Michael R Elliott's research while affiliated with Concordia University–Ann Arbor and other places

Publications (246)

Article
Full-text available
Propensity score (PS) based methods, such as matching, stratification, regression adjustment, simple and augmented inverse probability weighting, are popular for controlling for observed confounders in observational studies of causal effects. More recently, we proposed penalized spline of propensity prediction (PENCOMP), which multiply-imputes outc...
Preprint
Full-text available
A common practice in clinical trials is to evaluate a treatment effect on an intermediate endpoint when the true outcome of interest would be difficult or costly to measure. We consider how to validate intermediate endpoints in a causally-valid way when the trial outcomes are time-to-event. Using counterfactual outcomes, those that would be observe...
Article
Importance: Racial disparities in cardiometabolic health are consistently observed in cohort studies. However, most studies neither evaluate differences in age at onset nor account for systematic exclusion stemming from "weathering" (accelerated health declines for minoritized groups due to structural social and economic marginalization). Objecti...
Article
Background The prevalence of smoking is higher among sexual minority (SM) individuals compared with heterosexuals. The impact of televised anti-tobacco mass media campaigns on smoking outcomes among SMs is not known. Aims We examined whether televised anti-tobacco advertising was differentially associated with current smoking and smoking intensity...
Preprint
Longitudinal biomarker data and cross-sectional outcomes are routinely collected in modern epidemiology studies, often with the goal of informing tailored early intervention decisions. For example, hormones such as estradiol and follicle-stimulating hormone may predict changes in womens' health during the midlife. Most existing methods focus on con...
Article
Valid surrogate endpoints S can be used as a substitute for a true outcome of interest T to measure treatment efficacy in a clinical trial. We propose a causal inference approach to validate a surrogate by incorporating longitudinal measurements of the true outcomes using a mixed modeling approach, and we define models and quantities for validation...
Article
Introduction Anti-tobacco media campaigns can prevent youth smoking, but there is little research on how adult-targeted campaigns affect youth. We investigated the association between the Tips From Former Smokers (Tips) campaign and youth smoking behaviors and anti-tobacco attitudes, and variation by sex, race/ethnicity, or socioeconomic status (SE...
Article
Background: We examined associations between smoke-free laws and smoking outcomes in a nationally representative sample of US adults, including exploring whether these associations differed for heterosexual and sexual minority (SM) adults. Methods: We constructed county-level variables representing the percent of the population covered by state-...
Article
Full-text available
Importance: Despite US graduated driver licensing laws, young novice driver crash rates remain high. Study findings suggest comprehensive license policy that mandates driver education including behind-the-wheel (BTW) training may reduce crashes postlicensure. However, only 15 states mandate BTW training. Objective: To identify differences in lic...
Article
Full-text available
Motor vehicle crash rates are highest immediately after licensure, and driver error is one of the leading causes. Yet, few studies have quantified driving skills at the time of licensure, making it difficult to identify at-risk drivers before independent driving. Using data from a virtual driving assessment implemented into the licensing workflow i...
Article
Full-text available
The goal of our study is to understand the impact of Tobacco 21 (T21) laws on youth smoking and health equity. We conducted modified Poisson regression models using 2014-2019 Monitoring the Future data to measure the impact of attending school in a county 100% covered by a T21 law versus counties with <100% T21 coverage on past 30-day smoking parti...
Article
Full-text available
Background and Objectives Social support networks of older adults have been linked to their health and well-being; however, findings regarding the effects of specific network characteristics have been mixed. Additionally, due to demographic shifts increasing numbers of older adults live outside of traditional family structures. Previous studies hav...
Article
Significance This study examines the differential effects of Canadian point-of-sale (POS) tobacco display bans across provinces on quit attempts and smoking cessation, by sex, education and income. Methods We analysed survey data from five waves (waves 4–8) of the International Tobacco Control Canada Survey, a population-based, longitudinal survey...
