Search

High-quality survey data collection is getting more expensive to conduct because of decreasing response rates and rising data collection costs. Responsive and adaptive designs have emerged as a framework for targeting and reallocating resources during the data collection period to improve survey data collection efficiency. Here, we report on the implementation and evaluation of a responsive design experiment in the National Survey of College Graduates that optimizes the cost-quality tradeoff by minimizing a function of data collection costs and the root mean squared error of a key survey measure, self-reported salary. We used a Bayesian framework to incorporate prior information and generate predictions of estimated response propensity, self-reported salary, and data collection costs for use in our optimization rule. At three points during the data collection process, we implement the optimization rule and identify cases for which reduced effort would have minimal effect on the mean squared error (RMSE) of mean self-reported salary while allowing us to reduce data collection costs. We find that this optimization process allowed us to reduce data collection costs by nearly 10 percent, without a statistically or practically significant increase in the RMSE of mean salary or a decrease in the unweighted response rate. This experiment demonstrates the potential for these types of designs to more effectively target data collection resources to reach survey quality goals.

Mixed-mode designs have become increasingly common in survey data collection. Although different modes often have different measurement properties, the standard practice is to treat mixed-mode data as if they had been collected with a single mode, neglecting the potential impact of mode effects. To account for potential mode effects when making inferences for mixed-mode samples, we propose (i) a Testimator approach, (ii) a Bayesian approach, and (iii) a model averaging approach. In the Testimator approach, we test whether the means and the variances of mixed-mode samples are the same. If the means are the same, we take the average of mode-specific estimates. If the means are different, we take the average when we have no prior information about preferred modes and take the smaller (or larger) estimate when we have prior information about preferred modes (e.g., a smaller estimate is better). In the Bayesian approach, we assume some prior information. We use a data-driven method to determine whether there are mode effects. If there are no mode effects, we draw inferences using a common mean model. If there are mode effects, we draw inferences using the data collected with the mode that produces smaller estimates. In the model averaging approach, we combine estimates of different models (characterized by whether assume same means and variances across modes) using marginal posteriors as weights. We evaluate the approaches in simulation studies and find that they achieve robust inferences compared to the standard approach. We apply the methods to the Arab Barometer study, which employs a randomized mixed-mode design.

Introduction : Racial/ethnic discrimination (hereafter, discrimination) is associated with tobacco use. However, little is known about the relationship between discrimination and dual/polytobacco use and tobacco use disorder (TUD), including how these relationships vary by race/ethnicity. Methods: Data on adults 18 and older come from the 2012–2013 National Epidemiologic Survey on Alcohol and Related Conditions-III (n= 35,881). Past-year discrimination was measured using the Experiences of Discrimination scale. Past 30-day exclusive, dual, and polytobacco use was measured as the mutually exclusive use of any combination of four types of tobacco products: cigarettes, electronic nicotine delivery systems, other combustibles (i.e., cigars and pipe), and smokeless tobacco. Past-year TUD was defined according to DSM-5 criteria. Associations between discrimination and exclusive, dual, and polytobacco use and discrimination and TUD were estimated using multinomial logistic regression and logistic regression, respectively. Models were stratified by race/ethnicity (i.e., Hispanic, non-Hispanic (NH) White, NH Black, another race/ethnicity) to assess effect modification. Results: Adults who used tobacco and who had TUD was 24.2% and 19.2%, respectively. More discrimination was associated with higher odds of exclusive, dual, and polytobacco use as well as TUD. Models stratified by race/ethnicity suggest that discrimination was associated with dual/polytobacco use among NH Black adults (OR: 1.05, 95% CI: 1.002–1.11) and NH White adults (OR: 1.18, 95% CI: 1.13–1.22). While more discrimination was associated with TUD among all racial/ethnic groups, the relationship was the strongest for NH White adults. Conclusions: Discrimination was associated with more severe tobacco use outcomes among multiple racial/ethnic groups, but associations were the strongest for NH White adults.

Surveys estimate and monitor a variety of data collection parameters, including response propensity, number of contacts, and data collection costs. These parameters can be used as inputs to a responsive/adaptive design or to monitor the progression of a data collection period against predefined expectations. Recently, Bayesian methods have emerged as a method for combining historical information or external data with data from the in-progress data collection period to improve prediction. We develop a Bayesian method for predicting a measure of case-level progress or productivity, the estimated time lag, in days, between first contact attempt and first respondent contact. We compare the quality of predictions from the Bayesian method to predictions generated from more commonly-used predictive methods that leverage data from only historical data collection periods or the in-progress round of data collection. Using prediction error and misclassification as short- or long- day lags, we demonstrate that the Bayesian method results in improved predictions close to the day of the first contact attempt, when these predictions may be most informative for interventions or interviewer feedback. This application adds to evidence that combining historical and current information about data collection, in a Bayesian framework, can improve predictions of data collection parameters.

