Search

Attention to sexual misconduct has focused on acquaintance rape, leaving a need for research on less highly recognizable forms of harm. We estimated institution of higher education (IHE)-specific prevalence of yellow zone sexual harassment (SH) among students at 27 IHEs. We then examined SH and perceived risk of sexual assault/misconduct, knowledge regarding policies/resources, and perceptions of sexual misconduct response. Between 37.1% and 55.7% of students experienced SH. Harassed students were much more likely than non-harassed students to feel at risk for sexual misconduct and to have negative views of sexual misconduct response. Implications for research, policy, and prevention/response are discussed.

Quantitative methods are increasingly being used in domestic violence and abuse (DVA) settings to build evidence that can affect meaningful change. Ideally resulting in processes that are reproducible and results that can be comparable, quantitative methods are highly valued by many stakeholders, making them particularly useful to inform DVA practice, policy, and research. In essence, quantitative methods produce numerical estimates. When done well, these estimates provide insightful, precise, and valid scientific conclusions. This chapter provides an overview of common quantitative research approaches and methods, with salient examples interspersed throughout. The focus begins on important underlying principles, such as statistical significance, typologies of analysis, basic statistics, types of variables, association of variables, and levels of measurement. From there, the discussion proceeds to an overview of commonly used univariate and bivariate statistical methods. Next, more various advanced multivariable modelling methods such as regression analysis, multilevel regression analysis, factor analysis, and structural equation modelling are covered. Finally, limitations of quantitative methods are reviewed to provide important context for those conducting and interpreting the results of such analyses. Overall, readers will find this chapter to be an easy-to-digest and non-technical reference. For those who are somewhat familiar with quantitative research, the chapter will serve as a useful refresher. Moreover, for those new to quantitative methods, this chapter will provide a starting point for further exploration. In all, the chapter expertly weaves together a multitude of useful content for many stakeholders, and will hopefully inform future DVA-focused work.

One goal of university campus sexual assault (CSA) policies is to help prevent CSA. Federal guidance in the 2014 White House Task Force to Protect Students From Sexual Assault Checklist for Campus Sexual Misconduct Policies suggests 10 elements for inclusion in CSA policies (e.g., Policy Introduction, Grievance/Adjudication), and outlines policy topics to be included within each element (Policy Introduction includes two topics: statement of prohibition against sex discrimination including sexual misconduct and statement of commitment to address sexual misconduct). However, no research has examined whether CSA policies impact CSA prevalence. To begin addressing this gap, we studied 24 universities participating in the 2015 Association of American Universities Campus Climate Survey on Sexual Assault and Sexual Misconduct. We linked 2014-2015 data from these universities’ CSA policies and their CSA prevalence findings from the 2015 Association of American Universities (AAU) survey. To test whether the comprehensiveness of schools’ CSA policies was related to schools’ CSA prevalence, we examined the degree to which the CSA policies included recommended policy content from the aforementioned Checklist. Policies were characterized as more comprehensive if they included greater numbers of Checklist topics. We then correlated the number of topics within the policies with school-level CSA prevalence. We also explored whether there was lower CSA prevalence among schools with policies containing particular topics. Results suggested that greater comprehensiveness of schools’ entire CSA policies was negatively correlated with CSA prevalence; however, these findings did not approach statistical significance. The number of negative correlations observed between schools’ CSA policy elements and CSA prevalence among undergraduate women was greater than expected by chance alone, suggesting a possible connection between comprehensive CSA policies and CSA prevalence. Schools with policies that included a topic on their sexual assault response team had the lowest CSA prevalence for both women and men, and schools that included topics describing grievance/adjudication procedures had lower CSA prevalence. This study provides a novel examination of CSA and could inform needed research related to the impact of CSA policies on CSA.

Planned interventions and/or natural conditions often effect change on an ordinal categorical outcome (e.g., symptom severity). In such scenarios, it is sometimes desirable to assign a priori scores to observed changes in status, typically giving higher weight to changes of greater magnitude. We define change indices for such data based upon a multinomial model for each row of a c × c table, where the rows represent the baseline status categories. We distinguish an index designed to assess conditional changes within each baseline category from two others designed to capture overall change. One of these overall indices measures expected change across a target population. The other is scaled to capture the proportion of total possible change in the direction indicated by the data, so that it ranges from −1 (when all subjects finish in the least favorable category) to +1 (when all finish in the most favorable category). The conditional assessment of change can be informative regardless of how subjects are sampled into the baseline categories. In contrast, the overall indices become relevant when subjects are randomly sampled at baseline from the target population of interest, or when the investigator is able to make certain assumptions about the baseline status distribution in that population. We use a Dirichlet-multinomial model to obtain Bayesian credible intervals for the conditional change index that exhibit favorable small-sample frequentist properties. Simulation studies illustrate the methods, and we apply them to examples involving changes in ordinal responses for studies of sleep deprivation and activities of daily living. Copyright © 2015 John Wiley & Sons, Ltd.

