Mediation analysis with binary outcomes: Direct and indirect effects of pro-alcohol influences on alcohol use disorders

Heart rate reactivity mediates the relationship between trait gratitude and acute myocardial infarction

OBJECTIVE

This study examines the relationship between trait gratitude and acute myocardial infarction. A burgeoning body of literature suggests that gratitude can play a role in regulating individual's cardiovascular responses to stress which in turn, may reduce the incidence of cardiovascular disease such as acute myocardial infarction. However, to date no research has examined these effects.

METHOD

This study used the Mid-Life in the United States dataset (MIDUS; N = 1031) to assess these relationships. Participants completed a standardised cardiovascular stress-testing laboratory protocol and were assessed at a second time-point; on average 6.7 years later.

RESULTS

Results from logistic parallel mediation models suggest that trait gratitude was found to be significantly associated with reduced risk of acute myocardial infarction through the mechanism of increased heart rate reactivity, β = -0.098, 95%CI [- 0.331, - 0.010]. However, neither systolic nor diastolic blood pressure reactivity mediated this relationship.

CONCLUSIONS

These findings suggest that gratitude may be associated with certain aspects of physical health. Specifically, our study reveals a potential link between gratitude and cardiovascular reactivity, which could be a mechanism through which trait gratitude contributes to reductions in the risk of myocardial infarction. As such, this study highlights the potential utility of positive psychological factors, such as gratitude, in promoting cardiovascular health.

A Causal Approach to Functional Mediation Analysis with Application to a Smoking Cessation Intervention

The increase in the use of mobile and wearable devices now allows dense assessment of mediating processes over time. For example, a pharmacological intervention may have an effect on smoking cessation via reductions in momentary withdrawal symptoms. We define and identify the causal direct and indirect effects in terms of potential outcomes on the mean difference and odds ratio scales, and present a method for estimating and testing the indirect effect of a randomized treatment on a distal binary variable as mediated by the nonparametric trajectory of an intensively measured longitudinal variable (e.g., from ecological momentary assessment). Coverage of a bootstrap test for the indirect effect is demonstrated via simulation. An empirical example is presented based on estimating later smoking abstinence from patterns of craving during smoking cessation treatment. We provide an R package, funmediation, available on CRAN at https://cran.r-project.org/web/packages/funmediation/index.html, to conveniently apply this technique. We conclude by discussing possible extensions to multiple mediators and directions for future research.

Factors associated with melatonin use in child sleep disturbance: A survey of Australian parents

OBJECTIVE

Sleep disturbances are prevalent in Australian children and cause a significant amount of stress to both children and their parents. If sleep disturbances are left untreated, they can result in detrimental consequences to both child and parental wellbeing. While behavioural interventions are recognised as the gold standard treatment for paediatric insomnia, there is a growing use of melatonin from both prescription and non-prescription sources in paediatrics. However, empirical research on the efficacy and safety of melatonin for children is sparse, conflicting, and inconsistent. This first Australian study aimed to investigate correlates of melatonin use in a sample of children with sleep disturbance in Australia and assess whether melatonin is being administered to children contrary to Therapeutic Goods Administration Guidelines.

METHODS

An exploratory online study was undertaken with 318 parents of Australian children with sleep disturbance (95.30% females). Logistic regression, mediation and descriptive analysis were conducted.

RESULTS

As hypothesised, higher parental stress, older child age, and a diagnosis of ADHD were significantly associated with melatonin use in this sleep disturbed sample. Other potential correlates such as financial stress, number of household dependents, child sleep disturbance, and autism were not significantly associated with melatonin use in adjusted analyses.

CONCLUSION

Findings show off-label melatonin use in children with sleep disturbance is prevalent. For the first time, this study showed an indirect path linking child sleep disturbances with melatonin use via parental stress. Overall, while the safety of melatonin use remains unknown, there is a clear need to mitigate parental stress to minimise potentially unmonitored and improper use of melatonin in children.