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. Daily predictions of response propensity for all active sampled cases are among the most important quantities for live monitoring of data collection outcomes, making sound predictions of these propensities essential for the success of RSD. Because it relies on real-time updates of prior beliefs about key design quantities, such as predicted response propensities, RSD stands to benefit from Bayesian approaches. However, empirical evidence of the merits of these approaches is lacking in the literature, and the derivation of informative prior distributions is required for these approaches to be effective. In this paper, we evaluate the ability of two approaches to deriving prior distributions for the coefficients defining daily response propensity models to improve predictions of daily response propensity in a real data collection employing RSD. The first approach involves analyses of historical data from the same survey, and the second approach involves literature review. We find that Bayesian methods based on these two approaches result in higher-quality predictions of response propensity than more standard approaches ignoring prior information. This is especially true during the early-to-middle periods of data collection, when survey managers using RSD often consider interventions.

Introduction: Mental health problems in emerging adulthood are linked to tobacco and cannabis use, but whether race and ethnicity modifies these associations is unclear. Methods: We used data from wave 4 of the Population Assessment of Tobacco and Health Study (youth n= 6898, young adult n= 10,304) to conduct latent class analysis (LCA) of six past 30-day tobacco and cannabis use indicators (i.e., cigarettes, electronic nicotine delivery systems (ENDS), cigars, blunts, cannabis vaping, other cannabis). We estimated associations between past 30-day internalizing and externalizing (i.e., low (referent), moderate, high) problems and latent classes of tobacco/cannabis use (vs. never/former use) using adjusted multinomial logistic regression. We explored whether associations varied by race and ethnicity through stratification. Results: We identified four exclusive use latent classes and two dual/poly use latent classes for both youth and young adult samples. Race/ethnicity-stratified models identified associations between internalizing/externalizing problems and most use classes for Hispanic and non-Hispanic White youth/young adults, with mixed results for non-Hispanic Black youth/young adults. For example, Hispanic (OR: 2.50, 95% CI: 1.09–5.74) and non-Hispanic White (OR: 1.90, 95% CI: 1.18–3.06) youth with high internalizing problems had higher odds of ENDS + cannabis vaping. Externalizing problems were not associated with use among non-Hispanic Black youth while internalizing problems were not associated with use among non-Hispanic Black young adults. Conclusion: We observed racial/ethnic variation in mental health problems and tobacco and cannabis use. Understanding mental health problem and tobacco product and cannabis use comorbidity may better inform culturally relevant interventions aimed to prevent and reduce use.

SignificanceThis 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.MethodsWe analysed survey data from five waves (waves 4–8) of the International Tobacco Control Canada Survey, a population-based, longitudinal survey, where provinces implemented display bans between 2004 and 2010. Primary outcomes were quit attempts and successful cessation. We used generalised estimating equation Poisson regression models to estimate associations between living in a province with or without a POS ban (with a 24-month threshold) and smoking outcomes. We tested whether these associations varied by sex, education and income by including interaction terms.ResultsAcross survey waves, the percentage of participants in provinces with POS bans established for more than 24 months increased from 5.0% to 95.8%. There was no association between POS bans and quit attempts for provinces with bans in place for 0–24 months or more than 24 months, respectively (adjusted relative risk (aRR)=0.99, 95% CI: 0.89 to 1.10; 1.03, 95% CI: 0.88 to 1.20). However, we found a differential impact of POS bans on quit attempts by sex, whereby bans were more effective for women than men for bans of 0–24 months. Participants living in a province with a POS ban for at least 24 months had a higher chance of successful cessation (aRR=1.49; 95% CI: 1.08 to 2.05) compared with those in a province without a ban. We found no differences in the association between POS bans and quit attempts or cessation by education or income, and no differences by sex for cessation.ConclusionPOS bans are associated with increased smoking cessation overall and more quit attempts among women than men.

This article examines the driving records of young Michigan drivers in order to predict future offenses and accidents. Results indicate that ticketed violations and at-fault accidents significantly increase the likelihood of subsequent offenses and accidents, especially in female and more experienced young drivers.

