Home-based work and childbearing

We examine the timely yet greatly under-researched interplay between home-based work (HBW) and women's birth transitions. Past research has shown that HBW may facilitate and/or jeopardize work-family balance, depending on the worker's family and work circumstances. Following that research, we develop here a theoretical framework on how HBW can facilitate or hinder fertility. Using the UK Household Longitudinal Study 2009-19 and random-effects cloglog regression, we study the link between HBW and first- and second-birth risks. We find that HBW is negatively associated with the transition to motherhood and unrelated to the progression to a second child. We also show that HBW helps to enable women to have children if they would otherwise face a long commute. All in all, our findings do not support the idea that the spread of HBW will lead to an immediate increase in fertility.

Comparing methods for risk prediction of multicategory outcomes: dichotomized logistic regression vs. multinominal logit regression

Background

Medical outcomes of interest to clinicians may have multiple categories. Researchers face several options for risk prediction of such outcomes, including dichotomized logistic regression and multinomial logit regression modeling. We aimed to compare these methods and provide practical guidance needed.

Methods

We described dichotomized logistic regression and competing risks regression, and an alternative to standard multinomial logit regression, continuation-ratio logit regression for ordinal outcomes. We then applied these methods to develop prediction models of survival and growth outcomes based on the NICHD Extremely Preterm Birth Outcome Tool model. The statistical and practical advantages and flaws of these methods were examined and both discrimination and calibration of the estimated models were assessed.

Results

The dichotomized logistic models and multinomial continuation-ratio logit model had similar discrimination and calibration in predicting death and survival without neurodevelopmental impairment. But the continuation-ratio logit model had better discrimination and calibration in predicting probabilities of neurodevelopmental impairment. The sum of predicted probabilities of outcome categories from the logistic models did not equal 100% for about half of the study infants, ranging from 87.7% to 124.0%, and the logistic model of neurodevelopmental impairment greatly overpredicted the risk among low-risk infants and underpredicted among high-risk infants.

Conclusions

Estimating multiple logistic regression models of dichotomized outcomes may result in poorly calibrated predictions. For an outcome with multiple ordinal categories, continuation-ratio logit regression is a useful alternative to standard multinomial logit regression. It produces better calibrated predictions and has the advantages of simplicity in model interpretation and flexibility to include outcome category-specific predictors and random-effect terms for patient heterogeneity by hospital.

Epidemiology of emergency and elective caesarean section and its association with early neonatal mortality in sub-Saharan African countries

INTRODUCTION

Investigating elective and emergency caesarean section (CS) separately is important for a better understanding of birth delivery modes in the sub-Saharan Africa (SSA) region and identifying bottlenecks that prevent favourable childbirth outcomes in SSA. This study aimed at evaluating the prevalences of both CS types, determining their associated socioeconomic factors and their association with early neonatal mortality in SSA.

METHODS

SSA countries Demographic and Health Surveys data that had collected information on the CS' timing were included in our study. A total of 21 countries were included in this study, with a total of 155 172 institutional live births. Prevalences of both CS types were estimated at the countries' level using household sampling weights. Multilevel models were fitted to identify associated socioeconomic factors of both CS types and their associations with early neonatal mortality.

RESULTS

The emergency CS prevalence in SSA countries was estimated at 4.6% (95% CI 4.4-4.7) and was higher than the elective CS prevalence estimated at 3.4% (95% CI 3.3-3.6). Private health facilities' elective CS prevalence was estimated at 10.2% (95% CI 9.3-11.2) which was higher than the emergency CS prevalence estimated at 7.7% (95% CI 7.0-8.5). Conversely, in public health facilities, the emergency CS prevalence was estimated at 4.0% (95% CI 3.8-4.2) was higher than the elective CS prevalence estimated at 2.7% (95% CI 2.6-2.8). The richest women were more likely to have birth delivery by both CS types than normal vaginal delivery. Emergency CS was positively associated with early neonatal mortality (adjusted OR=2.37, 95% CI 1.64-3.41), while no association was found with elective CS.

CONCLUSIONS

Findings suggest shortcomings in pregnancy monitoring, delivery preparation and postnatal care. Beyond antenatal care (ANC) coverage, more attention should be put on quality of ANC, postnatal care, emergency obstetric and newborn care for favourable birth delivery outcomes in SSA.

Modelling representations in speech normalization of prosodic cues

The lack of invariance problem in speech perception refers to a fundamental problem of how listeners deal with differences of speech sounds produced by various speakers. The current study is the first to test the contributions of mentally stored distributional information in normalization of prosodic cues. This study starts out by modelling distributions of acoustic cues from a speech corpus. We proceeded to conduct three experiments using both naturally produced lexical tones with estimated distributions and manipulated lexical tones with f0 values generated from simulated distributions. State of the art statistical techniques have been used to examine the effects of distribution parameters in normalization and identification curves with respect to each parameter. Based on the significant effects of distribution parameters, we proposed a probabilistic parametric representation (PPR), integrating knowledge from previously established distributions of speakers with their indexical information. PPR is still accessed during speech perception even when contextual information is present. We also discussed the procedure of normalization of speech signals produced by unfamiliar talker with and without contexts and the access of long-term stored representations.