Presenting continuous outcomes in Summary of Findings tables presents particular challenges to interpretation. When each study uses the same outcome measure, and the units of that measure are intuitively interpretable (e.g., duration of hospitalization, duration of symptoms), presenting differences in means is usually desirable. When the natural units of the outcome measure are not easily interpretable, choosing a threshold to create a binary outcome and presenting relative and absolute effects become a more attractive alternative. When studies use different measures of the same construct, calculating summary measures requires converting to the same units of measurement for each study. The longest standing and most widely used approach is to divide the difference in means in each study by its standard deviation and present pooled results in standard deviation units (standardized mean difference). Disadvantages of this approach include vulnerability to varying degrees of heterogeneity in the underlying populations and difficulties in interpretation. Alternatives include presenting results in the units of the most popular or interpretable measure, converting to dichotomous measures and presenting relative and absolute effects, presenting the ratio of the means of intervention and control groups, and presenting the results in minimally important difference units. We outline the merits and limitations of each alternative and provide guidance for meta-analysts and guideline developers.

Women who experience intimate partner violence (IPV) during pregnancy also tend to experience depressive symptoms. Unfortunately, little is known about how victimized women’s levels of depressive symptoms change longitudinally before pregnancy, during pregnancy, and after infant delivery. In addition, few studies have used a comparison group of women to determine if levels of depressive symptoms among victimized women differ from depressive symptom levels in women who have not experienced IPV. To help address these knowledge gaps, we examined longitudinal trends in levels of depressive symptoms among a sample of 76 women who did ( n = 33) and did not ( n = 43) experience physical IPV during pregnancy. Using multilevel analysis, we estimated the relationship of physical IPV victimization and women’s depressive symptom levels across six time periods: (a) the year before pregnancy, (b) first and second trimesters, (c) third trimester, (d) the first month postpartum, (e) Months 2 to 6 postpartum, and (f) Months 7 to 12 postpartum. Women who experienced physical IPV victimization during pregnancy had significantly higher levels of depressive symptoms during each time period ( p < .05). No significant difference between the two groups was found in the rate of change in levels of depressive symptoms over time. These findings point to the importance of screening for IPV within health care settings and suggest that women physically abused during pregnancy need safety interventions that are coordinated with interventions targeting symptoms of depression.

A common goal in environmental epidemiologic studies is to undertake logistic regression modeling to associate a continuous measure of exposure with binary disease status, adjusting for covariates. A frequent complication is that exposure may only be measurable indirectly, through a collection of subject-specific variables assumed associated with it. Motivated by a specific study to investigate the association between lung function and exposure to metal working fluids, we focus on a multiplicative-lognormal structural measurement error scenario and approaches to address it when external validation data are available. Conceptually, we emphasize the case in which true untransformed exposure is of interest in modeling disease status, but measurement error is additive on the log scale and thus multiplicative on the raw scale. Methodologically, we favor a pseudo-likelihood (PL) approach that exhibits fewer computational problems than direct full maximum likelihood (ML) yet maintains consistency under the assumed models without necessitating small exposure effects and/or small measurement error assumptions. Such assumptions are required by computationally convenient alternative methods like regression calibration (RC) and ML based on probit approximations. We summarize simulations demonstrating considerable potential for bias in the latter two approaches, while supporting the use of PL across a variety of scenarios. We also provide accessible strategies for obtaining adjusted standard errors to accompany RC and PL estimates.

The past half-century has witnessed great advances in the characterization of exposures to toxic chemicals in the workplace. With the development of robust personal monitors, it is now a relatively simple matter to measure the daily air levels among various workers in an observational group (a group of workers sharing observable factors such as job, location, department, etc.). Unfortunately, the interpretation of levels of exposure has not achieved the same degree of sophistication. In fact, decisions still tend to rely upon one-to-one comparisons of the highest measured air concentration with occupational exposure limits (OELs), much as they did 50 years ago (see, for example, Refs. (10–14)). If the highest air concentration is less than the OEL, then exposure is acceptable (in compliance) and vice versa. While such interpretations would be reasonable if occupational exposures were relatively constant, they seem remarkably naive when considered in light of daily exposure levels which vary between 10 and 4000 fold. Because exposures must be evaluated in the face of great variability within and between workers in observational groups, statistical methods are central to proper assessment procedures. Likewise, since monitoring is motivated by, and must ultimately relate to, exposure limits, the impact of exposure variability on the interpretation of OELs is important. The purpose of this chapter is to integrate statistical methods and the philosophical bases for exposure limits within a conceptual framework for interpreting exposure data. The methods will be illustrated with the four sets of exposure data depicted in graphical form. The sources and characteristics of these measurements are summarized and all data are listed in Appendix A for those who wish to work through the various examples in the text. Keywords: Random process; Compliance testing; Variations; Sampling; Statistical models; Exposure limits; Testing relative to limits; Nomenclature

BACKGROUND: Research concerning suicide and homicide in North Carolina is needed so that medical providers and others who develop and implement preventive and therapeutic interventions related to violence have an empirical base from which to work. METHODS: North Carolina Violent Death Reporting System data composed of death certificates, medical examiner reports, and law enforcement reports were analyzed to examine the prevalence of suicide and homicide in North Carolina during 2004-2007 and to describe the sociodemographic characteristics of suicide and homicide victims. RESULTS: Suicides and homicides accounted for 2.3% of all North Carolina deaths during 2004-2007. There were 12.0 suicides (95% confidence interval [CI], 11.7-12.4) and 7.2 homicides (95% CI, 6.9-7.4) per 100,000 North Carolina residents. Suicide rates were higher among men and boys, whites, non-Hispanics, and persons aged > or = 35 years. Homicide rates were higher among men and boys, American Indians, blacks, Hispanics, and persons aged Language: en