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 the 2001-2015 Tobacco Use Supplement to the Current Population Survey.Measures: Area-level exposure to State-sponsored and “Tips from former smokers” anti-tobacco media campaigns was the primary predictor of this study. Outcome variables included: quit attempt in the past 12 months, past 30-day smoking cessation, and past 90-day smoking cessation among year-ago smokers.Analysis: We conducted modified Poisson regression models to examine the association between media campaign exposure and cessation behaviors. We also examined effect modification on the additive scale by sex, race/ethnicity, income, and education using average marginal effects.Results: Year-ago smokers with greater exposure to media campaigns were more likely to report 30-day (Prevalence Ratio [PR]: 1.18, CI: 1.03, 1.36) and 90-day cessation (PR: 1.18, CI: 1.00, 1.41) compared to respondents with less campaign exposure. We found no evidence of effect modification by sociodemographic variables.Conclusion: Exposure to anti-smoking media campaigns were associated with year-ago smokers’ cessation behaviors. However, there were no differences in the association by sex, race/ethnicity, income, or education, indicating that broadly focused media campaigns may be insufficient to reduce smoking cessation among priority populations, and thus health disparities generally.

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, potentially providing a better balance between bias correction and efficiency. Previous work in this area has focused on estimation of means and linear and generalized linear regression parameters; these methods do not allow for a general estimation of distributional functions such as quantile or quantile regression parameters. Here we adapt an extended Dirichlet Process Mixture model that allows the DP prior to be a mixture of DP random basis measures that are a function of covariates. These models allow many mixture components when necessary to accommodate the sample design, but can shrink to few components for more efficient estimation when the data allow. We provide an application to the estimation of relationships between serum dioxin levels and age in the US population, either at the mean level (via linear regression) or across the dioxin distribution (via quantile regression) using the National Health and Nutrition Examination Survey.

BackgroundSexual 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.MethodsWe conducted a descriptive analysis to explore differential exposure to smoke-free laws and televised anti-tobacco media using two sources of national data from the United States: Census data on same-sex couple households/all households and data on SM adults/all adults from the National Health Interview Survey (NHIS). We combined this information with variables representing the proportion of individuals in each county covered by smoke-free laws (2013–2017), and average county-level exposure to televised anti-tobacco media campaigns (2013–2015). We compared average coverage levels for SM populations to average coverage levels for the broader US population.ResultsBetween 2013 and 2017, same-sex couple households/SM adults lived in counties with higher levels of smoke-free law coverage compared with all US households/adults for workplaces (Census: 71.3% vs 68.0%; NHIS: 70.7% vs 67.9%) and hospitality venues (Census: 82.3% vs 77.0%; NHIS: 80.5% vs 77.2%). There were no consistent differences in exposures to anti-tobacco media campaigns across datasets.ConclusionsSM adults may be more likely to live in areas with smoke-free laws, compared with the general population. Findings point to the need to examine other potential drivers of smoking in SM populations.

ObjectiveThis 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.MethodsWe used US nationally representative, repeated cross-sectional data from the 2005 to 2016 Monitoring the Future study to evaluate the relationship between state cigarette taxes and past 30-day current smoking, smoking intensity, and first cigarette and daily smoking initiation using modified Poisson and linear regression models, stratified by grade. We tested for interactions between tax and gender, SES and race/ethnicity on the additive scale using average marginal effects.ResultsWe found that higher taxes were associated with lower smoking outcomes, with variation by grade. Across nearly all of our specifications, there were no statistically significant interactions between tax and gender, SES or race/ethnicity for any grades/outcomes. One exception is that among 12th graders, there was a statistically significant interaction between tax and college plans, with taxes being associated with a lower probability of 30-day smoking among students who definitely planned to attend college compared with those who did not.ConclusionWe conclude that higher taxes were associated with reduced smoking among adolescents, with little difference by gender, SES and racial/ethnicity groups. While effective at reducing adolescent smoking, taxes appear unlikely to reduce smoking disparities among youth.

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. However, there is very little research on proxy indicators for costs and almost none aimed at predicting costs under alternative design strategies. Predictions of survey costs and proxy error indicators can be used to optimize survey designs in real time. Using data from the National Survey of Family Growth, we evaluate alternative modeling strategies aimed at predicting survey costs (specifically, interviewer hours). The models include multilevel regression (with random interviewer effects) and Bayesian Additive Regression Trees (BART).

