American journal of epidemiology最新文献

Epidemiologists spend a great deal of time on confounding in our teaching, in our methods development, and in our assessment of study results. This may give the impression that uncontrolled confounding is the biggest problem observational epidemiology faces, when in fact, other sources of bias such as selection bias, measurement error, missing data, and misalignment of zero time may often (especially if they are all present in a single study) lead to a stronger deviation from the truth. Compared with the amount of time we spend teaching how to address confounding in data analysis, we spend relatively little time teaching methods for simulating confounding (and other sources of bias) to learn their impact and develop plans to mitigate or quantify the bias. Here we review the accompanying paper by Desai et al (Am J Epidemiol. 2024;193(11):1600-1608), which uses simulation methods to quantify the impact of an unmeasured confounder when it is completely missing or when a proxy of the confounder is measured. We discuss how we can use simulations of sources of bias to ensure that we generate better and more valid study estimates, and we discuss the importance of simulating realistic datasets with plausible bias structures to guide data collection. This article is part of a Special Collection on Pharmacoepidemiology.

Unmeasured confounding is often raised as a source of potential bias during the design of nonrandomized studies, but quantifying such concerns is challenging. We developed a simulation-based approach to assess the potential impact of unmeasured confounding during the study design stage. The approach involved generation of hypothetical individual-level cohorts using realistic parameters, including a binary treatment (prevalence 25%), a time-to-event outcome (incidence 5%), 13 measured covariates, a binary unmeasured confounder (u1; 10%), and a binary measured "proxy" variable (p1) correlated with u1. Strengths of unmeasured confounding and correlations between u1 and p1 were varied in simulation scenarios. Treatment effects were estimated with (1) no adjustment, (2) adjustment for measured confounders (level 1), and (3) adjustment for measured confounders and their proxy (level 2). We computed absolute standardized mean differences in u1 and p1 and relative bias with each level of adjustment. Across all scenarios, level 2 adjustment led to improvement in the balance of u1, but this improvement was highly dependent on the correlation between u1 and p1. Level 2 adjustments also had lower relative bias than level 1 adjustments (in strong u1 scenarios: relative bias of 9.2%, 12.2%, and 13.5% at correlations of 0.7, 0.5, and 0.3, respectively, vs 16.4%, 15.8%, and 15.0% for level 1). An approach using simulated individual-level data is useful to explicitly convey the potential for bias due to unmeasured confounding while designing nonrandomized studies, and can be helpful in informing design choices. This article is part of a Special Collection on Pharmacoepidemiology.

Background: Observational studies using real-world data (RWD) can address gaps in knowledge on deprescribing medications but are subject to methodological issues. Limited data exist on the methods employed to use RWD to measure the effects of deprescribing.

Objective: To describe methodological approaches used in observational studies of deprescribing medications in older adults.

Method: We conducted a systematic review in Medline for observational studies published in English (01/01/2000-09/14/2023) that examined the health effects of medication deprescribing in older adults. We described study characteristics and methods, focusing on the operationalization of deprescribing as an exposure and potential time-related biases.

Results: Forty-five studies were included, representing a variety of drug classes (e.g., statins, aspirin, bisphosphonates) and diseases. Most studies adequately addressed potential time-related biases. The definition of deprescribing was not clearly defined in 12 studies. There was heterogeneity regarding the minimum duration of time that qualified as deprescribing, even within a drug class; fewer than one-third of studies provided a justification for these definitions.

Conclusion: Observational studies are common to examine the effects of deprescribing; however, there were inconsistencies in measuring deprescribing and a lack of transparency in reporting. There is a need for minimum sufficient reporting criteria for observational studies on deprescribing.

Childhood adversity is an important risk factor for adverse health across the life course. Epigenetic modifications, such as DNA methylation (DNAm), are a hypothesized mechanism linking adversity to disease susceptibility. Yet, few studies have determined whether adversity-related DNAm alterations are causally related to future health outcomes or if their developmental timing plays a role in these relationships. Here, we used 2-sample mendelian randomization to obtain stronger causal inferences about the association between adversity-associated DNAm loci across development (ie, birth, childhood, adolescence, and young adulthood) and 24 mental, physical, and behavioral health outcomes. We identified particularly strong associations between adversity-associated DNAm and attention-deficit/hyperactivity disorder, depression, obsessive-compulsive disorder, suicide attempts, asthma, coronary artery disease, and chronic kidney disease. More of these associations were identified for birth and childhood DNAm, whereas adolescent and young adulthood DNAm were more closely linked to mental health. Childhood DNAm loci also had primarily risk-suppressing relationships with health outcomes, suggesting that DNAm might reflect compensatory or buffering mechanisms against childhood adversity rather than acting solely as an indicator of disease risk. Together, our results suggest adversity-related DNAm alterations are linked to both physical and mental health outcomes, with particularly strong impacts of DNAm differences emerging earlier in development.

