J Michael Oakes’s research while affiliated with Case Western Reserve University and other places

Investigators traditionally use randomized designs and corresponding analysis procedures to make causal inferences about the effects of interventions, assuming independence between an individual’s outcome and treatment assignment and the outcomes of other individuals in the study. Often, such independence may not hold. We provide examples of interdependency in model organism studies and human trials and group effects in aging research and then discuss methodologic issues and solutions. We group methodologic issues as they pertain to (1) single-stage individually randomized trials; (2) cluster-randomized controlled trials; (3) pseudo-cluster-randomized trials; (4) individually randomized group treatment; and (5) two-stage randomized designs. Although we present possible strategies for design and analysis to improve the rigor, accuracy and reproducibility of the science, we also acknowledge real-world constraints. Consequences of nonadherence, differential attrition or missing data, unintended exposure to multiple treatments and other practical realities can be reduced with careful planning, proper study designs and best practices. In this Perspective, the authors discuss experimental scenarios that breach the assumption of independence of all samples or participants in a study, specifically in aging research. They outline various strategies to improve the rigor and accuracy of the science with design and analysis solutions, while also considering real-world constraints.

Prevention and control of infectious diseases in livestock is dependent upon perceived risk and susceptibility, including the prevention of between-herd transmission of bovine tuberculosis through introductions of cattle to susceptible herds. To examine how perceived risk and susceptibility can help to inform policymaking in disease surveillance and control, we used factorial surveys to profile risk perceptions of cattle producers. We found that government indemnity and slaughtering policy did not impact the cattle purchasing behavior of producers who responded to our survey, but rather through other attributes such as the reliability or reputation of the seller. In addition, we identified significant production type and gender differences in purchasing behavior and risk perception. Finally, clustering analysis revealed a group of high-risk respondents characterized as experienced and very dedicated owners of established medium to large size herds. With the increasing availability of business data, assessment of producer’s behavior, personalities and attitudes allows policymakers to understand the needs of cattle producers and develop tailored programs that will improve producer cooperation with government agencies.

Background Cluster randomized controlled trials (cRCTs) are increasingly used but must be analyzed carefully. We conducted a simulation study to evaluate the validity of a parametric bootstrap (PB) approach with respect to the empirical type I error rate for a cRCT with binary outcomes and a small number of clusters. Methods We simulated a case study with a binary (0/1) outcome, four clusters, and 100 subjects per cluster. To compare the validity of the test with respect to error rate, we simulated the same experiment with K=10, 20, and 30 clusters, each with 2,000 simulated datasets. To test the null hypothesis, we used a generalized linear mixed model including a random intercept for clusters and obtained p-values based on likelihood ratio tests (LRTs) using the parametric bootstrap method as implemented in the R package “pbkrtest”. Results The PB test produced error rates of 9.1%, 5.5%, 4.9%, and 5.0% on average across all ICC values for K=4, K=10, K=20, and K=30, respectively. The error rates were higher, ranging from 9.1% to 36.5% for K=4, in the models with singular fits (i.e., ignoring clustering) because the ICC was estimated to be zero. Conclusion Using the parametric bootstrap for cRCTs with a small number of clusters results in inflated error rates and is not valid

Social epidemiology is concerned with how social forces influence population health. Rather than focusing on a single disease (as in cancer or cardiovascular epidemiology) or a single type of exposure (e.g., nutritional epidemiology), social epidemiology encompasses all the social and economic determinants of health, both historical and contemporary. These include features of social and physical environments, the network of relationships in a society, as well as the institutions, politics, policies, norms and cultures that shape all of these forces. This commentary presents the perspective of several editors at the American Journal of Epidemiology with expertise in social epidemiology. We articulate our thinking to encourage submissions to the journal that: (i) expand knowledge of emerging and under-researched social determinants of population health; (ii) advance new empirical evidence on the determinants of health inequities and solutions to advance health equity; (iii) generate evidence to inform the translation of research on social determinants of health into public health impact; (iv) contribute to innovation in methods to improve the rigor and relevance of social epidemiology; and (v) encourage critical self-reflection on the direction, challenges, successes, and failures of the field.