Purpose To determine the prevalence of patterns of intimate partner violence (IPV) victimization from adolescence to young adulthood, and document associations with selected sociodemographic and experiential factors. Methods We used prospective data from the National Longitudinal Study of Adolescent Health to group 4134 respondents reporting only opposite-sex romantic or sexual relationships in adolescence and young adulthood into four victimization patterns: no IPV victimization, adolescent-limited IPV victimization, young adult onset IPV victimization, and adolescent-young adult persistent IPV victimization. Results Forty percent of respondents reported physical or sexual victimization by young adulthood. Eight percent experienced IPV only in adolescence, 25% only in young adulthood, and 7% showed persistent victimization. Female sex, Hispanic and non-Hispanic black race/ethnicity, an atypical family structure (something other than two biologic parents, step-family, single parent), more romantic partners, experiencing childhood abuse, and early sexual debut (before age 16) were each associated with one or more patterns of victimization versus none. Number of romantic partners and early sexual debut were the most consistent predictors of violence, its timing of onset, and whether victimization persisted across developmental periods. These associations did not vary by biological sex. Conclusions Substantial numbers of young adults have experienced physical or sexual IPV victimization. More research is needed to understand the developmental and experiential mechanisms underlying timing of onset of victimization, whether victimization persists across time and relationships, and whether etiology and temporal patterns vary by type of violence. These additional distinctions would inform the timing, content, and targeting of violence prevention efforts.

Measurements of biomarkers have been used, in conjunction with those of external exposure, to provide information about the rates of human uptake, bioactivation, and detoxification of chemicals. Direct knowledge of these processes is valuable because it can reduce reliance on animal models for making extrapolations about disease risks in humans. In practice, however, valid and precise quantification of exposure—biomarker relationships has been hampered by a number of problems. These include errors in the measurement of true mean exposure and true mean biomarker levels, nonlinear exposure—biomarker relationships, background biomarker levels in unexposed individuals, and response levels that fall below analytic limits of detection. In this article we develop nonlinear mixed models and corresponding likelihood expressions that can address these concerns, and we describe maximum likelihood estimation techniques that permit valid statistical inferences to be made using standard software. The models and methods are applied to both experimental and simulated exposure—biomarker data to illustrate their utility and limitations.

Oxidative DNA damage is one of the key events thought to be involved in mutation and cancer. The present study examined the accumulation of M1dG, 3-(2'-deoxy-beta-D-erythro-pentofuranosyl)-pyrimido[1,2-a]-purin-10(3H)-one, DNA adducts after single dose or 1-year exposure to polyhalogenated aromatic hydrocarbons (PHAH) in order to evaluate the potential role of oxidative DNA damage in PHAH toxicity and carcinogenicity. The effect of PHAH exposure on the number of M1dG adducts was explored initially in female mice exposed to a single dose of either 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) or a PHAH mixture. This study demonstrated that a single exposure to PHAH had no significant effect on the number of M1dG adducts compared to the corn oil control group. The role of M1dG adducts in polychlorinated biphenyl (PCB)-induced toxicity and carcinogenicity was further investigated in rats exposed for a year to PCB 153, PCB 126, or a mixture of the two. PCB 153, at doses up to 3000 microg/kg/day, had no significant effect on the number of M1dG adducts in liver and brain tissues from the exposed rats compared to controls. However, 1000 ng/kg/day of PCB 126 resulted in M1dG adduct accumulation in the liver. More importantly, coadministration of equal proportions of PCB 153 and PCB 126 resulted in dose-dependent increases in M1dG adduct accumulation in the liver from 300 to 1000 ng/kg/day of PCB 126 with 300-1000 microg/kg/day of PCB 153. Interestingly, the coadministration of different amounts of PCB 153 with fixed amounts of PCB 126 demonstrated more M1dG adduct accumulation with higher doses of PCB 153. These results are consistent with the results from cancer bioassays that demonstrated a synergistic effect between PCB 126 and PCB 153 on toxicity and tumor development. In summary, the results from the present study support the hypothesis that oxidative DNA damage plays a key role in toxicity and carcinogenicity following long-term PCB exposure.

Objective This study examines links between women’s experiences of violence during adulthood (including physical and sexual violence) and women’s physical health, mental health, and functional status. Methods Data were analyzed from a representative sample of 9,830 North Carolina women surveyed by the North Carolina Behavioral Risk Factor Surveillance System (BRFSS). Results One-quarter of the women experienced violence as adults, with current or ex-partners being the most common perpetrators. Logistic regression analyses that controlled for the sociodemographic characteristics of the women found that women who experienced violence were significantly more likely than other women to have poor physical health, poor mental health, and functional limitations. Moreover, these negative health outcomes were most prevalent among the women who experienced a combination of both physical and sexual violence. Conclusions These findings underscore the need for trauma-informed women’s health services and policies.

Objectives Although partner violence during pregnancy has serious consequences for women’s health, little is known about how physical partner violence may change throughout pregnancy transitions. Even less is known about changes in sexual and psychological partner violence throughout pregnancy transitions. In addition, few research studies on pregnancy and partner violence have examined these changes among both victimized women (i.e., women who report physical partner violence at the beginning of their pregnancies) and comparison women (i.e., women who do not report physical partner violence during this same time period). Methods This longitudinal research study investigated 76 women’s experiences with partner violence beginning 1 year before their pregnancies, and continuing throughout their pregnancies until 1 year after delivery. Four structured interviews were administered to participants, and information was collected concerning the women’s partner violence experiences. Hierarchical linear modeling was used to determine whether there were statistically significant differences between the violence rates experienced by the victimized women relative to the comparison women at each time period. Results The results show that partner violence rates do change throughout pregnancy transitions, and that these changes are seen for both victimized and comparison women. Conclusions Pregnant and postpartum women are in need of comprehensive services that promote both their health and safety. This study offers care providers clinical implications for their work with pregnant and postpartum women, as well as policy and research recommendations.