Article
Full-text available
Research suggests that the way individuals are oriented towards the future is deeply embedded in their psychology, shaping how they perceive and react to opportunities and threats, even at unconscious levels. We argue that exposures to opportunities and threats over the life course can shape future orientation at a deep level, and that word-valence...
Article
Introduction To analyze the impact of Truth and state-sponsored anti-tobacco media campaigns on youth smoking in the U.S., and their potential to reduce tobacco-related health disparities. Methods Our study included data from the 2000-2015 Monitoring the Future study, an annual nationally representative survey of youth in 8 th (n=201,913), 10 th (...
Article
Background: There is a dearth of research on within-person relationships between tobacco price and cigarette smoking initiation and progression in young adulthood. This project examines the within-person association between cigarette pack price and smoking initiation and progression between age 18 and 21/22, focusing on differences across subgroup...
Article
A surrogate endpoint S in a clinical trial is an outcome that may be measured earlier or more easily than the true outcome of interest T. In this work, we extend causal inference approaches to validate such a surrogate using potential outcomes. The causal association paradigm assesses the relationship of the treatment effect on the surrogate with t...
Article
Introduction: This study examines whether smoke-free laws are differentially associated with youth smoking outcomes by parental education, race/ethnicity, sex, and college plans in a U.S. Sample: Methods: This study assessed the relationships between smoke-free laws in workplaces and hospitality venues (restaurants/bars) and past 30-day smokin...
Article
A persistent challenge is characterizing patterns of tobacco use in terms of product combinations and frequency. Using Wave 4 (2016–17) Population Assessment of Tobacco and Health Study adult data, we conducted latent class analyses (LCA) of past 30-day frequency of use for 9 tobacco products. One-step LCA with joint multinomial logistic regression...
Article
Background and aims: Tobacco control policies may differentially impact smoking initiation across sociodemographic groups. We measured longitudinal associations between exposure to smoke-free laws in grade 12 (modal age 18) and patterns of smoking initiation in the U.S. Design: Prospective longitudinal analysis. Setting and participants: We us...
Article
Full-text available
Background: Little is known regarding long-term impacts of anti-tobacco media campaigns on youth smoking and related disparities in the United States. Methods: We examined longitudinal cohort data from Monitoring the Future (MTF) between 2000 and 2017 in modified Poisson regression models to understand the long-term impacts of televised Truth an...
Article
Full-text available
Without randomization of treatments, valid inference of treatment effects from observational studies requires controlling for all confounders because the treated subjects generally differ systematically from the control subjects. Confounding control is commonly achieved using the propensity score, defined as the conditional probability of assignmen...
Article
Background Emerging evidence suggests many people have persistent symptoms after acute COVID-19 illness. Our objective was to estimate the prevalence and correlates of post-acute sequelae of SARS-CoV-2 infection (PASC). Methods We employed a population-based probability survey of adults with COVID-19 in Michigan. Living non-institutionalized adult...
Preprint
Full-text available
A surrogate endpoint S in a clinical trial is an outcome that may be measured earlier or more easily than the true outcome of interest T. In this work, we extend causal inference approaches to validate such a surrogate using potential outcomes. The causal association paradigm assesses the relationship of the treatment effect on the surrogate with t...
Article
With medical tests becoming increasingly available, concerns about over-testing, over-treatment and health care cost dramatically increase. Hence, it is important to understand the influence of testing on treatment selection in general practice. Most statistical methods focus on average effects of testing on treatment decisions. However, this may b...
Article
Full-text available
Little research examines how tobacco quitlines affect disparities in smoking cessation in the United States. Our study utilized data from the Tobacco Use Supplement to the Current Population Survey (2010, 2011, 2012, 2015, 2018) (TUS-CPS) and state-level quitline data from the North American Quitline Consortium and National Quitline Data Warehouse....
Preprint
Full-text available
Importance: Emerging evidence suggests many people have persistent symptoms after acute COVID-19 illness. Objective: To estimate the prevalence and correlates of persistent COVID-19 symptoms 30 and 60 days post onset using a population-based sample. Design & Setting: The Michigan COVID-19 Recovery Surveillance Study is a population-based cross-sect...
