American Journal of Epidemiology
ISSN / EISSN : 0002-9262 / 1476-6256
Published by: Oxford University Press (OUP) (10.1093)
Total articles ≅ 21,797
Latest articles in this journal
Published: 4 October 2022
American Journal of Epidemiology; https://doi.org/10.1093/aje/kwac170
Recombinant zoster vaccine (RZV) (Shingrix; GlaxoSmithKline, Brentford, United Kingdom) is an adjuvanted glycoprotein vaccine that was licensed in 2017 to prevent herpes zoster and its complications in older adults. In this prospective, post-licensure Vaccine Safety Datalink (VSD) study using electronic health records, we sequentially monitored a real-world population of adults aged 50 years and older who received care at multiple VSD health systems in the United States to identify potential increased risks of 10 pre-specified priority outcomes, including stroke, anaphylaxis, and Guillain-Barré Syndrome (GBS). Among 647,833 RZV doses administered from January 2018 through December 2019, we did not detect a sustained increased risk of any monitored outcome for RZV recipients relative either to historical (2013-2017) recipients of Zoster Vaccine Live (ZVL), a live-attenuated virus vaccine (Zostavax; Merck & Co., Inc., Kenilworth, New Jersey), or contemporary non-RZV vaccinated persons who had an annual well-visit during the 2018-2019 study period. We confirmed pre-licensure trial findings of increased risks of systemic and local reactions following RZV. Our study provides additional reassurance about the overall safety of RZV. Despite a large sample, uncertainty remains regarding potential associations with GBS due to the limited number of confirmed GBS cases that were observed.
Published: 4 October 2022
American Journal of Epidemiology; https://doi.org/10.1093/aje/kwac173
Along with age and race, sex has historically been a core stratification and control variable in epidemiological research. While in recent decades research guidelines and institutionalized requirements have incorporated an approach differentiating biological sex from social gender, neither sex nor gender is itself a unidimensional construct. The conflation of dimensions within and between sex and gender presents a validity issue wherein proxy measures are used for dimensions of interest, often without explicit acknowledgement or evaluation. Individual-level dimensions of sex and gender are outlined as a guide for epidemiologists. Two case studies are presented. The first demonstrates how unacknowledged use of a sex/gender proxy for a sexed dimension of interest (uterine status) resulted in decades of cancer research misestimating risks, racial disparities, and age trends. The second illustrates how a multidimensional sex and gender framework may be applied to strengthen research on COVID-19 incidence, diagnosis, morbidity and mortality. Considerations are outlined, including 1) addressing the match between measures and theory, and explicitly acknowledging and evaluating proxy use; 2) improving measurement across dimensions and social ecological levels; 3) incorporating multidimensionality into research objectives; 4) interpreting sex, gender, and their effects as biopsychosocial.
Published: 4 October 2022
American Journal of Epidemiology; https://doi.org/10.1093/aje/kwac171
Interest in using internet search data, such as that from the Google Health Trends Application Programming Interface (GHT-API), to measure epidemiologically relevant exposures or health outcomes is growing due to their accessibility and timeliness. Researchers input search term(s), geography and time period, and the GHT-API returns a scaled probability of that search term, given all searches within the specified geo-time period. In this study, we detail a method for using these data to measure a construct of interest in five iterative steps: first, identify phrases the target population may use to search for the construct of interest; second, refine candidate search phrases with incognito Google searches to improve sensitivity and specificity; third, craft the GHT-API search term(s) by combining the refined phrases; fourth, test search volume and choose geographic and temporal scales; and fifth, retrieve and average multiple samples to stabilize estimates and address missingness. An optional sixth step involves accounting for changes in total search volume by normalizing. We present a case study examining weekly state-level child abuse searches in the United States during the COVID-19 pandemic (January 2018-August 2020) as an application of this method and describe limitations.