Couples Therapy Delivered Through Videoconferencing: Effects on Relationship Outcomes, Mental Health and the Therapeutic Alliance

Changing technology, and the pervasive demand created by a greater need in the population for access to mental health interventions, has led to the development of technologies that are shifting the traditional way in which therapy is provided. This study investigated the efficacy of a behavioral couples therapy program conducted via videoconferencing, as compared to face-to-face. There were 60 participants, in couples, ranging in age from 21 to 69 years old. Couples had been in a relationship for between 1 to 49 years. The 30 couples were randomly allocated to the face-to-face or videoconferencing group. They all took part in Couple CARE—a couples behavioral education program which promotes self-change in order to increase relationship satisfaction. The six session manualized intervention was offered in an identical manner to all clients, in each condition. Data analysis was based on several questionnaires completed by each couple at pre, post and 3-months follow-up. Results showed that therapeutic alliance ratings did not differ between groups, but increased significantly over time for both groups. Additionally, the results indicated improvements in relationship satisfaction, mental health, and all other outcome scores over time, which did not differ based on condition. This study uniquely contributes to the literature exploring the use of technology to provide therapy. Specifically, the study provides evidence for couples therapy via videoconferencing as a viable alternative to face-to-face interventions, especially for those couples who may not have access to the treatment they require. It is anticipated that the results of this study will contribute to the field of online therapy, and add to fostering confidence in agencies to allow expansion of services conducted via videoconferencing.

Cumulative Evidence for the Association of Thrombosis and the Prognosis of COVID-19: Systematic Review and Meta-Analysis

Background

Although thrombosis events have been reported in patients with coronavirus disease 2019 (COVID-19), the association between thrombosis and COVID-19-related critical status or risk of mortality in COVID-19 has been inconsistent.

Objective

We conducted a meta-analysis of reports assessing the association between thrombosis and the prognosis of COVID-19.

Methods

The EMBASE, Ovid-MEDLINE, and Web of Science databases were searched up to December 9, 2021, and additional studies were retrieved via manual searching. Studies were included if they reported the risk of COVID-19-related critical status or COVID-19-related mortality in relation to thrombosis. The related data were extracted by two authors independently, and a random effects model was conducted to pool the odds ratios (ORs). In addition, stratified analyses were conducted to evaluate the association.

Results

Among 6,686 initially identified studies, we included 25 studies published in 2020 and 2021, with a total of 332,915 patients according to predefined inclusion criteria. The associations between thrombosis and COVID-19-related mortality and COVID-19-related critical status were significant, with ORs of 2.61 (95% CI, 1.91–3.55, p < 0.05) and 2.9 (95% CI, 1.6–5.24, p < 0.05), respectively. The results were statistically significant and consistent in stratified analyses.

Conclusions

Thrombosis is associated with an increased risk of mortality and critical status induced by COVID-19. Further prospective studies with large sample sizes are required to establish whether these associations are causal by considering more confounders and to clarify their mechanisms.

Observational studies cannot prove causality. However, autopsy studies show thrombosis events preceding COVID-19-related deaths. The results of this meta-analysis reported that thrombosis was associated with a 161% increased risk of mortality from COVID-19 and a 190% increased risk of COVID-19-related critical status. The type of thrombosis included in the original studies also seemed to be related to the results.

Analysis of the potential for a malaria vaccine to reduce gaps in malaria intervention coverage

Background

The RTS,S/AS01 malaria vaccine is currently being evaluated in a cluster-randomized pilot implementation programme in three African countries. This study seeks to identify whether vaccination could reach additional children who are at risk from malaria but do not currently have access to, or use, core malaria interventions.

Methods

Using data from household surveys, the overlap between malaria intervention coverage and childhood vaccination (diphtheria-tetanus-pertussis dose 3, DTP3) uptake in 20 African countries with at least one first administrative level unit with Plasmodium falciparum parasite prevalence greater than 10% was calculated. Multilevel logistic regression was used to explore patterns of overlap by demographic and socioeconomic variables. The public health impact of delivering RTS,S/AS01 to those children who do not use an insecticide-treated net (ITN), but who received the DTP3 vaccine, was also estimated.

Results

Uptake of DTP3 was higher than malaria intervention coverage in most countries. Overall, 34% of children did not use ITNs and received DTP3, while 35% of children used ITNs and received DTP3, although this breakdown varied by country. It was estimated that there are 33 million children in these 20 countries who do not use an ITN. Of these, 23 million (70%) received the DTP3 vaccine. Vaccinating those 23 million children who receive DTP3 but do not use an ITN could avert up to an estimated 9.7 million (range 8.5–10.8 million) clinical malaria cases each year, assuming all children who receive DTP3 are administered all four RTS,S doses. An additional 10.8 million (9.5–12.0 million) cases could be averted by vaccinating those 24 million children who receive the DTP3 vaccine and use an ITN. Children who had access to or used an ITN were 9–13% more likely to reside in rural areas compared to those who had neither intervention regardless of vaccination status. Mothers’ education status was a strong predictor of intervention uptake and was positively associated with use of ITNs and vaccination uptake and negatively associated with having access to an ITN but not using it. Wealth was also a strong predictor of intervention coverage.

Conclusions

Childhood vaccination to prevent malaria has the potential to reduce inequity in access to existing malaria interventions and could substantially reduce the childhood malaria burden in sub-Saharan Africa, even in regions with lower existing DTP3 coverage.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12936-021-03966-x.

Cumulative evidence for association of rhinitis and depression

Background

Several primary studies evaluated the association between rhinitis and the incidence of depression and yielded inconsistent results. We conducted a meta-analysis of studies evaluating the association between rhinitis and depression.