Preprint
Full-text available
Multiple imputation is a well-established general technique for analyzing data with missing values. A convenient way to implement multiple imputation is sequential regression multiple imputation (SRMI), also called chained equations multiple imputation. In this approach, we impute missing values using regression models for each variable, conditiona...
Article
Full-text available
Standard randomization-based inference conditions on the data in the population and makes inference with respect to the repeating sampling properties of the sampling indicators. In some settings these estimators can be quite unstable; Bayesian model-based approaches focus on the posterior predictive distribution of population quantities, potentiall...
Article
Purpose. Women's cardio-metabolic risk increases in midlife. Previous work relating menopause symptoms to diabetes/Metabolic Syndrome (MetS) does not consider the adverse impact of multiple concurrent physical and psychological symptoms in midlife. Methods. Data are from 3097 participants in the Study of Women's Health Across the Nation, a cohort o...
Article
Purpose To evaluate sociodemographic differences in the relationship between state and national anti-smoking media campaigns and cessation behaviors among adult smokers in the U.S. Design Repeated cross-sectional analysis. Setting U.S. nationally representative survey of adults ages 18 and older, 2001-2015. Subjects 76,278 year-ago smokers from...
Article
Full-text available
Responsive survey designs rely upon incoming data from the field data collection to optimize cost and quality tradeoffs. In order to make these decisions in real-time, survey managers rely upon monitoring tools that generate proxy indicators for cost and quality. There is a developing literature on proxy indicators for the risk of nonresponse bias....
Article
Full-text available
Increasing cigarette taxes has been the cornerstone of tobacco control policy. Recent work has argued that raising cigarette taxes alone may no longer be an effective strategy for lowering smoking rates. We largely confirm these findings but also find that increases in price continue to predict lower smoking participation in most model specificatio...
Article
Introduction The goal of the paper is to characterize the geographic and sociodemographic patterns of policies prohibiting tobacco sales to people aged <21 years (i.e., Tobacco 21) at the local, county, and state levels in the U.S. before the national law. Methods This study assessed area-level markers for region, race/ethnicity, education, povert...
Article
Purpose Limited research exists on tobacco taxes and cigarette smoking initiation and progression, particularly across different sociodemographic groups in young adulthood. This project examines how cigarette pack price in late adolescence prospectively relates to smoking initiation and progression by 21 years of age, focusing on differences across...
Article
Background: Sexual minority (SM) (gay, lesbian and bisexual) adults are at higher risk of smoking compared with heterosexual individuals, yet little is known about how the tobacco control policy landscape interacts with sexual orientation smoking disparities. Methods: We conducted a descriptive analysis to explore differential exposure to smoke-...
Preprint
With medical tests becoming increasingly available, concerns about over-testing and over-treatment dramatically increase. Hence, it is important to understand the influence of testing on treatment selection in general practice. Most statistical methods focus on average effects of testing on treatment decisions. However, this may be ill-advised, par...
Article
Background US iodine intake, estimated from the median urinary iodine concentration of population representative data, has declined by half since the 1970s which is problematic because maternal iodine intake is critical for fetal neurodevelopment. Relying on median urinary concentrations to assess iodine intake of populations is standard practice b...
Article
Objective This study examines the extent to which cigarette taxes affect smoking behaviour and disparities in smoking among adolescents by gender, socioeconomic status (SES) and race/ethnicity. Methods We used US nationally representative, repeated cross-sectional data from the 2005 to 2016 Monitoring the Future study to evaluate the relationship...
Article
This article presents results from an experimental study in Germany designed to test the effectiveness of a novel protocol for matching participants in a national panel survey with interviewers employing computer-assisted telephone interviewing (CATI) on selected sociodemographic features, including sex, age, and education. We specifically focus on...
Article
Full-text available
Body mass index (BMI) is generally used to classify adiposity. Despite the fact that the consequences of adiposity for chronic health accumulate and manifest over time, most population health research exploring the implications of high BMI measures only its recent intensity. Some studies have used retrospective measures involving maximum weight, an...