Furloughed Employees' Voluntary Turnover: The Role of Procedural Justice, Job Insecurity, and Job Embeddedness

During the COVID-19 lockdown period, several employers used furloughs, that is, temporary layoffs or unpaid leave, to sustain their businesses and retain their employees. While furloughs allow employers to reduce payroll costs, they are challenging for employees and increase voluntary turnover. This study uses a two-wave model (Time 1: n = 639/Time 2: n = 379) and confirms that furloughed employees' perceived justice in furlough management and job insecurity (measured at Time 1) explain their decision to quit their employer (measured at Time 2). In addition, our results confirm that furloughed employees' job embeddedness (measured at Time 1) has a positive mediator effect on the relationship between their perceived procedural justice in furlough management (measured at Time 1) and their turnover decision (Time 2). We discuss the contribution of this study to the fields of knowledge and practice related to turnover and furlough management to reduce their financial, human, and social costs.

Afraid but misinformed: Conspiracist beliefs cancel the positive influence of fear of COVID-19 on vaccination intentions - Findings from a Romanian sample

Understanding the factors that make people more likely to refuse vaccination against COVID-19 is crucial in order to design public health messages efficient in increasing vaccination rates. As COVID-19 creates risks of seriously damaging health effects, fear of this disease is as a significant determinant of vaccination intentions, as indicated by past research. Nevertheless, this positive influence may be limited in people who do not consider vaccines as a solution to protect against COVID-19, especially those who hold conspiracist beliefs about the new coronavirus and, implicitly, about the newly developed vaccines. The present study examined in a cross-sectional design on a convenience sample (N = 564) the joint effect of fear of COVID-19 and conspiracist beliefs on vaccination intentions, advancing past research on their independent influences. Furthermore, we investigated and controlled the effects of perceived risk of catching COVID-19, trust in medical experts, attitude towards vaccination and socio-demographical characteristics (i.e., gender, age, and education), previously found to be associated to COVID-19 vaccination intentions. We also tested the effect of ambivalence towards vaccination, i.e., the degree to which people simultaneously hold positive and negative evaluations of this intervention, as the widespread misinformation on the new coronavirus and its vaccines may induce ambivalence on this latter issue in many people. The results showed that the positive effect of fear of COVID-19 on vaccination intentions emerged only in participants who tend not to endorse conspiracist ideas on the new coronavirus. Moreover, higher vaccine hesitancy was found in participants with higher ambivalence towards vaccination, in those who perceive the risk of being contaminated by the new coronavirus as low, and in those with more negative attitudes towards vaccines in general. Vaccine ambivalence also emerged as a mediator of the negative effects of conspiracist beliefs about COVID-19 on vaccination intentions. This pattern of findings suggests the public messages emphasizing the risks of COVID-19 should also combat misinformation in order to maximize vaccine uptake.

Regression-Equivalent Effect Sizes for Latent Growth Modeling and Associated Null Hypothesis Significance Tests

The effect of an independent variable on random slopes in growth modeling with latent variables is conventionally used to examine predictors of change over the course of a study. This tutorial demonstrates that the same effect of a covariate on growth can be obtained by using final status centering for parameterization and regressing the random intercepts (or the intercept factor scores) on both the independent variable and a baseline covariate--the framework used to study change with classical regression analysis. Examples are provided that illustrate the application of an intercept-focused approach to obtain effect sizes--the unstandardized regression coefficient, the standardized regression coefficient, squared semi-partial correlation, and Cohen's f2 --that estimate the same parameters as respective effect sizes from a classical regression analysis. Moreover, statistical power to detect the effect of the predictor on growth was greater when using random intercepts than the conventionally used random slopes.

Evaluating Treatment Mechanisms of Varenicline: Mediation by Affect and Craving

INTRODUCTION

Negative reinforcement models posit that relapse to cigarette smoking is driven in part by changes in affect and craving during the quit attempt. Varenicline may aid cessation by attenuating these changes; however, this mediational pathway has not been formally evaluated in placebo-controlled trials. Thus, trajectories of negative affect (NA), positive affect (PA), and craving were tested as mediators of the effect of varenicline on smoking cessation.