Methods

We searched the EMBASE, PubMed and Cochrane Library databases for studies published in English before April 1, 2019. The studies were included if they reported any type of rhinitis in relation to depression. Two authors independently extracted the data. The odds ratios (ORs) were pooled using a random-effects model. Stratified analyses were conducted to evaluate the association.

Results

Among the 3472 initially identified studies, we included 14 studies involving a total of 19.36 ± 1.1 million participants according to predefined inclusion criteria. The associations between rhinitis (R), allergic rhinitis (AR), and nonallergic rhinitis (NAR) and depression were significant with ORs of 1.86 (95% CI 1.32 to 2.62, p < 0.05), 1.54 (95% CI 1.24 to 1.90, p < 0.05), and 2.15 (95% CI 1.49 to 3.09, p < 0.05), respectively. The results were consistent and statistically significant in all subgroup analyses.

Conclusions

Rhinitis was associated with an increased risk of depression. Further prospective studies involving large sample sizes are required to confirm the results by considering more confounders and clarify the mechanisms.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13223-021-00615-5.

Reflections on Murray Aitkin's contributions to nonparametric mixture models and Bayes factors

We describe two interesting and innovative strands of Murray Aitkin's research publications, dealing with mixture models and with Bayesian inference. Of his considerable publications on mixture models, we focus on a nonparametric random effects approach in generalized linear mixed modelling, which has proven useful in a wide variety of applications. As an early proponent of ways of implementing the Bayesian paradigm, Aitkin proposed an alternative Bayes factor based on a posterior mean likelihood. We discuss these innovative approaches and some research lines motivated by them and also suggest future related methodological implementations.

Multivariate ordinal random effects models including subject and group specific response style effects

Common random effects models for repeated measurements account for the heterogeneity in the population by including subject-specific intercepts or variable effects. They do not account for the heterogeneity in answering tendencies. For ordinal responses in particular, the tendency to choose extreme or middle responses can vary in the population. Extended models are proposed that account for this type of heterogeneity. Location effects as well as the tendency to extreme or middle responses are modelled as functions of explanatory variables. It is demonstrated that ignoring response styles may affect the accuracy of parameter estimates. An example demonstrates the applicability of the method.

Two-part models for repeatedly measured ordinal data with "don't know" category

Ordinal data (eg, "low," "medium," "high"; graded response on a Likert scale) with an additional "don't know" category are frequently encountered in the medical, social, and behavioral science literature. The handling of a "don't know" option presents unique challenges as it often "destroys" the ordinal nature of the data. Commonly, nominal models are employed which ignore the partial ordering and have a complicated interpretation, especially in situations with repeatedly measured outcomes. We propose two-part models that easily accommodate longitudinal partially ordered (semiordinal) data. The most easily interpretable formulation consists of a random effect logistic submodel for "don't know" vs all the other categories combined, and a random effect ordinal submodel for the ordered categories. Correlated random effects account for statistical dependence within individual. An extension allowing for nonproportionality of odds for the predictor effects in the ordinal submodel is also considered. Maximum likelihood estimation is performed using adaptive Gaussian quadrature in SAS PROC NLMIXED. A simulation study is performed to evaluate the performance of the estimation algorithm in terms of bias and efficiency, and to compare the results of joint and separate models of the two parts, and of proportional and nonproportional model formulations. The methods are motivated and illustrated on a dataset from a study of adolescents' perceptions of nicotine strength of JUUL e-cigarettes. Using the proposed approach we show that adolescents perceive 5% nicotine content as relatively low, a misconception more pronounced among past month nonusers than among past month users of JUUL e-cigarettes.

Adaptive Attention with Consumer Sentinel for Movie Box Office Prediction

To improve the movie box office prediction accuracy, this paper proposes an adaptive attention with consumer sentinel (LSTM-AACS) for movie box office prediction. First, the influencing factors of the movie box office are analyzed. Tackling the problem of ignoring consumer groups in existing prediction models, we add consumer features and then quantitatively analyze and normalize the box office influence factors. Second, we establish an LSTM (Long Short-Term Memory) box office prediction model and inject the attention mechanism to construct an adaptive attention with consumer sentinel for movie box office prediction. Finally, 10,398 pieces of movie box office dataset are used in the Kaggle competition to compare the prediction results with the LSTM-AACS model, LSTM-Attention model, and LSTM model. The results show that the relative error of LSTM-AACS prediction is 6.58%, which is lower than other models used in the experiment.

Polytomous Item Explanatory Item Response Theory Models

This study investigates polytomous item explanatory item response theory models under the multivariate generalized linear mixed modeling framework, using the linear logistic test model approach. Building on the original ideas of the many-facet Rasch model and the linear partial credit model, a polytomous Rasch model is extended to the item location explanatory many-facet Rasch model and the step difficulty explanatory linear partial credit model. To demonstrate the practical differences between the two polytomous item explanatory approaches, two empirical studies examine how item properties explain and predict the overall item difficulties or the step difficulties each in the Carbon Cycle assessment data and in the Verbal Aggression data. The results suggest that the two polytomous item explanatory models are methodologically and practically different in terms of (a) the target difficulty parameters of polytomous items, which are explained by item properties; (b) the types of predictors for the item properties incorporated into the design matrix; and (c) the types of item property effects. The potentials and methodological advantages of item explanatory modeling are discussed as well.