Article
Objectives: While understanding of complex within-person clustering of health behaviors into meaningful profiles of risk is growing, we still know little about whether and how U.S. adults transition from one profile to another as they age. This study assesses patterns of stability and change in profiles of tobacco and alcohol use and body mass ind...
Article
Introduction: This study sought to empirically evaluate whether the Medicaid expansion under the Affordable Care Act increased smoking cessation among low-income childless adult smokers. Methods: The effects of the Medicaid expansion on smoking quit attempts and the probability of 30- and 90-day smoking cessation were evaluated using logistic re...
Article
Latent class models have become a popular means of summarizing survey questionnaires and other large sets of categorical variables. Often these classes are of primary interest to better understand complex patterns in data. Increasingly, these latent classes are reified into predictors of other outcomes of interests, treating the most likely class a...
Article
Full-text available
Social networks are critical in maintaining late-life functional health, but, previous studies have focused on isolated dimensions of social networks. We examined whether network types, representing multiple interrelated network characteristics, are associated with mobility and disability among older adults in America. Data are from the National So...
Article
Background: The menopause has adverse effects on cardiometabolic profiles that are linked to an increased risk of atherosclerosis in women. A healthy diet during the menopausal transition may counteract the menopause-induced atherosclerotic risk. Objective: This prospective cohort study aimed to examine the associations between empirically deriv...
Article
Background and Purpose— Stroke is the leading cause of serious, long-term disability in the United States, and the number of stroke survivors is projected to rise. Physical functioning status may be compromised in survivors living in low socioeconomic status environments in comparison to higher socioeconomic status environments. Higher socioeconomi...
Article
Case studies reporting real-world experiences with survey falsification are uncommon. In this article, we document the experience of a panel survey in India that produced TV viewing estimates (“TV ratings”) where external parties were illegitimately trying to influence respondents’ behavior. The usual method to detect possible falsifications was th...
Article
Full-text available
(1) Background: Smoking restrictions have been shown to be associated with reduced smoking, but there are a number of gaps in the literature surrounding the relationship between smoke-free policies and cessation, including the extent to which this association may be modified by sociodemographic characteristics. (2) Methods: We analyzed data from th...
Article
Objective: Although reproductive function is influenced by season, few studies have evaluated seasonal effects on menopausal symptoms. We assessed the impact of season and proximity to the final menstrual period (FMP) on frequency of symptom reporting. Methods: In all, 955 participants in the Study of Women's Health Across the Nation recorded wh...
Preprint
Responsive Survey Design (RSD) aims to increase the efficiency of survey data collection via live monitoring of paradata and the introduction of protocol changes when survey errors and increased costs seem imminent. Unfortunately, RSD lacks a unifying analytical framework for standardizing its implementation across surveys. Bayesian approaches woul...
Article
Full-text available
Background: The potential impacts of beverage intake during the midlife on future subclinical atherosclerosis among women are unclear. The aim of this study was to evaluate the prospective associations between the intakes of eight beverage groups and subclinical carotid atherosclerosis in midlife women. Methods: Data came from the Study of Women...
Article
This article examines the influence of interviewers on the estimation of regression coefficients from survey data. First, we present theoretical considerations with a focus on measurement errors and nonresponse errors due to interviewers. Then, we show via simulation which of several nonresponse and measurement error scenarios has the biggest impac...
Article
Background: The All of Us Research Program is building a national longitudinal cohort and collecting data from multiple information sources (e.g., biospecimens, electronic health records, and mobile/wearable technologies) to advance precision medicine. Participant-provided information, collected via surveys, will complement and augment these infor...
Article
Full-text available
: media-1vid110.1542/5984244682001PEDS-VA_2018-2507Video Abstract BACKGROUND AND OBJECTIVE: Unrestrained child passengers are at significant risk of crash-related injury. Previous researchers using nationally representative crash data from 1992 to 1993 found an association between driver and passenger safety-belt use. Our objective in this study is...