Macrophage antibody (Ab)-dependent cellular phagocytosis (ADCP) is a major cytotoxic mechanism for both therapeutic unconjugated monoclonal Abs (mAbs) such as rituximab and Ab-induced hemolytic anemia and immune thrombocytopenia. Here, we studied the mechanisms controlling the rate and capacity of macrophages to carry out ADCP in settings of high target/effector cell ratios, such as those seen in patients with circulating tumor burden in leukemic phase disease. Using quantitative live-cell imaging of primary human and mouse macrophages, we found that, upon initial challenge with mAb-opsonized lymphocytes, macrophages underwent a brief burst (<1 hour) of rapid phagocytosis, which was then invariably followed by a sharp reduction in phagocytic activity that could persist for days. This previously unknown refractory period of ADCP, or hypophagia, was observed in all macrophage, mAb, and target cell conditions tested in vitro and was also seen in vivo in Kupffer cells from mice induced to undergo successive rounds of αCD20 mAb-dependent clearance of circulating B cells. Importantly, hypophagia had no effect on Ab-independent phagocytosis and did not alter macrophage viability. In mechanistic studies, we found that the rapid loss of activating Fc receptors from the surface and their subsequent proteolytic degradation were the primary mechanisms responsible for the loss of ADCP activity in hypophagia. These data suggest hypophagia is a critical limiting step in macrophage-mediated clearance of cells via ADCP, and understanding such limitations to innate immune system cytotoxic capacity will aid in the development of mAb regimens that could optimize ADCP and improve patient outcome.

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 the ability of the protocol to engender close matches between respondents and interviewers in terms of these features, using both theory and empirical evidence to suggest that this type of matching will improve cooperation rates in surveys employing CATI. We also focus on indicators of “success” at first contact (defined as a successful interview or establishment of an appointment for an interview) as a function of whether the matching protocol was in use on a given day and whether specific types of matches generated higher rates of success overall. We find strong evidence of the protocol effectively establishing close matches, and we also observe that matches based on education proved especially effective for rates of “success” in a panel survey that focused primarily on labor market topics. We conclude with thoughts on practical implementation of this approach in other settings and suggested directions for future work in this area.

Supplemental Digital Content is available in the text.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 as the true class to which an individual belongs even though there is uncertainty in class membership. This uncertainty can be viewed as a form of measurement error in predictors, leading to bias in the estimates of the regression parameters associated with the latent classes. Despite this fact, there is very limited literature treating latent class predictors as measurement error models. Most applications ignore this issue and fit a two-stage model that treats the modal class prediction as truth. Here, we develop two approaches—one likelihood-based, the other Bayesian—to implement a joint model for latent class analysis and outcome prediction. We apply these methods to an analysis of how acculturation behaviors predict depression in South Asian immigrants to the United States. A simulation study gives guidance for when a two-stage model can be safely implemented and when the joint model may be required.

Big Data are a “big challenge” for finite population inference. Lack of control over data-generating processes by researchers in the absence of a known random selection mechanism may lead to biased estimates. Further, larger sample sizes increase the relative contribution of selection bias to squared or absolute error. One approach to mitigate this issue is to treat Big Data as a random sample and estimate the pseudo-inclusion probabilities through a benchmark survey with a set of relevant auxiliary variables common to the Big Data. Since the true propensity model is usually unknown, and Big Data tend to be poor in such variables that fully govern the selection mechanism, the use of flexible non-parametric models seems to be essential. Traditionally, a weighted logistic model is recommended to account for the sampling weights in the benchmark survey when estimating the propensity scores. However, handling weights is a hurdle when seeking a broader range of predictive methods. To further protect against model misspecification, we propose using an alternative pseudo-weighting approach that allows us to fit more flexible modern predictive tools such as Bayesian Additive Regression Trees (BART), which automatically detect non-linear associations as well as high-order interactions. In addition, the posterior predictive distribution generated by BART makes it easier to quantify the uncertainty due to pseudo-weighting. Our simulation findings reveal further reduction in bias by our approach compared with conventional propensity adjustment method when the true model is unknown. Finally, we apply our method to the naturalistic driving data from the Safety Pilot Model Deployment using the National Household Travel Survey as a benchmark.

Introduction: We have recently shown that high frequency (2 x week) low dose (50 mg)(HFLD) rituximab (RTX) and acalabrutinib is highly effective initial therapy for patients (n=38) with progressive CLL (ClinicalTrials.gov NCT03788291, PMID:36689726). The first dose of 50 mg of IV rituximab administered at 25 mg/h decreased the median circulating CLL cell count by 84% from pre-treatment baseline at 1h with no further decrease in the count during the remainder of the infusion (PMID:37003030). During the first hour of infusion, median CLL cell membrane CD20 levels decreased to 65% of baseline and were 41% of baseline at the end of the infusion. Median serum rituximab concentration was 2 mcg/ml at 1h and 10 mcg/ml at the end of the infusion. Median complement (CH50) concentration decreased to 91% of baseline at 1 h and to 81% of baseline at the end of the infusion. The cessation of clearance of CLL cells from the circulation within 1h of starting rituximab therapy is thus unlikely to be caused by low rituximab levels, decreased CLL cell CD20 expression or low serum complement levels. We and others have previously shown that ADCP is the primary mechanism of clearance of anti-CD20 mAb opsonized B cells in both humans and mice. We hypothesized that clearance of circulating CLL cells decreases because residual circulating CLL cells are resistant to mAb induced ADCP. To test this hypothesis, we measured in vitro ADCP sensitivity of CLL cells sampled from patients before and after rituximab treatment.