AIMS AND METHODS

Secondary data analysis was conducted on 828 adults assigned to either varenicline or placebo in a randomized controlled trial for smoking cessation (NCT01314001). Self-reported NA, PA, and craving were assessed 1-week pre-quit, on the target quit day (TQD), and 1 and 4 weeks post-TQD.

RESULTS

Across time, NA peaked 1-week post-quit, PA did not change, and craving declined. Less steep rises in NA (indirect effect 95% CI: .01 to .30) and lower mean craving at 1-week post-quit (CI: .06 to .50) were mediators of the relationship between varenicline and higher cessation rates at the end of treatment. PA was associated with cessation but was not a significant mediator.

CONCLUSIONS

These results partially support the hypothesis that varenicline improves smoking cessation rates by attenuating changes in specific psychological processes and supported NA and craving as plausible treatment mechanisms of varenicline.

IMPLICATIONS

The present research provides the first evidence from a placebo-controlled randomized clinical trial that varenicline's efficacy is due, in part, to post-quit attenuation of NA and craving. Reducing NA across the quit attempt and craving early into the attempt may be important treatment mechanisms for effective interventions. Furthermore, post-quit NA, PA, and craving were all associated with relapse and represent treatment targets for future intervention development.

Cyber-Victimization Experience among Higher Education Students: Effects of Social Support, Loneliness, and Self-Efficacy

Most of the research literature on cyberbullying (CB) has focused on adolescents, but due to their intensive, unsupervised use of Information Communication Technologies (ICT), higher education students are at high risk of being involved in CB. The current study examined the nature of CB among 1004 higher education students. In addition, we explored the relationships between cyber-victimization, social support, loneliness, and self-efficacy. For that purpose, we applied a path analysis model (PA) to explain the effect of each variable on the cyber-victimization experience, expecting that high levels of loneliness and low levels of self-efficacy will predict cyber-victimization, but might be moderated and reduced by high levels of social support. Results revealed that social support moderated the relationships between these socio-emotional variables and cyber-victimization, and might serve as a protective factor. These findings on young adults may contribute to the understanding of the nature of cyber-victimization throughout the life cycle. Nowadays, academic institutions are facing an uphill effort in trying to restrain online misbehavior. In view of the findings, higher education policy could help facilitate coping with CB through student support and focused intervention programs.

Multidimensional Impulsive Personality Traits Mediate the Effect of Parent Substance Use Disorder on Adolescent Alcohol and Cannabis Use

Theory suggests that behavioral undercontrol mediates the effect of parental substance disorder on offspring substance use, but no studies have tested multidimensional impulsive personality traits as mechanisms of risk. Adolescents (N = 392; 48% female) from a multigenerational study of familial alcohol disorder self-reported impulsive personality traits via the UPPS-P (M_age = 16.09; Range = 13-19) and alcohol/cannabis frequency one year later. The UPPS-P assesses negative and positive urgency (i.e., rash action in a negative or positive mood state), lack of premeditation (i.e., lack of planning/forethought), lack of perseverance (i.e., inability to finish tedious/boring tasks), and sensation seeking (i.e., thrill seeking/risk taking). Parent substance disorder was assessed via diagnostic interviews. Two-part hurdle models tested predictors of any substance use (i.e., binary part) and frequency of use (i.e., continuous part). Parent substance disorder was indirectly associated with any alcohol/cannabis use (binary part) and higher cannabis frequency (continuous part) through negative urgency. Parental substance disorder was associated with higher alcohol frequency through a lack of premeditation. Sensation seeking was associated with any alcohol/cannabis use but unrelated to parental substance disorder. Despite indirect effects, strong effects of parental substance disorder on substance use remained. The findings are discussed in terms of theory and public health implications.

Effects of body weight and fat mass on back pain - direct mechanical or indirect through inflammatory and metabolic parameters?

BACKGROUND

While reports indicate the association between obesity and back pain, its mechanism is still unclear. Thus, we aimed to investigate the effects of weight and its components on back pain in middle-aged women while considering direct mechanical and indirect effects via inflammatory and metabolic parameters.