Article
Valid causal inference from observational studies requires controlling for confounders. When time-dependent confounders are present that serve as mediators of treatment effects and affect future treatment assignment, standard regression methods for controlling for confounders fail. Similar issues also arise in trials with sequential randomization,...
Preprint
The Health and Retirement Study is a longitudinal study of US adults enrolled at age 50 and older. We were interested in investigating the effect of a sudden large decline in wealth on the cognitive score of subjects. Our analysis was complicated by the lack of randomization, confounding by indication, and a substantial fraction of the sample and p...
Article
Full-text available
Background Measures of subclinical atherosclerosis are predictors of future cardiovascular outcomes as well as of physical and cognitive functioning. The menopausal transition is associated with accelerated progression of atherosclerosis in women. The prospective association between a healthy lifestyle during the midlife and subclinical atheroscler...
Article
Declining response rates and increasing costs have led to greater use of non‐probability samples in election polling. But non‐probability samples may suffer from selection bias due to differential access, degrees of interest and other factors. Here we estimate voting preference for 19 elections in the US 2014 midterm elections by using large non‐pr...
Data
Table S1. Alternate Healthy Eating Index (AHEI) Scores in SWAN Table S2. Average HLS and Measures of Subclinical Carotid Atherosclerosis Among 1143 SWAN Participants Table S3. Long‐Term Smoking Status and Measures of Subclinical Carotid Atherosclerosis Among 1143 SWAN Participants Table S4. Average AHEI Score and Measures of Subclinical Carotid...
Article
Psychologists and other behavioral scientists are frequently interested in whether a questionnaire measures a latent construct. Attempts to address this issue are referred to as construct validation. We describe and extend nonparametric hypothesis testing procedures to assess matrix structures, which can be used for construct validation. These meth...
Chapter
Combining estimates from multiple surveys can be very useful, especially when the question of interest cannot be addressed well by a single, existing survey. In this article, we provide a brief review of the methodology for combining estimates, with a focus on dual frame, weighting‐based, joint‐modeling, missing‐data, and small‐area methods. Many s...
Article
Importance A sudden loss of wealth—a negative wealth shock—may lead to a significant mental health toll and also leave fewer monetary resources for health-related expenses. With limited years remaining to regain lost wealth in older age, the health consequences of these negative wealth shocks may be long-lasting. Objective To determine whether a n...
Article
Privacy (or the lack of it) is an important feature of the interview. Researchers rely on interviewers to ensure a private setting. In reality, interviewers are guests in respondents’ homes and might find it difficult to achieve privacy. Thus, a substantial proportion of interviews are conducted in the presence of a third party. We investigate whet...
Article
Purpose: In May 2010, New Jersey implemented the first-in-the-nation decal provision to increase intermediate drivers' compliance with Graduated Driver Licensing restrictions and ultimately reduce young driver crashes. We previously found that the provision was associated with a 9.5% decline in crash rates. This study evaluates whether the decal p...
Article
Examples of "doubly robust" estimator for missing data include augmented inverse probability weighting (AIPWT) models (Robins et al., 1994) and penalized splines of propensity prediction (PSPP) models (Zhang and Little, 2009). Doubly-robust estimators have the property that, if either the response propensity or the mean is modeled correctly, a cons...
Article
Purpose: Although motor vehicle crashes are the leading cause of death for adolescents, there is a scarcity of research addressing adolescents' lack of pre-licensure practical driving experience, which is theorized to increase their post-licensure crash risk. Methods: Utilizing police-reported crashes and survey data from a randomized and quasi-...
Article
As connected autonomous vehicles (CAVs) enter the fleet, there will be a long period when these vehicles will have to interact with human drivers. One of the challenges for CAVs is that human drivers do not communicate their decisions well. Fortunately, the kinematic behavior of a human-driven vehicle may be a good predictor of driver intent within...
Conference Paper
Statement of purpose ADHD often persists into adolescence when motor vehicle crash risk peaks. Small studies have suggested that adolescents with ADHD have more driving errors and increased crash risk. We compared monthly rates of police-reported crashes, traffic-related citations, and license suspensions over the first four years of licensure for...