In the bone marrow, tissue-resident macrophages serve several specific roles including engulfing senescent neutrophils, maintaining hematopoietic stem cells (HSCs), and comprising an important component of the erythroid microenvironment within erythroblastic islands (EBIs). These latter macrophages are thought to “nurture” maturing erythroblasts with iron and cytokines, and to serve as a sink for erythroblast mitochondria, while simultaneously phagocytosing newly formed pyrenocytes. Here we sought to delineate the developmental origin of marrow macrophages, as well as their cellular heterogeneity within the complex cellular milieu of the marrow. Macrophages form an extensive cellular network extending long processes throughout the marrow, which results in cellular injury during dissociation and the presence of macrophage remnants on diverse hematopoietic cell types. We visualized macrophage remnants utilizing imaging flow cytometry to optimize conditions for marrow dissociation and to develop immunophenotyping that removes contaminants to enrich for bona fide macrophages. Importantly, in vitro coculture of marrow macrophages with late-stage erythroid progenitors significantly expanded erythroid output >1.5-fold, consistent with a role of EBI-macrophages in erythroblast maturation.

A comparative analysis of risk of child passenger injury during nighttime and daytime driving and teenage and adult driving is presented.

Supplemental Digital Content is available in the text.

Rheumatoid arthritis is characterized by progressive joint inflammation and affects ~1% of the human population. We noted single-nucleotide polymorphisms (SNPs) in the apoptotic cell–engulfment genes ELMO1, DOCK2, and RAC1linked to rheumatoid arthritis. As ELMO1 promotes cytoskeletal reorganization during engulfment, we hypothesized that ELMO1 loss would worsen inflammatory arthritis. Surprisingly, Elmo1-deficient mice showed reduced joint inflammation in acute and chronic arthritis models. Genetic and cell-biology studies revealed that ELMO1 associates with receptors linked to neutrophil function in arthritis and regulates activation and early neutrophil recruitment to the joints, without general inhibition of inflammatory responses. Further, neutrophils from the peripheral blood of human donors that carry the SNP in ELMO1associated with arthritis display increased migratory capacity, whereas ELMO1knockdown reduces human neutrophil migration to chemokines linked to arthritis. These data identify ‘noncanonical’ roles for ELMO1 as an important cytoplasmic regulator of specific neutrophil receptors and promoter of arthritis.

Toxoplasma gondiiis a common protozoan parasite that infects up to one third of the world’s population. Notably, very little is known about innate immune sensing mechanisms for this obligate intracellular parasite by human cells. Here, by applying an unbiased biochemical screening approach, we show that human monocytes recognized the presence of T. gondiiinfection by detecting the alarmin S100A11 protein, which is released from parasite-infected cells via caspase-1-dependent mechanisms. S100A11 induced a potent chemokine response to T. gondiiby engaging its receptor RAGE, and regulated monocyte recruitment in vivo by inducing expression of the chemokine CCL2. Our experiments reveal a sensing system for T. gondiiby human cells that is based on the detection of infection-mediated release of S100A11 and RAGE-dependent induction of CCL2, a crucial chemokine required for host resistance to the parasite.