METHODS

We used data from the Chingford 1000 Women Study, two follow-ups seven years apart. We assessed effects of weight, body mass index (BMI), total fat mass (TFM), total lean mass (TLM) and total bone mineral density (TBMD), measured by dual-energy X-ray absorptiometry, on back pain episode. We used inflammatory (C-reactive protein, interleukin-6, and tumour necrosis factor-alpha) and metabolic parameters (systolic and diastolic blood pressure, triglyceride, high-density lipoprotein cholesterol, and fasting blood glucose) as mediators of indirect effects. We investigated associations of interest cross-sectionally and longitudinally using binary logistic regression and parallel mediation model.

RESULTS

We included 826 Chingford middle-aged women (mean age=60.7, SD=5.9) from the first used follow-up in cross-sectional and mediation analyses and 645 women that attended the follow-up seven years later, in longitudinal analyses. We found that increased weight was directly associated with increased odds of having back pain episode (OR=1.02; 95% CI 1.01-1.03), similarly as BMI (OR=1.05; 95% CI 1.02-1.08) and TFM (OR=1.03; 95% CI 1.01-1.04) consistently across the cross-sectional and longitudinal models, but not TLM or TBMD. However, we did not find consistent indirect effects of weight or its components through measured inflammatory or metabolic parameters on back pain.

CONCLUSIONS

Our results show that in middle-aged women, weight, BMI and TFM are directly related to back pain, indicating prominence of mechanical loading effect.

Mediating Effect of Work Stress on the Associations Between Psychological Job Demands, Social Approval, and Workplace Violence Among Health Care Workers in Sichuan Province of China

Objective: The aim of this study was to investigate the prevalence of workplace violence against health care workers, to explore the combined association of work stress, psychological job demands, and social approval with workplace violence and their respective mechanisms among health care workers. Methods: Using data from the Chinese Sixth National Health Service Survey (NHSS) in 2018 conducted among 1,371 health care workers in Sichuan province of China. A self-administered structured questionnaire was used to collect data on health care workers' socio-demographic and work-related characteristics, work stress, psychological job demands, social approval, and workplace violence. We used structural equation modeling (SEM) to test the hypothesized relationship among the variables. Results: The results showed that a total of 77.0% health care workers were exposed to workplace violence. Work stress was directly related to workplace violence (β = 2.167, 95%CI: 1.707, 2.627), while psychological job demands and social approval had indirect associations with workplace violence via work stress [β = 0.427, 95%CI: 0.297, 0.557; β = -0.787, 95%CI: (-0.941)-(-0.633)]. Both psychological job demands (β = 0.197, 95%CI: 0.139, 0.255) and social approval [β = -0.346, 95%CI: (-0.399)-(-0.294)] had direct associations with work stress, while social approval had direct association with psychological job demands [β = -0.085, 95%CI: (-0.136)-(-0.034)]. Psychological job demands mediated the relationship between social approval and work stress. Conclusion: Overall, decreasing workplace violence among health care workers requires to promote interventions to reduce work stress and psychological job demands by improving social approval.

Air Pollution Impairs Subjective Happiness by Damaging Their Health

Background: The impact of air pollution on residents' happiness remains unclear and the underlying mechanism remains unknown. We aimed to examine the direct effect of air pollution on residents' happiness and indirect effect through mediating their health. Methods: Based on the 2017 China Comprehensive Social Survey Data (CGSS), data on happiness were retrieved from 11,997 residents in 28 provinces in China. An ordered-probit model was used to examine the effect of air pollution on residents' happiness and health, respectively. A stepwise regression was used to derive the direct effect of air pollution on residents' happiness and indirect effect from health impairment attributable to air pollution. Results: Air pollution was associated with lower levels of health (coef. -0.190, 95% CI -0.212, -0.167, p < 0.001), while health was positively associated with happiness (coef. 0.215, 95% CI 0.196, 0.234, p < 0.001). Mediation analysis methods showed that air pollution impacted residents' happiness directly and indirectly: the percent of total effect that was mediated through health was 36.97%, and the ratio of indirect to direct effect was 0.5864. Conclusions: Health plays a major mediating role in the relation between air pollution and residents' happiness. In order to alleviate the impact of air pollution on residents' happiness, future strategies should focus on health promotion besides reducing air pollutant emission.