Estimating overdispersion in sparse multinomial data

Multinomial data arise in many areas of the life sciences, such as mark-recapture studies and phylogenetics, and will often by overdispersed, with the variance being higher than predicted by a multinomial model. The quasi-likelihood approach to modeling this overdispersion involves the assumption that the variance is proportional to that specified by the multinomial model. As this approach does not require specification of the full distribution of the response variable, it can be more robust than fitting a Dirichlet-multinomial model or adding a random effect to the linear predictor. Estimation of the amount of overdispersion is often based on Pearson's statistic X² or the deviance D. For many types of study, such as mark-recapture, the data will be sparse. The estimator based on X² can then be highly variable, and that based on D can have a large negative bias. We derive a new estimator, which has a smaller asymptotic variance than that based on X² , the difference being most marked for sparse data. We illustrate the numerical difference between the three estimators using a mark-recapture study of swifts and compare their performance via a simulation study. The new estimator has the lowest root mean squared error across a range of scenarios, especially when the data are very sparse.

Association between vitamin D status and sepsis in children: A meta-analysis of observational studies

BACKGROUND

The consequences of vitamin D deficiency regarding sepsis in children remain controversial. We conducted a meta-analysis of studies evaluating the association between vitamin D status and sepsis in children.

METHODS

We used EMBASE, Ovid Medline and Cochrane Library to conduct a meta-analysis of studies published in English before November 21, 2017.

RESULTS

Among 1146 initially identified studies, we included 13 studies according to predefined inclusion criteria comprising 975 patients and 770 control participants. According to a random effects model, the mean difference in 25(OH)D levels (nmol/L) between participants with sepsis (444) and controls (528) was (mean difference, -18.55; 95% confidence interval (CI), -19.45 to -17.66, p < 0.05). The association between vitamin D deficiency and sepsis was significant, with an odds ratio (OR) = 1.13 (95% CI, 1.18 to 1.50, p < 0.05). Factors that could explain differences in the results include the study location/medical conditions, study design, 25(OH)D assay methods, diagnostic sepsis at different ages, diagnostic criteria for sepsis, and sepsis with comorbidities.

CONCLUSIONS

The association between vitamin D deficiency/lower 25(OH)D levels and sepsis was significant in children and neonates. Further studies are required to confirm the results by considering more confounders.

The intensity order illusion: temporal order of different vibrotactile intensity causes systematic localization errors

Haptic illusions serve as important tools for studying neurocognitive processing of touch and can be utilized in practical contexts. We report a new spatiotemporal haptic illusion that involves mislocalization when the order of vibrotactile intensity is manipulated. We tested two types of motors mounted in a 4 × 4 array in the lower thoracic region. We created apparent movement with two successive vibrotactile stimulations of varying distance (40, 20, or 0 mm) and direction (up, down, or same) while changing the temporal order of stimulation intensity (strong-weak vs. weak-strong). Participants judged the perceived direction of movement in a 2-alternative forced-choice task. The results suggest that varying the temporal order of vibrotactile stimuli with different intensity leads to systematic localization errors: when a strong-intensity stimulus was followed by a weak-intensity stimulus, the probability that participants perceived a downward movement increased, and vice versa. The illusion is so strong that the order of the strength of stimulation determined perception even when the actual presentation movement was the opposite. We then verified this "intensity order illusion" using an open response format where observers judged the orientation of an imaginary line drawn between two sequential tactor activations. The intensity order illusion reveals a strong bias in vibrotactile perception that has strong implications for the design of haptic information systems.NEW & NOTEWORTHY We report a new illusion involving mislocalization of stimulation when the order of vibrotactile intensity is manipulated. When a strong-intensity stimulus follows a weak-intensity stimulus, the probability that participants perceive an upward movement increases, and vice versa. The illusion is so strong that the order of the strength of stimulation determined perception even when the actual presentation movement was the opposite. This illusion is important for the design of vibrotactile stimulation displays.

A joint model for mixed and truncated longitudinal data and survival data, with application to HIV vaccine studies

In HIV vaccine studies, a major research objective is to identify immune response biomarkers measured longitudinally that may be associated with risk of HIV infection. This objective can be assessed via joint modeling of longitudinal and survival data. Joint models for HIV vaccine data are complicated by the following issues: (i) left truncations of some longitudinal data due to lower limits of quantification; (ii) mixed types of longitudinal variables; (iii) measurement errors and missing values in longitudinal measurements; (iv) computational challenges associated with likelihood inference. In this article, we propose a joint model of complex longitudinal and survival data and a computationally efficient method for approximate likelihood inference to address the foregoing issues simultaneously. In particular, our model does not make unverifiable distributional assumptions for truncated values, which is different from methods commonly used in the literature. The parameters are estimated based on the h-likelihood method, which is computationally efficient and offers approximate likelihood inference. Moreover, we propose a new approach to estimate the standard errors of the h-likelihood based parameter estimates by using an adaptive Gauss-Hermite method. Simulation studies show that our methods perform well and are computationally efficient. A comprehensive data analysis is also presented.

Dynamic prediction of repeated events data based on landmarking model: application to colorectal liver metastases data

BACKGROUND

In some clinical situations, patients experience repeated events of the same type. Among these, cancer recurrences can result in terminal events such as death. Therefore, here we dynamically predicted the risks of repeated and terminal events given longitudinal histories observed before prediction time using dynamic pseudo-observations (DPOs) in a landmarking model.