Background Exposure to adverse childhood experiences (ACEs) and being female are distinct risk factors for having a major depressive episode (MDE) or an anxiety disorder (AD) in adulthood, but it is unclear whether these two risk factors are synergistic. The purpose of this study was to determine whether exposure to ACEs and being female are more than additive (synergistic) in their association with MDE and AD in US adults. Methods We pooled cross-sectional survey data in the Midlife in the United States study from two nationally-representative cohorts of English-speaking US adults. Data from the first cohort were collected in 2004–2006 and from the second in 2011–2014. Data from both cohorts included the 12-month prevalence of MDE and AD (generalized anxiety disorder or panic disorder) assessed with the Composite International Diagnostic Interview Short Form, gender (here termed female and male), and the count of five categories of exposure to ACEs: physical, sexual, or emotional abuse; household alcohol or substance abuse; and parental separation or divorce. Results Of the 5834 survey respondents, 4344 (74.5%) with complete data on ACEs were included in the analysis. Mean (SD) age was 54.1 (13.8) years and 53.9% were female. The prevalences of MDE, AD, and exposure to 3–5 categories of ACEs were 13.7, 10.0, and 12.5%, respectively. After adjusting for covariates (age, race, and current and childhood socioeconomic disadvantage), for those with both risk factors (female and 3–5 ACEs) the prevalence of MDE was 26.9%. This was 10.2% (95% CI: 1.8, 18.5%) higher than the expected prevalence based on the additive associations of the two risk factors. The adjusted prevalence of AD among females with 3–5 ACEs was 21.9%, which was 11.4% (95% CI: 4.0, 18.9%) higher than the expected prevalence. Conclusions For both MDE and AD, there was synergy between the two risk factors of exposure to ACEs and being female. Identification and treatment of MDE and AD may benefit from understanding the mechanisms involved in the synergistic interaction of gender with ACEs.

Randomization is an important tool used to establish causal inferences in studies designed to further our understanding of questions related to obesity and nutrition. To take advantage of the inferences afforded by randomization, scientific standards must be upheld during the planning, execution, analysis, and reporting of such studies. We discuss ten errors in randomized experiments from real-world examples from the literature and outline best practices for their avoidance. These ten errors include: representing nonrandom allocation as random, failing to adequately conceal allocation, not accounting for changing allocation ratios, replacing subjects in nonrandom ways, failing to account for non-independence, drawing inferences by comparing statistical significance from within-group comparisons instead of between-groups, pooling data and breaking the randomized design, failing to account for missing data, failing to report sufficient information to understand study methods, and failing to frame the causal question as testing the randomized assignment per se. We hope that these examples will aid researchers, reviewers, journal editors, and other readers to endeavor to a high standard of scientific rigor in randomized experiments within obesity and nutrition research.

Objective To report perspectives of participants in a food benefit program that includes FAS restrictions and FAS restrictions paired with F/V incentives. Design Randomized experimental trial in which participant perspectives were an exploratory study outcome. Setting Participants were randomized into one of three SNAP-like food benefit program groups - (1) Restriction: not allowed to buy FAS with benefits; (2) Restriction paired with incentive: not allowed to buy FAS with benefits and 30% financial incentive on eligible F/V purchased using benefits; or (3) Control: Same food purchasing rules as SNAP. Participants were asked questions to assess program satisfaction. Participants Adults in the Minneapolis-St. Paul, MN metropolitan area, eligible for but not currently participating in SNAP who completed baseline and follow-up study measures (n=254). Results Among remaining households in each group, most found the program helpful in buying nutritious foods (88.2%-95.7%) and were satisfied with the program (89.1%-93.0%). Sensitivity analysis results indicate that reported helpfulness and satisfaction with the program may in some instances be lower among the Restriction and the Restrictions paired with Incentive groups in comparison to the control group. Conclusions A food benefit program that includes restriction on purchase of FAS or restriction paired with a financial incentive for F/V purchases may be acceptable to most SNAP-eligible households with children.

Rational & Objective Screening for chronic kidney disease (CKD) is recommended for patients with diabetes and hypertension as stated by their respective societies. However, CKD, a silent disease usually detected at later stages, is associated with low socioeconomic status (SES). We assessed whether adding census tract SES status to the standard screening approach improves our ability to identify patients with CKD. Study Design Screening test analysis Settings & Participants Electronic health records (EHR) of 256,162 patients seen at a healthcare system in the 7 county Minneapolis/St.Paul area and linked census tract data. Exposures First quartile of census tract SES [median value of owner occupied housing units (<$165,200), average household income (<$35,935), percentage of residents >25 years with a Bachelor’s degree or more (<20.4%], hypertension and diabetes Outcomes CKD [eGFR< 60/mL/min/1.73m² or UACR>30mg/g or UPCR >150mg/g or UA>30mg/g] Analytic Approach Sensitivity, specificity, and number needed to screen (NNS) to detect CKD if we screen patients who have hypertension and/or diabetes and/or who live in low SES tracts (belonging to the first quartile of any of the three measures of tract SES) vs. standard approach. Results CKD was prevalent in 13% of our cohort. Sensitivity, specificity, and NNS of detecting CKD after adding tract SES to the screening approach was 67% [95%CI: 66.2, 67.2] , 61% [95%CI: 61.1, 61.5] and 5, respectively. With the standard approach, sensitivity of detecting CKD was 60% [95% CI: 59.4, 60.4], specificity was 73% [95%CI: 72.4, 72.7] and NNS was 4, respectively. Limitations One healthcare system and selection bias Conclusions Leveraging patients’ addresses from the EHR and adding tract level SES to standard screening approach modestly increases the sensitivity of detecting patients with CKD at a cost of decreased specificity. Identifying further factors that improve CKD detection at an early stage are needed to slow the progression of CKD and prevent cardiovascular complications.