Parametric-Regression-Based Causal Mediation Analysis of Binary Outcomes and Binary Mediators: Moving Beyond the Rareness or Commonness of the Outcome

In the causal mediation framework, several parametric-regression–based approaches have been introduced in the last decade for estimating natural direct and indirect effects. For a binary outcome, a number of proposed estimators use a logistic model and rely on specific assumptions or approximations that may be delicate or not easy to verify in practice. To circumvent the challenges prompted by the rare outcome assumption in this context, an exact closed-form natural-effects estimator on the odds ratio scale was recently introduced for a binary mediator. In this work, we further push this exact approach and extend it for the estimation of natural effects on the risk ratio and risk difference scales. Explicit formulas for the delta method standard errors are provided. The performance of our proposed exact estimators is demonstrated in simulation scenarios featuring various levels of outcome rareness/commonness. The total effect decomposition property on the multiplicative scales is also examined. Using a SAS macro (SAS Institute, Inc., Cary, North Carolina) we developed, our approach is illustrated to assess the separate effects of exposure to inhaled corticosteroids and placental abruption on low birth weight mediated by prematurity. Our exact natural-effects estimators are found to work properly in both simulations and the real data example.

Non-verbal and verbal parental mentalization as predictors of infant attachment security: Contributions of parental embodied mentalizing and mind-mindedness and the mediating role of maternal sensitivity

Interest in studying the relative contributions of verbal (e.g., maternal mind-mindedness [MM]) and non-verbal dimensions (i.e., parental embodied mentalizing [PEM]) of parental mentalization to child socio-emotional development is relatively recent. To date, only one study has addressed this issue in relation to child attachment security, suggesting a complementary and unique contribution of each one. The purpose of the present study was to further examine the specific contribution of PEM to infant attachment security by considering MM. In addition, this study aimed to explore the mediating role of maternal sensitivity linking PEM, MM to infant attachment security within 110 mother-infant dyads at moderate psychosocial risk. The two dimensions of parental mentalization (PEM and MM) were assessed on the basis of observations made during a videorecorded sequence of mother-child interactions in a context of free play with and without toys when the infants were 8 months old. The Maternal Behavior Q-Sort was used to measure the mothers' sensitivity in a natural setting based on observations of daily mother-child interactions, also when the infants were about 8 months old. Attachment security was measured using The Strange Situation Procedure at infant age 16 months. The results showed positive correlations between maternal sensitivity and both verbal and non-verbal measures of parental mentalization. The mediation analyses first revealed that PEM had a significant indirect effect on attachment security, with sensitivity being identified as a mediator in this association. No indirect effect linking MM and attachment security via sensitivity was observed. These results highlight the contribution of PEM to maternal sensitivity and show maternal sensitivity to be a factor that partly explains the influence of PEM on attachment security in children.

Are neighbourhood restaurants related to frequency of restaurant meals and dietary quality? Prevalence and changes over time in the Multi-Ethnic Study of Atherosclerosis

OBJECTIVE

To examine whether the density of neighbourhood restaurants affected the frequency of eating restaurant meals and subsequently affected diet quality.

DESIGN

Cross-sectional and longitudinal designs. Structural equation models assessed the indirect relationship between restaurant density (≤3 miles (4·8 km) of participant addresses) and dietary quality (Healthy Eating Index 2010 (HEI)) via the frequency of eating restaurant meals, after adjustment for sociodemographics, select health conditions, region, residence duration and area-level income.

SETTING

Urbanised areas in multiple regions of the USA, years 2000-2002 and 2010-2012.

PARTICIPANTS

Participants aged 45-84 years were followed for 10 years (n 3567).