We examine the effects of modelling errors, such as underfitting and overfitting, on the asymptotic power of tests of association between an explanatory variable x and an outcome in the setting of generalized linear models. The regression function for x is approximated by a polynomial or another simple function, and a chi-square statistic is used to test whether the coefficients of the approximation are simultaneously equal to zero. Adding terms to the approximation increases asymptotic power if and only if the fit of the model increases by a certain quantifiable amount. Although a high degree of freedom approximation offers robustness to the shape of the unknown regression function, a low degree of freedom approximation can yield much higher asymptotic power even when the approximation is very poor. In practice, it is useful to compute the power of competing test statistics across the range of alternatives that are plausible a priori. This approach is illustrated through an application in epidemiology.

North Carolina women were surveyed to examine whether women’s disability status was associated with their risk of being assaulted within the past year. Women’s violence experiences were classified into three groups: no violence, physical assault only (without sexual assault), and sexual assault (with or without physical assault). Multivariable analysis revealed that women with disabilities were not significantly more likely than women without disabilities to have experienced physical assault alone within the past year (odds ratio [OR] = 1.18, 95% Confidence Interval [CI] = 0.62 to 2.27); however, women with disabilities had more than 4 times the odds of experiencing sexual assault in the past year compared to women without disabilities (OR = 4.89, 95% CI = 2.21 to 10.83).

Objective: To identify demographic and work related factors that predict blood levels of styrene and styrene-7,8-oxide (SO) in the fibreglass reinforced plastics (FRP) industry. Methods: Personal breathing-zone air samples and whole blood samples were collected repeatedly from 328 reinforced plastics workers in the Unuted States between 1996 and 1999. Styrene and its major metabolite SO were measured in these samples. Multivariable linear regression analyses were applied to the subject-specific levels to explain the variation in exposure and biomarker levels. Results: Exposure levels of styrene were approximately 500-fold higher than those of SO. Exposure levels of styrene and SO varied greatly among the types of products manufactured, with an 11-fold range of median air levels among categories for styrene and a 23-fold range for SO. Even after stratification by job title, median exposures of styrene and SO among laminators varied 14- and 31-fold across product categories. Furthermore, the relative proportions of exposures to styrene and SO varied among product categories. Multivariable regression analyses explained 70% and 63% of the variation in air levels of styrene and SO, respectively, and 72% and 34% of the variation in blood levels of styrene and SO, respectively. Overall, air levels of styrene and SO appear to have decreased substantially in this industry over the last 10–20 years in the US and were greatest among workers with the least seniority. Conclusions: As levels of styrene and SO in air and blood varied among product categories in the FRP industry, use of job title as a surrogate for exposure can introduce unpredictable measurement errors and can confound the relation between exposure and health outcomes in epidemiology studies. Also, inverse relations between the intensity of exposure to styrene and SO and years on the job suggest that younger workers with little seniority are typically exposed to higher levels of styrene and SO than their coworkers.

In case-control studies, it is common for a categorical exposure variable to be misclassified. It is also common for exposure status to be informatively missing for some individuals, in that the probability of missingness may be related to exposure. Procedures for addressing the bias due to misclassification via validation data have been extensively studied, and related methods have been proposed for dealing with informative missingness based on supplemental sampling of some of those with missing data. In this paper, we introduce study designs and analytic procedures for dealing with both problems simultaneously in a 2x2 analysis. Results based on convergence in probability illustrate that the combined effects of missingness and misclassification, even when the latter is non-differential, can lead to naive exposure odds ratio estimates that are inflated or on the wrong side of the null. The motivating example comes from a case-control study of the association between low birth weight and the diagnosis of breast cancer later in life, where self-reported birth weight for some women is supplemented by accurate information from birth certificates.

Jet propulsion fuel 8 (JP-8) is the major jet fuel used worldwide and has been recognized as a major source of chemical exposure, both inhalation and dermal, for fuel-cell maintenance workers. We investigated the contributions of dermal and inhalation exposure to JP-8 to the total body dose of U.S. Air Force fuel-cell maintenance workers using naphthalene as a surrogate for JP-8 exposure. Dermal, breathing zone, and exhaled breath measurements of naphthalene were obtained using tape-strip sampling, passive monitoring, and glass bulbs, respectively. Levels of urinary 1- and 2-naphthols were determined in urine samples and used as biomarkers of JP-8 exposure. Multiple linear regression analyses were conducted to investigate the relative contributions of dermal and inhalation exposure to JP-8, and demographic and work-related covariates, to the levels of urinary naphthols. Our results show that both inhalation exposure and smoking significantly contributed to urinary 1-naphthol levels. The contribution of dermal exposure was significantly associated with levels of urinary 2-naphthol but not with urinary 1-naphthol among fuel-cell maintenance workers who wore supplied-air respirators. We conclude that dermal exposure to JP-8 significantly contributes to the systemic dose and affects the levels of urinary naphthalene metabolites. Future work on dermal xenobiotic metabolism and toxicokinetic studies are warranted in order to gain additional knowledge on naphthalene metabolism in the skin and the contribution to systemic exposure.