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 ADHD and both driver licensing and crash involvement and whether it varies by sex, licensing age, and/or being prescribed ADHD medication at licensure. DESIGN, SETTING, AND PARTICIPANTS: This retrospective cohort study was conducted at 6 primary care practices of the Children’s Hospital of Philadelphia, a large pediatric health care network in southeastern Pennsylvania and southern New Jersey. Using electronic health records, we defined a cohort of 2479 adolescents and young adults with ADHD and 15 865 without ADHD who were (1) born from 1987 to 1997; (2) residents of New Jersey and patients at 1 of 6 New Jersey primary care practices at age 12 years or older; and (3) age-eligible to obtain a driver’s license from 2004 through 2014. Electronic health records data were then linked with New Jersey’s statewide driver licensing and crash databases for 2004 through 2014. MAIN OUTCOMES AND MEASURES: Acquisition of a driver’s license and first involvement as a driver in a police-reported crash. Survival analysis was used to estimate adjusted hazard ratios for licensing and crash outcomes through age 25 years. RESULTS: The median age of individuals at the end of the study was 22.2 years (interquartile range, 19.7-24.8). Compared with individuals without ADHD, the licensing probability of individuals with ADHD 6 months after eligibility was 35% lower (for males: adjusted hazard ratio, 0.65; 95% CI, 0.61-0.70; females: adjusted hazard ratio, 0.64; 95% CI, 0.58-0.70). Among individuals with a driver’s license, 764 of 1785 with ADHD (42.8%) and 4715 of 13 221 without ADHD (35.7%) crashed during the study period. The adjusted risk for first crash among licensed drivers with ADHD was 1.36 times higher than for those without ADHD (95% CI, 1.25-1.48) and did not vary by sex, licensing age, or over time. Only 129 individuals with ADHD (12.1%) were prescribed medication in the 30 days before licensure. CONCLUSIONS AND RELEVANCE: Adolescents with ADHD get licensed less often and at an older age. Once licensed, this cohort has a greater risk of crashing. Additional research is needed to understand the specific mechanisms by which ADHD influences crash risk.

In civil airspace, the methods needed to produce software compliant with airworthiness have been considered overly burdensome, expensive and process-heavy. This view is based largely on experience with DO-178B/ED12B[5], which was the standard for software development in civil airspace beginning in 1992. Finalized in 2011, DO-178C/ED12C[6] was created to address these concerns and to apply more modern software practice to issues of software production and verification. Of special interest is DO-332/ED-217[7], the Object-oriented Technology and Related Techniques supplement to DO-178C/ED-12C, which addresses modern software practice and how it can be utilized in the production of software in systems that need airworthiness certification. This article addresses the traditional view of software development in this area and the significant cost and schedule reduction which can be realized by using the guidance and recommendations described in DO-332/ED-217 over those practices based on DO-178B/ED-12B. [ABSTRACT FROM AUTHOR]

The phagocytosis of apoptotic cells (efferocytosis) shifts macrophages to an anti-inflammatory state through a set of still poorly understood soluble and cell-bound signals. Apoptosis is a common feature of inflamed tissues, and efferocytosis by tissue macrophages is thought to promote the resolution of inflammation. However, it is not clear how the exposure of tissue macrophages to inflammatory cues (e.g., PAMPs, DAMPs) in the early stages of inflammation affects immune outcomes of macrophage-apoptotic cell interactions occurring at later stages of inflammation. To address this, we used low-dose endotoxin conditioning (LEC, 1 ng/ml LPS 18 h) of mouse resident peritoneal macrophages (RPMФ) to model the effects of suboptimal (i.e., non-tolerizing), antecedent TLR activation on macrophage inflammatory responses to apoptotic cells. Compared with unconditioned macrophages (MФ), LEC-MФ showed a significant enhancement of apoptotic cell-driven suppression of many inflammatory cytokines (e.g., TNF, MIP-1β, MCP-1). We then found that enzymatic depletion of adenosine or inhibition of the adenosine receptor A2a on LEC-MФ abrogated apoptotic cell suppression of TNF, and this suppression was entirely dependent on the ecto-enzyme CD73 (AMPadenosine) but not CD39 (ATPAMP), both of which are highly expressed on RPMФ. In addition to a requirement for CD73, we also show that Adora2a levels in macrophages are a critical determinant of TNF suppression by apoptotic cells. LEC treatment of RPMФ led to a ~3-fold increase in Adora2a and a ~28-fold increase in adenosine sensitivity. Moreover, in RAW264.7 cells, ectopic expression of both A2a and CD73 was required for TNF suppression by apoptotic cells. In mice, mild, TLR4-dependent inflammation in the lungs and peritoneum caused a rapid increase in macrophage Adora2a and Adora2b levels, and CD73 was required to limit neutrophil influx in this peritonitis model. Thus immune signaling via the CD73–A2a axis in macrophages links early inflammatory events to subsequent immune responses to apoptotic cells.