METHODS

The proposed DPOs were calculated using Aalen-Johansen estimator for the event processes described in the multi-state model. Furthermore, in the absence of a terminal event, a more convenient approach without matrix operation was described using the ordering of repeated events. Finally, generalized estimating equations were used to calculate probabilities of repeated and terminal events, which were treated as multinomial outcomes.

RESULTS

Simulation studies were conducted to assess bias and investigate the efficiency of the proposed DPOs in a finite sample. Little bias was detected in DPOs even under relatively heavy censoring, and the method was applied to data from patients with colorectal liver metastases.

CONCLUSIONS

The proposed method enabled intuitive interpretations of terminal event settings.

Protein Intake, Nutritional Status and Outcomes in ICU Survivors: A Single Center Cohort Study

Background: We hypothesized that protein delivery during hospitalization in patients who survived critical care would be associated with outcomes following hospital discharge. Methods: We studied 801 patients, age ≥ 18 years, who received critical care between 2004 and 2012 and survived hospitalization. All patients underwent a registered dietitian formal assessment within 48 h of ICU admission. The exposure of interest, grams of protein per kilogram body weight delivered per day, was determined from all oral, enteral and parenteral sources for up to 28 days. Adjusted odds ratios for all cause 90-day post-discharge mortality were estimated by mixed- effects logistic regression models. Results: The 90-day post-discharge mortality was 13.9%. The mean nutrition delivery days recorded was 15. In a mixed-effect logistic regression model adjusted for age, gender, race, Deyo-Charlson comorbidity index, acute organ failures, sepsis and percent energy needs met, the 90-day post-discharge mortality rate was 17% (95% CI: 6–26) lower for each 1 g/kg increase in daily protein delivery (OR = 0.83 (95% CI 0.74–0.94; p = 0.002)). Conclusions: Adult medical ICU patients with improvements in daily protein intake during hospitalization who survive hospitalization have decreased odds of mortality in the 3 months following hospital discharge.

Maternal chorioamnionitis and neurodevelopmental outcomes in preterm and very preterm neonates: A meta-analysis

Context

No consensus exists regarding the association between maternal chorioamnionitis and neurodevelopmental outcomes in preterm and very preterm neonates.

Objectives

To investigate whether maternal chorioamnionitis affects neurodevelopmental outcomes and to identify the factors that may explain these effects.

Data sources

We used Ovid Medline, EMBASE and Web of Science to conduct a meta-analysis of studies published in English before August 25, 2017, with titles or abstracts that discussed an association between maternal chorioamnionitis and mental/motor development.

Study selection

Among the 603 initially identified studies, we selected those that addressed an association between maternal chorioamnionitis and mental/motor development according to our preselected inclusion criteria as follows: (1) the study compared infants with and without exposure to maternal chorioamnionitis and (2) the neurodevelopmental outcome was followed up using the Bayley Scales of Infant Development 2^nd edition.

Data synthesis

Our meta-analysis included 10 studies. According to a random effect model, infants with maternal chorioamnionitis exposure had poorer mental development (d = -2.25 [95%CI, -4.33, -0.17], p<0.05) than infants without maternal chorioamnionitis, and infants with maternal clinical chorioamnionitis exposure had poorer motor development (d = -2.37 [95%CI, -4.62 to -0.12], p<0.05) than infants without maternal clinical chorioamnionitis exposure. Factors in the meta-analysis that showed differences between the two patient groups included an MDI assessment blinded to medical history, MDI assessment at the correct age, and time of the MDI assessment.

Conclusion

This study suggests that maternal chorioamnionitis may affect mental development in preterm and very preterm neonates, and that maternal clinical chorioamnionitis may affect motor development in offspring. Further studies are required to confirm these results and to detect the influence of variables across studies.

Association between maternal overweight or obesity and cerebral palsy in children: A meta-analysis

CONTEXT

There is no consensus regarding the association between maternal obesity or overweight and cerebral palsy (CP) in children.

OBJECTIVES

To investigate whether maternal obesity or overweight is associated with CP and identify the factors that explain the differences in the study results.

DATA SOURCES

We conducted a meta-analysis of studies published in English with titles or abstracts that discussed the relationships between maternal obesity or overweight and CP before August 23, 2017, using Ovid Medline, EMBASE and Web of Science.

STUDY SELECTION

Of 2699 initially identified studies, 8 studies that addressed the association between maternal obesity and CP met our final inclusion criteria.

DATA EXTRACTION

Information from the individual studies was abstracted using standardized forms by 2 independent observers who were blinded to the authors' names and journal titles.

DATA SYNTHESIS

According to a random effects model, maternal overweight was significantly associated with CP in offspring [RR = 1.29 (95% CI, 1.04-1.60), heterogeneity (I2 = 45.5%, P = 0.103)]; maternal obesity was significantly associated with CP in offspring [RR = 1.45 (95% CI, 1.25-1.69), heterogeneity (I2 = 24.1%, P = 0.253)]; and maternal obesity III was significantly associated with CP in offspring [RR = 2.25 (95% CI, 1.82-2.79), heterogeneity (I2 = 0%, P = 0.589)]. However, maternal underweight was not significantly associated with CP in offspring [RR = 1.11 (95% CI, 0.88-1.38), low heterogeneity (I2 = 0%, P = 0.435)]. Factors that explained the differences in the meta-analysis results included study design, study location, and whether individual studies adjusted for potential confounders.

CONCLUSION

This study suggests that maternal obesity and overweight increase the risk of CP in offspring. Further studies are required to confirm these results and determine the influence of variables across studies.