Establishing and characterizing exposure-biomarker relationships represent important steps to understanding how exposure to a harmful environmental toxin ultimately leads to disease in human populations. Here, we present a statistical model to characterize a nonlinear exposure-biomarker relationship and use occupational benzene exposures to illustrate the application. We also attempt to estimate the range of linear metabolism of benzene by fitting our model to data from a recent study of biomarkers (benzene-oxide-albumin adducts) measured in a population of Chinese workers exposed to benzene. Estimating the parameters of interest is difficult due to within and between subject variability in exposure and biomarker levels and because of exposure measurement error.

Summary For a univariate linear model, the Box–Cox method helps to choose a response transformation to ensure the validity of a Gaussian distribution and related assumptions. The desire to extend the method to a linear mixed model raises many vexing questions. Most importantly, how do the distributions of the two sources of randomness (pure error and random effects) interact in determining the validity of assumptions? For an otherwise valid model, we prove that the success of a transformation may be judged solely in terms of how closely the total error follows a Gaussian distribution. Hence the approach avoids the complexity of separately evaluating pure errors and random effects. The extension of the transformation to the mixed model requires an exploration of its potential effect on estimation and inference of the model parameters. Analysis of longitudinal pulmonary function data and Monte Carlo simulations illustrate the methodology discussed.

Superfund Research Program and Center for Exposure Biology, School of Public Health, University of California, Berkeley, CA 94720, USA, 1Department of Biostatistics and Bioinformatics, Rollins School of Public Health, Emory University, Atlanta, GA 30322, USA, 2Department of Toxicology, Universite de Technologie de Compiegne, 60200 Compiegne, France, 3Institut National de l'Environnement Industriel et des Risques, DRC/VIVA/METO Unit, 60550 Verneuil en Halatte, France, 4Department of Biostatistics, Gillings School of Global Public Health, University of North Carolina, Chapel Hill, NC 27599-7420, USA and 5Department of Environmental Health, School of Public Health, Seoul National University, Seoul 151–742, Republic of Korea

Despite the high prevalence of violence in the lives of women of reproductive age, and the adverse consequences posed to pregnant women suffering from PTSD, few studies have examined violence-related trauma and PTSD among pregnant women. A structured research interview was administered to a convenience sample of 85 prenatal care women to collect information regarding their experiences of violence victimization, PTSD symptoms related to violence-related traumatic events, and whether these symptoms impaired various aspects of the women's daily functioning. Results indicated that 69% of the women experienced a violent traumatic event, with the majority of events occurring before the current pregnancy and being perpetrated by the women's intimate partners. The traumatized women evidenced high levels of PTSD symptoms during pregnancy, with 58% of the women meeting criteria for a PTSD diagnosis. Moreover, many of the women with PTSD symptoms reported that the symptoms adversely affected aspects of their daily functioning.

Statistical analyses of data from a classroom-based study illustrate the need to account for intra-class clustering in studies involving schools, classrooms, and other higher order units of analysis. Students were clustered in homerooms that were assigned to intervention and comparison conditions. Standard multiple linear regression analysis yielded a significant group effect but incorrectly ignored intra-cluster response correlations. A multilevel model appropriately accounting for the dependency among responses in the same cluster yielded a nonsignificant group effect. Implications for the analysis of intervention research data are discussed.

This pilot study examined the impact of a prototype problem-solving skills-training program, Making Choices, on proximal outcomes in 70 6th-grade students. Method: Students received three components of Making Choices and completed pretest and posttest measures of skills on each component. Paired-sample t tests were used to assess proximal effects. Baseline measures were used to identify four subgroups of children, and differences in skill acquisition across subgroups were assessed. Results: Students displayed significantly higher scores at posttest on measures of two of three proximal skills. Nonaggressive-accepted and aggressive-accepted students displayed stronger skills at posttest, and aggressive-rejected and nonaggressive-rejected students failed to show significant gains. Conclusions: Results provided preliminary evidence of the effectiveness of Making Choices and guided refinement of the prototype program.

Polyenergetic, high-photon-flux X-ray absorption methods promise to make possible rapid, high-resolution measurements of phase fractions in multiphase systems. Improved statistical methods for data analysis, however, are needed. We develop a set of three new path-length estimators as alternatives to the traditional lumped-energy approach, and we compared this set of estimators for two classes of examples. The experimental comparisons were performed with clean systems to allow reliable measures of accuracy and precision of all material estimates; however, soil systems, for which X-ray attenuation methods have been routinely applied, have radiation attenuation properties similar to the systems used in this work, making these results directly applicable to soil systems. Results show that the new estimators provide more accurate and precise estimates of path length than the traditional lumped-energy approach, especially for challenging, multiple material estimation problems. The bias of estimated path lengths was reduced by at least 80% in all cases and was typically around 90% compared with the standard approach. Our sample results also demonstrate the reliable simultaneous measurement of three material path lengths, a new and desirable capability for multiphase flow experiments. Alternatively, the improved estimators developed in this work can be used to reduce measurement times compared with traditional methods for estimates of similar quality.