BackgroundExperiencing a negative wealth shock in late middle age may cause high levels of stress and induce reductions in health-related consumption.MethodsWe 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 and short-term outcomes that serve as markers of the pathways from wealth shock to health: elevated depressive symptoms, as a marker of the stress pathway and cost-related medication non-adherence (CRN), as a marker of the consumption pathway. Negative wealth shock was considered to be a loss of total net worth of 75% or more.ResultsUsing a nested cross-over approach—a within-person design among exposed individuals only that adjusts by design for all time-invariant individual characteristics—we found that negative wealth shock was significantly associated with increased odds of elevated depressive symptoms (OR=1.50, CI 1.10 to 2.05), but was not significantly associated with higher odds of CRN (OR=1.18, CI 0.76 to 1.82), even after further adjustment for time-varying sociodemographic and health covariates.ConclusionsNegative wealth shock during late middle age confers an increased risk of elevated depressive symptoms, but does not change levels of CRN. Personal and policy factors that may buffer the mental health risks of negative wealth shock, such as social support and social welfare policy, should be considered.

Racial/ethnic discrimination (hereafter, discrimination) is associated with use of individual tobacco and cannabis products. However, we know little about how discrimination affects dual/polytobacco and cannabis use and associated use disorders.

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-up data collection, provides a means to reduce nonresponse bias. However, traditional weighting methods to analyze data from NTS do not make full use of auxiliary variables. This article proposes a method called nonrespondent subsample multiple imputation (NSMI), where multiple imputation (Rubin 1987) is performed within the subsample of nonrespondents in Phase I using additional data collected in Phase II. The properties of the proposed methods by simulation are illustrated and the methods applied to a quality of life study. The simulation study shows that the gains from using the NTS scheme can be substantial, even if NTS sampling only collects data from a small proportion of the initial nonrespondents.

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 with the descriptive or model estimators of interest, highly disproportional sample designs resulting in large weights can introduce unnecessary variability, leading to an overall larger mean square error compared to unweighted methods.

Multiple imputation (MI) is commonly used when item-level missing data are present. However, MI requires that survey design information be built into the imputation models. For multistage stratified clustered designs, this requires dummy variables to represent strata as well as primary sampling units (PSUs) nested within each stratum in the imputation model. Such a modeling strategy is not only operationally burdensome but also inferentially inefficient when there are many strata in the sample design. Complexity only increases when sampling weights need to be modeled. This article develops a generalpurpose analytic strategy for population inference from complex sample designs with item-level missingness. In a simulation study, the proposed procedures demonstrate efficient estimation and good coverage properties. We also consider an application to accommodate missing body mass index (BMI) data in the analysis of BMI percentiles using National Health and Nutrition Examination Survey (NHANES) III data. We argue that the proposed methods offer an easy-to-implement solution to problems that are not well-handled by current MI techniques. Note that, while the proposed method borrows from the MI framework to develop its inferential methods, it is not designed as an alternative strategy to release multiply imputed datasets for complex sample design data, but rather as an analytic strategy in and of itself.

Privacy is an important feature of the interview interaction mainly due to its potential effect on reporting information, especially sensitive information. Here we examine the effect of third-party presence on reporting both sensitive and relatively neutral outcomes. We investigate whether the effect of third-party presence on reporting sensitive information is moderated by the respondent’s need for social conformity and the respondent’s country of residence. Three types of outcomes are investigated: behavioral, attitudinal, and relatively neutral health events. Using data from 22,070 interviews and nine countries in the cross-national World Mental Health Survey Initiative, we fit multilevel logistic regression to study reporting effects on questions about suicidal behavior and marital ratings, and contrast these with questions about having high blood pressure, asthma, or arthritis. We find that there is an effect of third-party presence on reporting sensitive information and no effect on reporting of neutral information. Further, the effect of the interview privacy setting on reporting sensitive information is moderated by the need for social conformity and the cultural setting.

The article presents an interview with American statistician Rod Little, who discusses motivation behind studying statistics, working at the World Fertility Survey and important factors for success in securing research grants.

Few studies have evaluated factors that influence menstrual cycle length (MCL) during the menopausal transition (MT), a life stage during which very long cycles become more likely to occur. The objective of this article was to assess how body mass index and raceethnicity—factors associated with MCL in young women—influence MCL during the MT.