RESULTS

Median HEI (out of 100) was fifty-nine at baseline and sixty-two at follow-up. The cross-sectional analysis found that residing in areas with a high density of restaurants (highest-ranked quartile) was associated with 52 % higher odds of frequently eating restaurant meals (≥3 times/week, OR: 1·52, 95 % CI 1·18, 1·98) and 3 % higher odds of having lower dietary quality (HEI lowest quartile < 54, OR: 1·03, 95 % CI 1·01, 1·06); associations were not sustained in longitudinal analyses. The cross-sectional analysis found 34 % higher odds of having lower dietary quality for those who frequently ate at restaurants (OR: 1·34, 95 % CI 1·12, 1·61), and more restaurant meals (over time increase ≥ 1 time/week) were associated with higher odds of having worse dietary quality at follow-up (OR: 1·21, 95 % CI 1·00, 1·46).

CONCLUSIONS

Restaurant density was associated with frequently eating out in cross-sectional and longitudinal analyses but was associated with the lower dietary quality only in cross-sectional analyses. Frequent restaurant meals were negatively related to dietary quality. Interventions that encourage less frequent eating out may improve population dietary quality.

Effect of Parameterization on Statistical Power and Effect Size Estimation in Latent Growth Modeling

The difference between groups in their random slopes is frequently examined in latent growth modeling to evaluate treatment efficacy. However, when end centering is used for model parameterization with a randomized design, the difference in the random intercepts is the model-estimated mean difference between the groups at the end of the study, which has the same expected value as the product of the coefficient for the slope difference and study duration. A Monte Carlo study found that (a) the statistical power to detect the treatment effect was greater when determined from the intercept instead of the slope difference, and (b) the standard error of the model-estimated mean difference was smaller when obtained from the intercept difference. Investigators may reduce Type II errors by comparing groups in random intercepts instead of random slopes to test treatment effects, and should therefore conduct power assessments using end centering to detect each difference.

Patients With Higher Treatment Outcome Expectations Are More Satisfied With the Results of Nonoperative Treatment for Thumb Base Osteoarthritis: A Cohort Study

OBJECTIVE

To investigate how satisfaction with treatment outcome is associated with patient mindset and Michigan Hand Outcome Questionnaire (MHQ) scores at baseline and 3 months in patients receiving nonoperative treatment for first carpometacarpal joint (CMC-1) osteoarthritis (OA).

DESIGN

Cohort study SETTING: A total of 20 outpatient locations of a clinic for hand surgery and hand therapy in the Netherlands.

PARTICIPANTS

Patients (N=308) receiving nonoperative treatment for CMC-1 OA, including exercise therapy, an orthosis, or both, between September 2017 and February 2019.

INTERVENTIONS

Nonoperative treatment (ie, exercise therapy, an orthosis, or both) MAIN OUTCOME MEASURES: Satisfaction with treatment outcomes was measured after 3 months of treatment. We measured total MHQ score at baseline and at 3 months. As baseline mindset factors, patients completed questionnaires on treatment outcome expectations, illness perceptions, pain catastrophizing, and psychological distress. We used multivariable logistic regression analysis and mediation analysis to identify factors associated with satisfaction with treatment outcomes.

RESULTS

More positive pretreatment outcome expectations were associated with a higher probability of being satisfied with treatment outcomes at 3 months (odds ratio, 1.15; 95% confidence interval, 1.07-1.25). Only a relatively small part (33%) of this association was because of a higher total MHQ score at 3 months. None of the other mindset and hand function variables at baseline were associated with satisfaction with treatment outcomes.

CONCLUSIONS

This study demonstrates that patients with higher pretreatment outcome expectations are more likely to be satisfied with treatment outcomes after 3 months of nonoperative treatment for CMC-1 OA. This association could only partially be explained by a better functional outcome at 3 months for patients who were satisfied. Health care providers treating patients nonoperatively for CMC-1 OA should be aware of the importance of expectations and may take this into account in pretreatment counseling.

Effect of antenatal depression on adverse birth outcomes in Gondar town, Ethiopia: A community-based cohort study

BACKGROUND

The impact of antenatal depression on pregnancy outcomes has been well investigated in developed countries, but few studies have been conducted in low-income countries. As depression is significantly affected by socio-economic and cultural factors, it would be difficult to generalize evidence from high-income countries to low-income countries. We conducted a community-based cohort study to estimate the incidence of adverse birth outcomes and the direct and indirect pathways via which depression and other psychosocial risk factors may impact such birth outcomes within Gondar town, Ethiopia.