Association between bronchopulmonary dysplasia and cerebral palsy in children: a meta-analysis

OBJECTIVE

To investigate the association between bronchopulmonary dysplasia (BPD) and the risk of cerebral palsy (CP) in children.

DATA SOURCES

We used EMBASE, PubMed and Web of Science to conduct a meta-analysis of studies published before 1 September 2017, written in English whose titles or abstracts discussed an association between BPD and CP.

STUDY SELECTION

Observational studies, for example, case-control and cohort studies were included.

DATA EXTRACTION AND SYNTHESIS

All review stages were conducted by two reviewers independently. Data synthesis was undertaken via meta-analysis of available evidence.

MAIN OUTCOMES AND MEASURES

The prevalence of developing CP was measured after exposure to BPD.

RESULTS

Among 1234 initially identified studies, we selected those that addressed an association between BPD and CP according to our preselected inclusion criteria. Our meta-analysis included 11 studies. According to a random effect model, BPD was significantly associated with CP (ORs 2.10; 95% CI 1.57 to 2.82) in preterm infants. Factors explaining differences in the study results included study design, the definition of BPD, the time of diagnosis of CP and whether the studies adjusted for potential confounders.

CONCLUSION

This study suggests that BPD is a risk factor for CP. Further studies are required to confirm these results and to detect the influence of variables across studies.

The association between vitamin D status and tuberculosis in children: A meta-analysis

BACKGROUND

Vitamin D deficiency (VDD) has been implicated in the pathogenesis of tuberculosis (TB), but most studies have not reported a significant association. We conducted a meta-analysis to explore the association between vitamin D status and TB in children.

METHODS

Web of Science, Ovid Medline, and EMBASE were searched for studies in English that discussed vitamin D status and TB in children before January 22, 2018.

RESULTS

From the 585 initially identified studies, we selected those that addressed an association between vitamin D status and TB according to our preselected inclusion criteria. Our meta-analysis included 10 studies. According to the random effects model, TB was significantly associated with VDD (ORs, 1.70; 95% CI, 1.20-2.42; P < .05) in children. Vitamin D levels were significantly lower in TB patients than in controls, with a mean difference d = -5.49 nmol/L (95% CI, -10.42 to -0.55; P < .05), indicating that VDD was significantly associated with TB (OR, 1.78; 95% CI, 1.30-2.44; P < .05) in children.

CONCLUSION

This study suggests that vitamin D levels are significantly lower in children with TB/latent TB infection than in controls. TB may contribute to VDD in children. Therefore, VDD may be associated with TB in children.

Leveling up the analysis of the reminiscence bump in autobiographical memory: A new approach based on multilevel multinomial models

In many studies of autobiographical memory, participants are asked to generate more than one autobiographical memory. The resulting data then have a hierarchical or multilevel structure, in the sense that the autobiographical memories (Level 1) generated by the same person (Level 2) tend to be more similar. Transferred to an analysis of the reminiscence bump in autobiographical memory, at Level 1 the prediction of whether an autobiographical memory will fall within the reminiscence bump is based on the characteristics of that memory. At Level 2, the prediction of whether an individual will report more autobiographical memories that fall in the reminiscence bump is based on the characteristics of the individual. We suggest a multilevel multinomial model that allows for analyzing whether an autobiographical memory falls in the reminiscence bump at both levels of analysis simultaneously. The data come from 100 older participants who reported up to 33 autobiographical memories. Our results showed that about 12% of the total variance was between persons (Level 2). Moreover, at Level 1, memories of first-time experiences were more likely to fall in the reminiscence bump than were emotionally more positive memories. At Level 2, persons who reported more emotionally positive memories tended to report fewer memories from the life period after the reminiscence bump. In addition, cross-level interactions showed that the effects at Level 1 partly depended on the Level 2 effects. We discuss possible extensions of the model we present and the meaning of our findings for two prominent explanatory approaches to the reminiscence bump, as well as future directions.

Childhood Sexual Abuse and Early Timing of Puberty

PURPOSE

The purpose was to examine whether the timing of puberty, indexed by breast development and pubic hair development, was earlier for sexually abused females compared with a matched comparison group of nonabused females, controlling for key alternative confounds.

METHODS

A cohort of sexually abused females and matched comparisons was followed longitudinally at mean ages 11 through 20 years. Sexually abused participants (N = 84) were referred by protective services. Comparison participants (N = 89) were recruited to be comparable in terms of age, ethnicity, income level, family constellation, zip codes, and nonsexual trauma histories. Stage of puberty was indexed at each assessment by nurse and participant ratings of breast and pubic hair development using Tanner staging-the gold standard for assessing pubertal onset and development. Cumulative logit mixed models were used to estimate the association between sexual abuse status and the likelihood of transitioning from earlier to later Tanner stage categories controlling for covariates and potential confounds.

RESULTS

Sexual abuse was associated with earlier pubertal onset: 8 months earlier for breasts (odds ratio: 3.06, 95% CI: 1.11-8.49) and 12 months earlier for pubic hair (odds ratio: 3.49, 95% CI: 1.34-9.12). Alternative explanations including ethnicity, obesity, and biological father absence did not eradicate these findings.

CONCLUSIONS

This study confirms an association between exposure to childhood sexual abuse and earlier pubertal onset. Results highlight the possibility that, due to this early onset, sexual abuse survivors may be at increased risk for psychosocial difficulties, menstrual and fertility problems, and even reproductive cancers due to prolonged exposure to sex hormones.