Objectives Population sexual mixing patterns can be quantified using Newman9s assortativity coefficient (r). Suggested methods for estimating the SE for r may lead to inappropriate statistical conclusions in situations where intracluster correlation is ignored and/or when cluster size is predictive of the response. We describe a computer-intensive, but highly accessible, within-cluster resampling approach for providing a valid large-sample estimated SE for r and an associated 95% CI. Methods We introduce needed statistical notation and describe the within-cluster resampling approach. Sexual network data and a simulation study were employed to compare within-cluster resampling with standard methods when cluster size is informative. Results For the analysis of network data when cluster size is informative, the simulation study demonstrates that within-cluster resampling produces valid statistical inferences about Newman9s assortativity coefficient, a popular statistic used to quantify the strength of mixing patterns. In contrast, commonly used methods are biased with attendant extremely poor CI coverage. Within-cluster resampling is recommended when cluster size is informative and/or when there is within-cluster response correlation. Conclusions Within-cluster resampling is recommended for providing valid statistical inferences when applying Newman9s assortativity coefficient r to network data.

SUl\lhIARY. Information from detectable exposure measurements randomly sampled from a left-truncated log-normal distribution may be used to evaluate the distribution of nondetectable values that fall below all analytic limit of detection_ If the proportion of nondetects is larger than expected under log normality. alternative models to account for these unobserved data should be considered. We discuss one such model that incorporates a mixture of true zero exposures and a log-normal distribution with possible left censoring, previously considered ill a different context by ManItou and Halsey (1995, Biometrics 51, 1570-1578). A particular relationship is demonstrated between maximum likelihood parameter estimates based on this mixture model and those assuming either left-tnlllcated or left-censored data. These results emphasize the Heed for caution when choosing a model to fit data involving nondetectable values. A one-sided likelihood ratio test for COlli paring Illeall exposure under the mixture model to an occupational exposure limit is then developed and evaluated via simulations. All example demonstrates the potential impact of specifying an incorrect model for the Ilondetectable value::;.

We propose an iterative method of estimation for discrete missing data problems that is conceptually different from the Expectation–Maximization (EM) algorithm and that does not in general yield the observed data maximum likelihood estimate (MLE). The proposed approach is based conceptually upon weighting the set of possible complete-data MLEs. Its implementation avoids the expectation step of EM, which can sometimes be problematic. In the simple case of Bernoulli trials missing completely at random, the iterations of the proposed algorithm are equivalent to the EM iterations. For a familiar genetics-oriented multinomial problem with missing count data and for the motivating example with epidemiologic applications that involves a mixture of a left censored normal distribution with a point mass at zero, we investigate the finite sample performance of the proposed estimator and find it to be competitive with that of the MLE. We give some intuitive justification for the method, and we explore an interesting ...

Background: Prenatal care provides an opportunity for counseling about behaviors and experiences that increase the likelihood of adverse maternal and fetal outcomes. Objectives: To document (1) prevalence of preventive health counseling during prenatal care, (2) prevalence of women in higher need of counseling about specific health concerns, and (3) whether women in higher need for counseling were more likely than women in lower need to have received counseling. Methods: Analysis of the Pregnancy Risk Assessment Monitoring System (PRAMS), a state-specific, population-based, random sample of postpartum women, was performed by using data from 14 states for births during 1997 or 1998, for a total of 24,620 participants. Outcome measures included report of preventive health counseling during prenatal visits by specific topic as well as behaviors and experiences about cigarette use, alcohol use, breast-feeding, partner violence, and preterm labor. Results: The percentage of women that report preventive counseling during prenatal care is relatively high (≥75%) for 9 of 13 topics. However, the percentage of women that report counseling is relatively low (

Objectives: This study estimates the prevalence of stressful life events and physical abuse among North Carolina women prior to infant delivery, and examines potential associations between abuse and the other stressors. Methods: Data were from the North Carolina Pregnancy Risk Assessment Monitoring System, a statewide representative survey of over 2,600 postpartum women. The survey assessed women's sociodemographic characteristics and their experiences of physical abuse and 13 other stressful life events before delivery. The prevalences of each life event and abuse were estimated. Logistic regression modeled the probability of women having high levels of stressful life events in relation to physical abuse and sociodemographics. Results: Most women were married, white, high school graduates, aged 20 or older. The most common stressful life events were residential moves, increased arguing with husbands/partners, family member hospitalizations, financial hardship, and deaths of loved ones. Fourteen percent of women had high levels of stressful events (5 or more), and almost 9% were physically abused. Abuse was positively associated with increased arguing with husbands/partners, physical fighting, having someone close with an alcohol/drug problem, becoming separated/divorced, and financial hardship. Logistic regression analysis showed that a high level of stressful life events was significantly more likely among women abused both before and during pregnancy (OR = 11.94) and among women abused before but not during pregnancy (OR = 14.19). Conclusions: The high frequency of multiple stressful events and abuse in women's lives suggests that women's care providers should ask their patients about these issues, and offer appropriate referral/interventions to those in need.