Published: 29 September 2022
American Journal of Epidemiology; https://doi.org/10.1093/aje/kwac164
Arterial blood oxygen saturation measured by pulse oximetry (SpO2) may be differentially less accurate for people with darker skin pigmentation, which could potentially affect COVID-19 treatment course. We analyzed pulse oximeter accuracy and association with COVID-19 treatment outcomes using electronic health record (EHR) data from Sutter Health, a large, mixed-payer, integrated healthcare delivery system in northern California, United States (US). We analyzed two cohorts: (1) 43,753 concurrent arterial blood gas (ABG) oxygen saturation (SaO2)/SpO2 measurement pairs taken January 2020-February 2021 for Non-Hispanic white (NHW) or Non-Hispanic Black/African American (NHB) adults, and (2) 8,735 adults who went to the emergency department (ED) with COVID-19 July 2020-February 2021. Pulse oximetry systematically overestimated blood oxygenation by 1% more in NHB individuals than in NHW individuals. For people with COVID-19, this was associated with lower admission probability (-3.1 percentage-points), dexamethasone treatment (-3.1 percentage-points), and supplemental oxygen treatment (-4.5 percentage-points), as well as increased time-to-treatment: +37.2 minutes before dexamethasone initiation and +278.5 minutes before initiation of supplemental oxygen. These results call for additional investigation of pulse oximeters, and suggest that current guidelines for development, testing, and calibration of these devices should be revisited, investigated, and revised.
Published: 28 September 2022
American Journal of Epidemiology; https://doi.org/10.1093/aje/kwac167
Published: 26 September 2022
American Journal of Epidemiology; https://doi.org/10.1093/aje/kwac168
This study aimed to examine the associations of increases in the duration of education with back pain using the exogenous variation generated by the English schooling reforms of 1947 and 1972. We analyzed cross-sectional data derived from nine waves (waves 1–9; 2002–2019) of the English Longitudinal Study of Ageing. An instrumental variables regression using two-stage least squares with the two-way cluster-robust standard error was used. The mean severity of back pain, measured using the Numerical Rating Scale, was used as the outcome. A total of 22,868 observations from 5,070 participants were included (the 1947 reform = 16,565 observations from 3,231 participants, mean age = 74.5; the 1972 reform = 6,303 observations from 1,839 participants, mean age = 59.3). The schooling reforms significantly extended years of school attendance by a mean of 0.57 years for the 1942 reform cohort and 0.66 years for 1972 reform cohort. For participants born within ±5 years of the pivotal cohorts, an additional year of education decreased the severity of back pain by 0.78 points (95% confidence interval, 0.65–0.92) for the 1972 reform cohort. Our finding underscores the importance of the length of education in the reduction of back pain in middle-aged individuals.
Published: 21 September 2022
American Journal of Epidemiology; https://doi.org/10.1093/aje/kwac166
Little research has investigated the long-term relationship between low wages and memory decline, despite the growing share of low-wage workers in the US labor market. Here, we examine whether cumulative exposure to low wages over 12 years in midlife is associated with memory decline in later life. Using 1992-2016 data from the Health and Retirement Study, we analyzed data from 2,879 individuals born 1936-1941 using confounder-adjusted linear mixed-effects models. Low-wage was defined as hourly wage lower than two-thirds of the federal median wage for the corresponding year and categorized into ‘never’, ‘intermittent’, and ‘sustained’ based on wages earned from 1992-2004. Memory function was measured at each visit from 2004-2016 by a memory composite score. The confounder-adjusted annual rate of memory decline among ‘never’ low-wage earners was -0.12 standard units, 95% CI: [-0.13, -0.10]. Compared with this, memory decline among workers with sustained earning of low midlife wages was significantly faster (βtime*sustained:-0.014, 95% CI: [-0.02, -0.01]), corresponding to an annual rate of -0.13 standard units for this group. Sustained low-wage earning in midlife was significantly associated with a downward trajectory of memory performance in older age. Enhancing social policies to protect low-wage workers may be especially beneficial for their cognitive health.