Generalized log-linear models with random effects, with application to smoothing contingency tables

We define a class of generalized log-linear models with random effects. For a vector of Poisson or multinomial means m and matrices of constants C and A, the model has the form C log Aμ = Xβ + Zu, where β are fixed effects and u are random effects. The model contains most standard models currently used for categorical data analysis. We suggest some new models that are special cases of this model and are useful for applications such as smoothing large contingency tables and modeling heterogeneity in odds ratios. We present examples of its use for such applications. In many cases, maximum likelihood model fitting can be handled with existing methods and software. We outline extensions of model fitting methods for other cases. We also summarize several challenges for future research, such as fitting the model in its most general form and deriving properties of estimates used in smoothing contingency tables.

Random effect models for repeated measures of zero-inflated count data

For count responses, the situation of excess zeros (relative to what standard models allow) often occurs in biomedical and sociological applications. Modeling repeated measures of zero-inflated count data presents special challenges. This is because in addition to the problem of extra zeros, the correlation between measurements upon the same subject at different occasions needs to be taken into account. This article discusses random effect models for repeated measurements on this type of response variable. A useful model is the hurdle model with random effects, which separately handles the zero observations and the positive counts. In maximum likelihood model fitting, we consider both a normal distribution and a nonparametric approach for the random effects. A special case of the hurdle model can be used to test for zero inflation. Random effects can also be introduced in a zero-inflated Poisson or negative binomial model, but such a model may encounter fitting problems if there is zero deflation at any settings of the explanatory variables. A simple alternative approach adapts the cumulative logit model with random effects, which has a single set of parameters for describing effects. We illustrate the proposed methods with examples.

On exchangeable multinomial distributions

We derive an expression for the joint distribution of exchangeable multinomial random variables, which generalizes the multinomial distribution based on independent trials while retaining some of its important properties. Unlike de Finneti's representation theorem for a binary sequence, the exchangeable multinomial distribution derived here does not require that the finite set of random variables under consideration be a subset of an infinite sequence. Using expressions for higher moments and correlations, we show that the covariance matrix for exchangeable multinomial data has a different form from that usually assumed in the literature, and we analyse data from developmental toxicology studies. The proposed analyses have been implemented in R and are available on CRAN in the CorrBin package.

A STATISTICAL MODEL TO ASSESS (ALLELE-SPECIFIC) ASSOCIATIONS BETWEEN GENE EXPRESSION AND EPIGENETIC FEATURES USING SEQUENCING DATA

Sequencing techniques have been widely used to assess gene expression (i.e., RNA-seq) or the presence of epigenetic features (e.g., DNase-seq to identify open chromatin regions). In contrast to traditional microarray platforms, sequencing data are typically summarized in the form of discrete counts, and they are able to delineate allele-specific signals, which are not available from microarrays. The presence of epigenetic features are often associated with gene expression, both of which have been shown to be affected by DNA polymorphisms. However, joint models with the flexibility to assess interactions between gene expression, epigenetic features and DNA polymorphisms are currently lacking. In this paper, we develop a statistical model to assess the associations between gene expression and epigenetic features using sequencing data, while explicitly modeling the effects of DNA polymorphisms in either an allele-specific or nonallele-specific manner. We show that in doing so we provide the flexibility to detect associations between gene expression and epigenetic features, as well as conditional associations given DNA polymorphisms. We evaluate the performance of our method using simulations and apply our method to study the association between gene expression and the presence of DNase I Hypersensitive sites (DHSs) in HapMap individuals. Our model can be generalized to exploring the relationships between DNA polymorphisms and any two types of sequencing experiments, a useful feature as the variety of sequencing experiments continue to expand.

Fast and highly efficient pseudo-likelihood methodology for large and complex ordinal data

In longitudinal studies, continuous, binary, categorical, and survival outcomes are often jointly collected, possibly with some observations missing. However, when it comes to modeling responses, the ordinal ones have received less attention in the literature. In a longitudinal or hierarchical context, the univariate proportional odds mixed model (POMM) can be regarded as an instance of the generalized linear mixed model (GLMM). When the response of the joint multivariate model encompass ordinal responses, the complexity further increases. An additional problem of model fitting is the size of the collected data. Pseudo-likelihood based methods for pairwise fitting, for partitioned samples and, as introduced in this paper, pairwise fitting within partitioned samples allow joint modeling of even larger numbers of responses. We show that that pseudo-likelihood methodology allows for highly efficient and fast inferences in high-dimensional large datasets.

IMPAIRED FERTILITY AND PERCEIVED DIFFICULTIES CONCEIVING IN GHANA: MEASUREMENT PROBLEMS AND PROSPECTS

What is the most appropriate measure of impaired fertility for understanding its social consequences in sub-Saharan Africa? The dearth of subjective measures in surveys in the region has prevented comparisons of subjective and objective measures. Perceived difficulties conceiving may have a greater impact than objective measures for social outcomes such as divorce, stigmatization and distress. This study compares 12- (clinical) and 24- (epidemiological) month measures from biomedicine and 5- and 7-year measures from demography with a subjective measure of impaired fertility using correlations, random effects models and test–retest models to assess relationships between measures, their association with sociodemographic characteristics and the stability of measures across time. Secondary panel data (1998–2004) from 1350 Ghanaian women aged 15–49 of all marital statuses are used. Longer waiting times to identification of impaired fertility required by demographic measures result in more stable measures, but perceived difficulties conceiving are most closely aligned with clinical infertility (r=0.61; p<0.05). Epidemiological infertility is also closely aligned with the subjective measure. A large proportion of those identified as having impaired fertility based purely on waiting times are successful contraceptors. Where subjective measures are not available, epidemiological (24-month) measures may be most appropriate for studies of the social consequences of impaired fertility. Accounting for contraceptive use is important in order to avoid false positives. Future research should consider a variety of measures of perceived difficulties conceiving and self-identified infertility to assess which is most valid; in order to accomplish this, it is imperative that subjective measures of infertility be included in social surveys in sub-Saharan Africa.