High and low daily ambient temperatures are associated with higher mortality in the general population. People experiencing homelessness (PEH) are thought to be particularly vulnerable, but there is almost no direct evidence available. We examined the temperature-mortality association among PEH in 2 populous, urban counties in hot-climate regions of the United States, focusing on heat effects. Study setting was Los Angeles County, CA, and Clark County, NV, which encompass the cities of Los Angeles and Las Vegas, respectively. Outcomes were 2015-2022 deaths among decedents categorized as homeless in county administrative records. We used quasi-Poisson distributed lag nonlinear models to estimate the association of mortality with daily temperatures and with 7-day lagged temperatures, adjusting for day of week, seasonality, and long-term trends. We estimated the minimum mortality temperature and fraction of mortality attributable to temperatures above and below minimum mortality temperature. The association between daily temperature and PEH mortality was skewed towards greater risk at higher temperatures, especially in Clark County. Temperature-attributable mortality equaled 50.1% of deaths in Clark County (95% CI, 29.0-62.8) and 7.0% in Los Angeles County (95% CI, 1.4-12.1). In both counties, most temperature-attributable deaths were attributable to heat rather than cold. In these hot-climate urban counties, our estimates of heat-attributable mortality among PEH were orders of magnitude greater than those reported in prior research on the general population. These results indicate that temperature vulnerability, particularly heat vulnerability, requires stronger public health and policy responses. This article is part of a Special Collection on Environmental Epidemiology.

Clinicians, researchers, regulators, and other decision-makers increasingly rely on evidence from real-world data (RWD), including data routinely accumulating in health and administrative databases. RWD studies often rely on algorithms to operationalize variable definitions. An algorithm is a combination of codes or concepts used to identify persons with a specific health condition or characteristic. Establishing the validity of algorithms is a prerequisite for generating valid study findings that can ultimately inform evidence-based health care. In this paper, we aim to systematize terminology, methods, and practical considerations relevant to the conduct of validation studies of RWD-based algorithms. We discuss measures of algorithm accuracy, gold/reference standards, study size, prioritization of accuracy measures, algorithm portability, and implications for interpretation. Information bias is common in epidemiologic studies, underscoring the importance of transparency in decisions regarding choice and prioritizing measures of algorithm validity. The validity of an algorithm should be judged in the context of a data source, and one size does not fit all. Prioritizing validity measures within a given data source depends on the role of a given variable in the analysis (eligibility criterion, exposure, outcome, or covariate). Validation work should be part of routine maintenance of RWD sources. This article is part of a Special Collection on Pharmacoepidemiology.

Suicide remains a leading cause of death in the United States, and recent data suggest suicide deaths involving opioids are increasing. Given unprecedented increases in drug-poisoning deaths, suicidality, and suicide deaths in recent years, an updated examination of the trends in suicide deaths involving opioids is warranted. In this descriptive epidemiologic analysis, we leverage final and provisional mortality data from the US Centers for Disease Control and Prevention's WONDER database to examine trends in suicide deaths involving opioid poisoning from 1999 to 2021 by biological sex. Results reveal complex changes over time: the number and age-adjusted rate of suicide deaths involving opioid poisoning among male and female residents tended to track together, and both increased through 2010, but then diverged, with the number and rate of suicide deaths involving opioid poisoning among female residents outpacing that of male residents. However, the number and rate of suicide deaths involving opioid poisoning among male residents then began to stabilize, while that of female residents declined, closing the sex-based gap. Across all years of data, the proportion of suicide deaths that involved opioid poisoning was consistently higher among female decedents (5.8%-11.0%) compared with male decedents (1.4%-2.8%). Findings have implications for improved suicide prevention and harm reduction efforts. This article is part of a Special Collection on Mental Health.

Real-world evidence (RWE) studies are increasingly used to inform policy and clinical decisions. However, there remain concerns about the credibility and reproducibility of RWE studies. While there is universal agreement on the critical importance of transparent and reproducible science, the building blocks of open science practice that are common across many disciplines have not yet been built into routine workflows for pharmacoepidemiology and outcomes researchers. Observational researchers should highlight the level of transparency of their studies by providing a succinct statement addressing study transparency with the publication of every paper, poster, or presentation that reports on an RWE study. In this paper, we propose a framework for an explicit transparency statement that declares the level of transparency a given RWE study has achieved across 5 key domains: (1) protocol, (2) preregistration, (3) data, (4) code-sharing, and (5) reporting checklists. The transparency statement outlined in the present paper can be used by research teams to proudly display the open science practices that were used to generate evidence designed to inform public health policy and practice. While transparency does not guarantee validity, such a statement signals confidence from the research team in the scientific choices that were made.