The stability of cysteinyl adducts of benzene oxide (BO) and mono-S-substituted cysteinyl adducts of 1,4-benzoquinone (1,4-BQ) was investigated in both hemoglobin (Hb) and albumin (Alb) following administration of a single oral dose of 400 mg [U-14C/13C6]benzene/kg body weight to F344 rats. Total radiobound adducts to Hb were stable, as were adducts formed by the reaction of [13C6]BO with cysteinyl residues on Hb. In both cases adduct stability was indicated by zero-order kinetics with decay rates consistent with the lifetime of rat erythrocytes. Hb adducts of 1,4-BQ were not detected, possibly due to the production of multi-S-substituted adducts within the erythrocyte. Regarding Alb binding, total radiobound adducts decayed more rapidly than expected (half-life of 0.4 days), suggesting that uncharacterized benzene metabolites were noncovalently bound or formed unstable adducts with Alb. Although adducts from reactions of BO and 1,4-BQ with Alb both decayed with rates consistent with those of Alb turnover in the rat, the half-life for 1,4-BQ-Alb (2.5 days) was shorter than that for BO-Alb (3.1 days), suggesting some instability of 1,4-BQ-Alb. Assuming similar rates of adduct instability in humans and rats, the 1,4-BQ-Alb adducts would be eliminated with a half-life of approximately 8 days, compared with BO-Alb, which would be expected to turnover with Alb (half-life of approximately 21 days).

This mansscript studies analytically the consequences of changing the scale of measurement of a continuous independent variable in a multiple linear regression setting. Assuming the continuous outcome variable, a continuous exposure variable, and a continuous control variable follow a trivariate Gaussian distribution, we examine the effect upon the structure of the modei of dichotomizing the continuous control variable. It is shown that, after dichotomizaiion, the condirionai expected vaiiie of the response is a quotient of two non-hear functions and hence is no longer linear in the exposure variable. Thus, when an underlying continuous independent variable is dichotomized in multiple linear regression, and one fits a linear model using the dichotomous variable, this model's linear structure is misspecified. The estimates obtained from this model are incorrect and potentially misleading.

Methods for optimal sample size determination are developed using four popular multiple comparison procedures (Scheffe's, Bonferroni's, Tukey's and Dunnett's procedures), where random samples of the same size n are to be selected from k (>/=2) normal populations with common variance sigma2, and where primary interest concerns inferences about a family of L linear contrasts among the k population means. For a simultaneous coverage probability of (1-alpha), the optimal sample size is defined to be the smallest integer value n*m such that, simultaneously for all L confidence intervals, the width of the lth confidence interval will be no greater than tolerance 2deltal (l=1,2,...,L) with tolerance probability at least (1-gamma), treating the pooled sample variance S2p as a random variable. Using Scheffe's procedure as an illustration, comparisons are made to usual sample size methods that incorrectly ignore the stochastic nature of S2p. The latter approach can lead to serious underestimation of required sample sizes and hence to unacceptably low values of the actually tolerance probability (1-gamma'). Our approach guarantees a lower bound of [1-(alpha+gamma)] for the probability that the L confidence intervals will both cover the parametric functions of interest and also be sufficiently narrow. Recommendations are provided regarding the choices among the four multiple comparison procedures for sample size determination and inference-making.

OBJECTIVES: The authors analyzed interview responses of patients at a prenatal care clinic to explore whether women who had been victims of sexual and physical abuse were more likely than non-victimized women to have experienced a sexually transmitted disease (STD). METHODS: A consecutive sample of 774 prenatal patients of a large health department in North Carolina were interviewed concerning a variety of health issues, including violence and STDs. Logistic regression analysis was used to model the women's STD status as a function of their experiences of sexual and physical abuse, controlling for several potentially confounding factors. RESULTS: Thirty percent of the women reported having experienced at least one STD, with the most common infections being chlamydia and gonorrhea. Twenty-eight percent of the women reported having been victims of abuse; 16% reported physical abuse only, while 12% reported both physical and sexual abuse. The majority of violence was domestic in nature, perpetrated by the victims' husbands, boyfriends, male friends, and relatives. After controlling for confounding variables, the authors found that women who reported both physical and sexual abuse were significantly more likely to have experienced STDs than non-victims (odds ratio [OR] = 2.25; 95% confidence interval [CI] 1.37, 3.69). The logistic regression analysis also showed a relationship of borderline statistical significance between non-sexual physical abuse and STDs. CONCLUSIONS: Health care providers should routinely screen patients for both abuse and STDs, and they should assist identified women in accessing appropriate health, social, and legal services.

During the manufacturing of reinforced plastics, large amounts of styrene and trace quantities of styrene-7,8-oxide (SO) are released. Since previous work suggests that inhalation of even small amounts of SO might be an important health risk, we investigated several possible factors contributing to styrene and SO exposure during the manufacture of reinforced plastics. Factors related to job type, worker and the type and quantity of styrene-containing resins were investigated using mixed-effects multiple linear regression models. Overall, SO exposure levels were positively correlated with styrene exposure levels. However, this correlation was statistically significant only among hand laminators who had the highest exposures to both styrene and SO. An important factor for predicting both styrene and SO concentrations was the type of resin used, while the quantity of resin consumed was predictive of styrene but not of SO exposure. Since SO exposure appears to be associated with factors other than coexposure to styrene, more effort should be placed on investigating emissions of SO per se. The type of mixed-models regression analysis employed in this study can be used for clarifying the underlying patterns for exposures to styrene and SO as well as for evaluating preventive measures. 1999 British Occupational Hygiene Society. Published by Elsevier Science Ltd. All rights reserved.