METHODS

The study followed 916 pregnant women who were screened for antenatal depression using the Edinburgh Postnatal Depression Scale (EPDS). We also assessed the incidence of preterm births, Low Birth Weight (LBW) and stillbirths. Modified Poisson regression was used to estimate the relative risk of predictors on adverse birth outcomes and a Generalized Structural Equation Model (GSEM) was used to estimate the direct and indirect effect of antenatal depression and other psychological risk factors on adverse birth outcomes.

RESULTS

The cumulative incidence of stillbirth, LBW and preterm was 1.90%, 5.25%, and 16.42%, respectively. The risk of preterm birth was 1.61, 1.46, 1.49, and 1.77 times higher among participants who identified as Muslim, reported being fearful of delivery, were government employee's, and who had no antenatal care services, respectively. Partner support moderated the association between depression, preterm birth, and LBW. Depression had no direct effect on birth outcomes but indirectly affected preterm birth via partner support. Religion had both direct and indirect effects on preterm birth, while occupation and fear of delivery had direct effects. The risk of LBW was 9.44 and 2.19 times higher among preterm births and those who had exposure to tobacco, respectively. Stress coping was indirectly associated, and preterm birth and tobacco exposure were directly associated with LBW. The risk of stillbirth was 3.22 times higher in women with antenatal depression and 73% lower in women with higher coping abilities.

CONCLUSIONS

There was a high incidence of all adverse birth outcomes in Gondar Town. Depression and psychosocial risk factors had important indirect negative effects on risk, while partner support provided a positive indirect effect on the incidence of adverse birth outcomes. Interventions that focus on increasing partner engagement and participation in antenatal support may help reduce adverse birth outcomes by enhancing maternal resilience.

Neighborhood characteristics and transport walking: Exploring multiple pathways of influence using a structural equation modeling approach

In studies of neighborhood effects on transport walking, residential preferences along with other personal characteristics (walking attitude, factors that affect decision to walk for transport) are important factors to consider. However, few studies have examined relationships between neighborhood characteristics and transport walking accounting for a complex suite of personal factors. This study employed a structural equation modeling approach to examine associations between neighborhood characteristics and transport walking behaviors by accounting for residential preferences, self-assessed factors that affect decision to walk for transportation, and socio-demographics; and to examine whether neighborhood effects were modified by walking attitude and residential preferences. Based on a cross-sectional phone and mail survey of 2,848 residents of New York City, Baltimore, Chicago, Los Angeles, St. Paul, and Winston Salem in 2011-12, we found neighborhood characteristics had significant independent associations with transport walking, even after adjusting for residential preferences and other personal factors. Neighborhood effects on transport walking did not differ significantly by walking attitude or residential preferences. Our findings highlight the importance of neighborhood effects in promoting transport walking, and suggest that while residential preferences are important factors, they do not obviate the neighborhood effects on transport walking.

New Approaches for Estimation of Effect Sizes and their Confidence Intervals for Treatment Effects from Randomized Controlled Trials

Although Cohen's d and the growth modeling analysis (GMA) d from linear models are common standardized effect sizes used to convey treatment effects, popular statistical software packages do not include them in their standard outputs. This article demonstrated the use of statistical software with user-prescribed parameter functions (e.g., Mplus) to produce d for treatment effects from both classical analysis and GMA--along with their associated standard errors (SEs) and confidence intervals (CIs). A Monte Carlo study was conducted to examine bias in the SE and CI for GMA d obtained with Mplus and found that both estimates were more accurate when calculated by the software with the standard bootstrap than with the delta method, but the delta method estimates were less biased than respective estimates from extant post hoc equations. Thus, users of many statistical software packages (including SAS, R, and LISREL) should obtain d or GMA d and associated CIs directly. Researchers employing less versatile software--and meta-analysts including ds and GMA ds in their syntheses of treatment effects--should continue to use the conventional post hoc equations. Biases in SEs and CIs for effect sizes obtained with them are ignorable and point estimates of d and GMA d are the same whether obtained directly from the software or with post hoc equations.