate the impact of adjuvant therapy on recurrence-free survival (RFS) and overall survival (OS) in early-stage USC. Patients with stage I and II USC treated at a single institution from January 2006 to December 2019 were identified. Demographic, clinicopathologic, treatment, and outcome data were collected. Data were compared using descriptive statistics. Survival analyses were performed using Kaplan-Meier and Cox proportional hazards methods. Ninety-four patients were identified. Median follow-up time was 33.5 months. The median age was 68 years (range, 49-87), the majority of patients were White (n = 78, 83.0%), and the median BMI was 30.7 (range, 14.2-57.3). Minimally invasive surgical staging was performed in 59.6% of cases (n = 56). Most patients had stage IA disease (n = 70, 74.5%). Most patients (n = 79, 84.0%) received adjuvant therapy, and a majority of patients received a combination of systemic chemotherapy and radiation therapy (n = 55, 58.5%), with the most common combination being chemotherapy plus vaginal brachytherapy (n = 42, 44.7%). Most patients (n = 77, 81.9%) remain without evidence of disease, while 17 patients (18.1%) have recurred. Patients receiving 6 cycles of adjuvant chemotherapy experienced improved OS (P = .004) and improved RFS (P = .02) compared to those receiving no adjuvant chemotherapy. Patients with early-stage USC who received 6 cycles of adjuvant chemotherapy had significantly improved OS and RFS when compared to those patients who did not receive adjuvant chemotherapy. This article is part of a Special Collection on Gynecological Cancer.

Previous studies on the impact of the COVID-19 pandemic on suicide trends have yielded inconsistent findings, with inadequate differentiation of short- and long-term effects. Addressing this gap, our study evaluates the pandemic's short- and long-term trends in suicide attempts across various demographics. Illinois statewide hospital data covering January 1, 2016, through December 31, 2022, were analyzed. Compared to the prepandemic baseline period, changes in daily rates of suicide attempts treated across 4 periods of state executive orders limiting public gatherings were evaluated. Segmented time-series analysis was developed to evaluate specific temporal patterns across 4 pandemic periods. A total of 153 476 suicide attempts were registered, with 38.6% of cases occurring after March 16, 2020. A significant increase of 0.8353 attempts per day per 1 million (P < .0001) was found after the second "stay-at-home orders" period. A significant increase in suicide attempts was observed in females 11 to 16 years old, starting from the second "stay-at-home order" through the end of 2022, and in females 19 to 24 years old after the end of the second "stay-at-home order." The rise in suicide attempts among young females during the pandemic's later stages is alarming. Further investigation into specific risk factors is needed. This article is part of a Special Collection on Methods in Social Epidemiology.

Tree-based scan statistics (TBSS) are data mining methods that screen thousands of hierarchically related health outcomes to detect unsuspected adverse drug effects. TBSS traditionally analyze claims data with outcomes defined via diagnosis codes. TBSS have not been previously applied to rich clinical information in electronic health records (EHR). We developed approaches for integrating EHR data in TBSS analyses, including outcomes derived from natural language processing (NLP) applied to clinical notes and laboratory results, related via multipath hierarchical structures. We consider 4 settings that sequentially add sources of outcomes to the TBSS tree: (1) diagnosis code, (2) NLP-derived outcomes, (3) binary outcomes from lab results, and (4) continuous lab results. In a comparative cohort study involving second-generation sulfonylureas (SUs) and dipeptidyl peptidase 4 (DPP-4) inhibitors among adults with type 2 diabetes, with an a priori expected signal of hypoglycemia, diagnosis code data showed no statistical alerts for inpatient or emergency department settings. Adding NLP-derived outcomes resulted in an alert for "Headaches" (P = .047), a nonspecific symptom of hypoglycemia. Progressively adding binary and continuous lab results produced the same alert. Integrating EHR in TBSS can be useful for the detection of safety signals for further investigation.