IMPORTANCE: Many studies have failed to show an effect of parent-supervised practice driving on the driving performance of teenagers; nevertheless, most Graduated Driver Licensing programs have provisions that require supervised practice. OBJECTIVE: To determine whether a web-based intervention, the Teen Driving Plan (TDP), can improve the driving performance of teenagers before licensure as measured by the Teen On-road Driving Assessment (tODA). DESIGN, SETTING, AND PARTICIPANTS: Randomized, single-blind, clinical trial among 217 dyads (1 parent: 1 teenaged learner’s-permit holder) to test TDP effectiveness on increasing the quantity and diversity of supervised practice and improving the teenagers’ prelicensed driving performance. The study was conducted from December 2011 through January 2013 in Southeastern Pennsylvania. INTERVENTIONS: Dyads were randomized (3:2) to receive the TDP or the Pennsylvania driver’s manual (control group). The TDP is a psychoeducational intervention designed to increase the quantity and diversity of parent-supervised practice. Materials are grouped by the following driving environments: empty parking lots, suburban residential streets, intermediate (1- or 2-lane) roads, highways, rural roads with curves and elevation changes, and commercial districts. MAIN OUTCOMES AND MEASURES: The main outcomes were self-reported practice driving across 6 environments and 2 conditions and driving performance as measured by the teenagers’ completion of the standardized and validated tODA 24 weeks after enrollment. Certified professional driving evaluators blinded to randomization status terminated the tODA if they determined that the teenager could not safely complete it. We examined mean differences in the quantity of supervised practice, differences in the overall proportion of teenagers in each group that had assessments terminated for unsafe driving, and the point of termination during the assessment. RESULTS: The TDP dyads reported more practice in 5 of the 6 environments and at night and in bad weather compared with the control dyads. Overall, 5 of 86 TDP teenagers (6%) had the tODA terminated compared with 10 of 65 control teenagers (15%) (risk difference [TDP − control], −9% [95% CI, −21% to 2%]; P = .06). The hazard ratio for exposure to TDP was 0.35 (95% CI, 0.12-1.03; P = .05, log-rank test). CONCLUSIONS AND RELEVANCE: Preliminary evidence suggests that the TDP improves supervised practice and the driving performance of prelicensed teenaged drivers. Future studies can explore how to revise the TDP to enhance the treatment effect and how best to disseminate the TDP without compromising implementation fidelity. TRIAL REGISTRATION: clinicaltrials.gov Identifier: NCT01498575

The Moving Ahead for Progress in the 21st Century Act requires a new emphasis on measurement of serious injuries in crashes. Current practice in states that measure serious injuries is to use the police-reported injury severity of A (suspected serious injury) to define serious injury. However, serious injury is more appropriately defined as a medical diagnosis. The biggest challenge to using a medical diagnosis as the definition of serious injury is that medical outcome data are typically not part of a state crash data set. The most comprehensive solution to this challenge is to link state crash databases with state trauma registries. However, for most states, data linkage is still many years away. This paper discusses sampling medical records as an interim solution. Sampling approaches are discussed, and a stratified sampling approach is recommended. Optimal allocation of a sample provides a framework for determining the proportion of cases to sample from each subgroup, and the desired precision of the estimate and the cost per sample determine the number of cases that should be sampled. A roadmap for developing a stratified sampling plan is presented as is a hypothetical numerical example that uses data from Michigan. Sampling as an interim solution has several advantages, including comparability, progress toward comprehensive linkage, and scalability. Logistical challenges, costs of sampling, and privacy considerations are also discussed. Although sampling is not free, it represents an important way to move ahead with the necessary step of enabling measurement of serious injuries by using a diagnosis-based definition.

This study aims to assess the agreement between the menopausal transition stages defined by annual interviews or annual follicle-stimulating hormone levels and the menopausal transition stages defined by monthly menstrual calendars, as well as factors associated with discordance.

The goal of the present study was to quantify the population-based background serum concentrations of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) by using data from the reference population of the 2005 University of Michigan Dioxin Exposure Study (UMDES) and the 2003-2004 National Health and Nutrition Examination Survey (NHANES).

Background: We conducted a population‐based case–control study to better delineate the relationship between individual alcohol consumption, alcohol outlets in the surrounding environment, and being assaulted with a gun.

Passenger compartment intrusion, loss of integrity of the vehicle occupant compartment due to a motor vehicle crash, has frequently been used as a triage criterion. Data to support intrusion as a proxy for injury severity in child occupants are lacking. The purpose of this study was to examine the association between intrusion and injury to children in motor vehicle crashes.

The article discusses the collaboration of a clinical science and statistics in the Healthy for Life project of the University of Pennsylvania and Children's Hospital of Philadelphia in Philadelphia, Pennsylvania. According to the author, two problems, which are the missing fraction of height data and implausibility in weights and heights for the age, are encountered in the project. Furthermore, these issues, along with problems on childhood obesity, are solved using the imputation model.

Background: Alcohol consumption and drink/driving are positively correlated and many predictors of alcohol use also predict drink/driving. Past research has not fully distinguished the contributions of personal risk factors from the level of alcohol use in the prediction of drink/driving. As a result, the extent to which predictors are specific to drink/driving, versus due to a mutual association to alcohol use, is unclear.