Managing nonignorable missing data with clustered multinomial responses

Clustered multinomial responses are common in public health studies. In this situation, the baseline logit random effects model is usually suggested as a general modelling approach. When nonignorable missing outcomes exist, naïve methods such as complete case analysis or likelihood methods ignoring missing information may distort the conclusions that are drawn. While methods to deal with binary and ordinal outcomes have been proposed, no easily implementable method is specifically available for missing clustered nominal responses. Joint modelling is usually one of the available choices but has high complexity in terms of likelihood. The numerical integration of both missing data and random effects is challenging. In this study, we have derived a closed form of likelihood. A simplified likelihood is also proposed, which is an extension of a previous study. One advantage is that both methods are easily implemented with commonly used software. We illustrate our proposed methods using the Global Youth Tobacco Survey and compare the results obtained by naïve methods that ignore missing data with the results obtained using the proposed methods. Our approaches restore the parameter estimates and predicted probability of each category to an acceptable extent. Analysis guidelines for the use of our methods are provided.

Fashion versus perception: the impact of surface lightness on the perceived dimensions of interior space

OBJECTIVES

We compare expert opinion with perceptual judgment regarding the influence of color on the perceived height and width of interior rooms.

BACKGROUND

We hypothesize that contrary to popular belief, ceiling and wall lightness have additive effects on perceived height, whereas the lightness contrast between these surfaces is less important. We assessed the intuitions of architectural experts as to which surface colors maximize apparent height and compared these intuitions with psychophysical height and width estimates for rooms differing in ceiling, floor, and wall lightness.

METHOD

Experiment 1 was a survey of architectural experts and nonexperts. Experiments 2 and 3 presented virtual rooms varying in physical height, physical width, and surface lightness.

RESULTS

In Experiment 1, both experts and nonexperts erroneously assumed that the lightness contrast between ceiling and walls influences perceived height Experiment 2 showed that the lightness contrast does not determine apparent height but that ceiling and wall lightness have additive effects. Experiment 3 demonstrated a decrease in perceived width with physical height, whereas the perceived height was not related to physical width. Apparent width was unaffected by ceiling lightness.

CONCLUSION

Light ceiling and light walls make a room appear higher, whereas floor color has a weaker effect. We also found evidence for an asymmetric interaction between height and width.

APPLICATION

The question of how to color walls and ceiling to maximize the apparent size of a room can be answered empirically. Aesthetic considerations may interfere with the correct assessment of the effects of color in experts.

Transitional ideal point models for longitudinal multinomial outcomes

For the analysis of longitudinal data, three families of models are generally distinguished: the marginal, the transitional and the subject-specific family. In this paper, we will propose a transitional model for the analysis of change for a nominal response variable. Such an analysis is often hampered by the dimensionality of the problem. We use multidimensional scaling techniques, more specifically the ideal point model, in order to reduce the dimensionality. The model can handle pure transitional data but also allows for explanatory variables. Two empirical examples will be discussed in order to illustrate all the virtues of the model.

Estimating time-to-event from longitudinal ordinal data using random-effects Markov models: application to multiple sclerosis progression

Longitudinal ordinal data are common in many scientific studies, including those of multiple sclerosis (MS), and are frequently modeled using Markov dependency. Several authors have proposed random-effects Markov models to account for heterogeneity in the population. In this paper, we go one step further and study prediction based on random-effects Markov models. In particular, we show how to calculate the probabilities of future events and confidence intervals for those probabilities, given observed data on the ordinal outcome and a set of covariates, and how to update them over time. We discuss the usefulness of depicting these probabilities for visualization and interpretation of model results and illustrate our method using data from a phase III clinical trial that evaluated the utility of interferon beta-1a (trademark Avonex) to MS patients of type relapsing-remitting.

Ranking of simultaneously presented choice options in animal preference experiments

We considered experiments where animals chose one of all possible simultaneously presented options. The animals might be observed at repeated occasions. In the ethological literature the analysis is often focused on testing the global hypothesis of no difference in preferences by non-parametric methods. This fails to address the estimation of a ranking. Often this approach cannot adequately reflect the experimental setting and the repeated measurement structure. Therefore, we propose to model the choice probabilities for the options with a multinomial logistic model. The correlation induced by repeated measurements is incorporated by animal specific random intercepts. The ranking of the options is taken as the order of the choice probabilities. Adopting a Bayesian approach samples from the posterior distribution of the choice probabilities provide directly samples from the posterior of the rankings. Based on this an estimate of the ranking and description of its variability can be derived. The computation was performed via Markov chain Monte Carlo sampling and was implemented using WinBUGS. We illustrate our approach with an experiment to determine the preference of pigs for three different rooting materials. The proposed method allowed deriving an overall ranking for different combinations of the materials and the spatial positioning.