Published: 20 September 2022
American Journal of Epidemiology; https://doi.org/10.1093/aje/kwac163
Thyroid cancer incidence is higher in women than men, especially during the reproductive years, for reasons that remain poorly understood. Using population-based registry data from four Nordic countries through 2015, we examined associations of perinatal characteristics with risk of maternal thyroid cancer. Cases were women diagnosed with thyroid cancer ≥2 years after last birth (n=7,425, 83% papillary). Cases were matched to controls (n=67,903) by mother’s birth year, country, and county of residence. Odds ratios (ORs) were estimated using conditional logistic regression models adjusted for parity. Older age at first pregnancy, postpartum hemorrhage (OR=1.18, 95% CI 1.08−1.29), and benign thyroid conditions (ORs ranging from 1.64 for hypothyroidism to 10.35 for thyroid neoplasms) were associated with increased thyroid cancer risk, as were higher offspring birth weight (per 1-kg increase, OR=1.17, 95% CI 1.12−1.22) and large-for-gestational-age (OR=1.26, 95% CI 1.11−1.43). Unmarried/non-cohabiting status (OR=0.91, 95% CI 0.84−0.98), maternal smoking (OR=0.75, 95% CI 0.67−0.84), and preterm birth (OR=0.90, 95% CI 0.83−0.98) were associated with reduced risk. Several factors (e.g., older age at first pregnancy, maternal smoking, goiter, benign neoplasms, postpartum hemorrhage, and hyperemesis gravidarum, neonatal jaundice) were associated with advanced thyroid cancer. These findings suggest that some perinatal exposures may influence maternal thyroid cancer risk.
Published: 20 September 2022
American Journal of Epidemiology; https://doi.org/10.1093/aje/kwac165
Mediation analysis can be applied to data from randomized trials of health and social interventions to draw causal inference concerning their mechanisms. We used data from a cluster randomized trial in Nicaragua, fielded between 2000 and 2002, to investigate whether the impact of providing access to a conditional cash transfer program on child nutritional outcomes was mediated by child health checkups and household dietary diversity. In a sample of 443 children 6–35 months old, we estimated the controlled direct effect (CDE) of random assignment on measured height-for-age z-scores had we intervened so that all children received a health checkup and had the same level of household dietary diversity, using inverse probability weighted marginal structural models to account for mediator-outcome confounding. Sensitivity analyses corrected the CDE for potential non-differential error in the measurement of dietary diversity. Treatment assignment increased height-for-age by 0.37 (95%CI=0.05, 0.69) standard deviations (SDs). The CDE was 0.20 (95%CI=-0.17, 0.57) SDs, suggesting nearly one-half of the program’s impact on child nutrition would be eliminated had we intervened on these factors, although estimates were relatively imprecise. This study provides an illustration of how causal mediation analysis can be applied to examine the mechanisms of multifaceted interventions.
Published: 16 September 2022
American Journal of Epidemiology; https://doi.org/10.1093/aje/kwac162
Inverse probability weighting (IPW) and g-computation are commonly used in time-varying analyses. To inform decisions on which to use, we compared these methods using a plasmode simulation, based on the Effects of Aspirin in Gestation and Reproduction trial. In our main analysis, we simulated 1226 individuals, followed for up to 10 weeks. The exposure was weekly exercise, and the outcome was time to pregnancy. We controlled for 6 confounders: 4 baseline (race, ever smoker, age, and BMI) and 2 time-varying (compliance to assigned treatment and nausea). We sought to estimate the average causal risk difference by 10 weeks, using IPW and g-computation implemented using a Monte Carlo (MC) estimator and iterated conditional expectations (ICE). Across 500 simulations, we compared the bias, empirical standard error (ESE), average standard error, standard error ratio, and 95% confidence interval coverage of each approach. IPW (bias: 0.017; ESE: 0.039; coverage: 92.6%) and MC g-computation (bias: -0.012; ESE: 0.031; coverage: 94.2%) performed similarly. ICE g-computation was the least biased but least precise estimator (bias: 0.010; ESE: 0.058; coverage: 93.4%). When choosing an estimator, one should consider factors like the research question, prevalence of the exposure and outcome, and the number of time points being analyzed.