Article
Two paradigms for the evaluation of surrogate markers in randomized clinical trials have been proposed: the causal effects paradigm and the causal association paradigm. Each of these paradigms rely on assumptions that must be made to proceed with estimation and to validate a candidate surrogate marker (S) for the true outcome of interest (T). We co...
Article
P-splines are penalized B-splines, in which finite order differences in coefficients are typically penalized with an $\ell_2$ norm. P-splines can be used for semiparametric regression and can include random effects to account for within-subject variability. In addition to $\ell_2$ penalties, $\ell_1$-type penalties have been used in nonparametric a...
Article
Full-text available
Background Patterns of symptom clustering in midlife women may suggest common underlying mechanisms or may identify women at risk of adverse health outcomes or, conversely, likely to experience healthy aging. This paper assesses symptom clustering in the Study of Women’s Health Across the Nation (SWAN) longitudinally by stage of reproductive aging...
Article
Importance Attention-deficit/hyperactivity disorder (ADHD) often persists into adolescence, when motor vehicle crash risk peaks. We know little about when adolescents with ADHD get licensed and, once they do, the extent to which they have increased crash risk compared with adolescents without ADHD. Objectives To examine the association between ADH...
Article
Although selecting a probability sample has been the standard for decades when making inferences from a sample to a finite population, incentives are increasing to use nonprobability samples. In a world of "big data", large amounts of data are available that are faster and easier to collect than are probability samples. Design-based inference, in w...
Article
In sample surveys, the sample units are typically chosen using a complex design. This may lead to a selection effect and, if uncorrected in the analysis, may lead to biased inferences. To mitigate the effect on inferences of deviations from a simple random sample a common technique is to use survey weights in the analysis. This article reviews appr...
Article
Background: Experiencing a negative wealth shock in late middle age may cause high levels of stress and induce reductions in health-related consumption. Methods: We used data on late middle age individuals (51-64 years) from the longitudinal US-based Health and Retirement Study (N=19 281) to examine the relationship between negative wealth shock...
Article
Objectives: Recent interest has been generated about reports of declining incidence in cognitive impairment among more recently born cohorts. At the same time, attained education, which is related to cognition, has increased in recent cohorts of older adults. We examined cohort differences in cognitive function in a nationally representative sampl...
Article
Objective: Newly licensed adolescent drivers have skill deficits that increase risk for motor vehicle crashes. Development of programs targeted to prelicensed adolescents has been hindered by concerns about encouraging overconfidence and early licensure. The study had 2 primary objectives: (a) determine whether an Internet-based intervention desig...
Article
Introduction: Graduated Driver Licensing (GDL) is the most effective strategy to reduce the burden of young driver crashes, but the extent to which young intermediate (newly licensed) drivers comply with, and police enforce, important GDL passenger and night-time restrictions is largely unknown. Population-level rates of intermediate drivers' comp...
Article
Nonresponse is very common in epidemiologic surveys and clinical trials. Common methods for dealing with missing data (e.g., complete-case analysis, ignorable-likelihood methods, and nonignorable modeling methods) rely on untestable assumptions. Nonresponse two-phase sampling (NTS), which takes a random sample of initial nonrespondents for follow-u...
Article
The development of driverless vehicles has spurred the need to predict human driving behavior to facilitate interaction between driverless and human driven vehicles. Predicting human driving movements can be challenging, and poor prediction models can lead to accidents between the driverless and human driven vehicles. We used the vehicle speed obta...
Poster
Background Several small studies suggesting teens with ADHD are at heightened crash risk were conducted among more severely affected teens in highly specialised samples but had substantial methodological limitations. Thus, we conducted the first truly longitudinal study focused on comparing crash risk between teen drivers with and without ADHD. We...
Article
Full-text available
When analyzing data sampled with unequal inclusion probabilities, correlations between the probability of selection and the sampled data can induce bias if the inclusion probabilities are ignored in the analysis. Weights equal to the inverse of the probability of inclusion are commonly used to correct possible bias. When weights are uncorrelated wi...