Using data from a consecutive sample of approximately 700 prenatal patients aged 12 to 19, the extent of violence and substance use in the lives of these women was examined, as were associations between violence and substance use. The findings show that 29% of the study participants had been victims of violence, with 15% experiencing physical violence only, 5% sexual violence only, and 9% both physical and sexual violence. Adjusted odds ratios (ORs) and 95% confidence intervals (CIs) from logistic regression analyses showed that, after controlling for confounding factors, victims were more likely than nonvictims to use cigarettes, alcohol, and illicit drugs, with victims of both physical and sexual violence being the most likely to use each type of substance. Ordinal logistic regression analysis found that, compared to nonvictims, violence victims were more likely to have more severe patterns of substance use (use multiple types of substances), with victims of both physical and sexual violence being the most likely to be multiple substance users. These findings underline the importance of both violence and substance use as health concerns among our youth.

The large concentration of female illicit drug users in state correctional facilities prompted an examination of the associations among different types of drug use and sexual risk factors related to human immunodeficiency virus (HIV) among women inmates. A consecutive sample of 805 women felons admitted to the North Carolina Correctional Institution for Women between July 1991 and November 1992 was interviewed. Of these inmates, 651 had complete information on relevant characteristics. Of the women inmates, 73% had used drugs prior to incarceration; most women were crack smokers only (33%), followed by non-drug users (27%), other drug users (19%), crack-smoking injecting drug users (15%), and injecting drug users only (6%). Inconsistent condom use with multiple sex partners, a history of a diagnosed sexually transmitted disease (STD), a drug-injecting sex partner, or exchanging sex for money or drugs prior to incarceration were reported by 55% of the women. Sexual risk factors differed across different types of drug users, with crack-smoking injectors being placed at greatest potential risk for exposure to heterosexually transmitted HIV, followed by injecting drug users, crack smokers, and then other drug users. Given the differential associations between sexual risk factors and types of drug use, prison-based sexual-risk reduction strategies should be tailored to specific types of drug users. In times of limited resources, special attention should be given to crack smokers and/or drug injectors.

OBJECTIVES: To conduct a comprehensive evaluation of long term changes in occupational exposure among a broad cross section of industries worldwide. METHODS: A review of the scientific literature identified studies that reported historical changes in exposure. About 700 sets of data from 119 published and several unpublished sources were compiled. Data were published over a 30 year period in 25 journals that spanned a range of disciplines. For each data set, the average exposure level was compiled for each period and details on the contaminant, the industry and location, changes in the threshold limit value (TLV), as well as the type of sampling method were recorded. Spearman rank correlation coefficients were used to identify monotonic changes in exposure over time and simple linear regression analyses were used to characterise trends in exposure. RESULTS: About 78% of the natural log transformed data showed linear trends towards lower exposure levels whereas 22% indicated increasing trends. (The Spearman rank correlation analyses produced a similar breakdown between exposures monotonically increasing or decreasing over time.) Although the rates of reduction for the data showing downward trends ranged from -1% to -62% per year, most exposures declined at rates between -4% and -14% per year (the interquartile range), with a median value of -8% per year. Exposures seemed to increase at rates that were slightly lower than those of exposures which have declined over time. Data sets that showed downward (versus upward) trends were influenced by several factors including type and carcinogenicity of the contaminant, type of monitoring, historical changes in the threshold limit values (TLVs), and period of sampling. CONCLUSIONS: This review supports the notion that occupational exposures are generally lower today than they were years or decades ago. However, such trends seem to have been affected by factors related to the contaminant, as well as to the period and type of sampling.

This study examined the associations between preincarceration risky sex and drug behaviors and HIV status among incarcerated women. A consecutive sample of 805 women felons admitted to the North Carolina Correctional Institution for Women between July 1991 and November 1992 was interviewed. Of these women, 700 granted permission to access their prison medical records and had complete information on relevant variables. Four percent of the women inmates were HIV positive. Over 80% of the women inconsistently used condoms during intercourse, and over 15% injected drugs, had a drug-injecting sex partner, and exchanged sex for money/drugs. The exchange of sex for money/drugs was associated with being HIV positive. The study findings suggest that prison-based HIV prevention programs should emphasize sexual and drug risk-reduction strategies as a means to reduce the heterosexual HIV risks facing women inmates once released back into the general community.

OBJECTIVES: This study examined whether incarceration during pregnancy is associated with infant birthweight. METHODS: Multivariable analyses compared infant birthweight outcomes among three groups of women: 168 women incarcerated during pregnancy, 630 women incarcerated at a time other than during pregnancy, and 3910 women never incarcerated. RESULTS: After confounders were controlled for, infant birthweights among women incarcerated during pregnancy were not significantly different from women never incarcerated; however, infant birthweights were significantly worse among women incarcerated at a time other than during pregnancy than among never-incarcerated women and women incarcerated during pregnancy. CONCLUSIONS: Certain aspects of the prison environment (shelter, food, etc.) may be health-promoting for high-risk pregnant women.

For estimating the difference (μ1 - μ2) between two population means based on random samples of sizes n 1 and n 2 from two normal populations with possibly unequal variances optimal sample sizes are determined based on the (unconditional) criterion that the length of a 100(l-α)% confidence interval for (μ1μ2) is no greater than 2h with probability at least (1 -δ). Consideration is also given to a conditional criterion involving the further requirement that the confidence interval must contain the value of (μ1μ2). and results based on the use of these two criteria are compared. For either the unconditional or conditional criterion, the optimal value of the ratio that minimizes the total sample size (n 1 + n 2) is found numerically to be approximately equal to the ratio .