Combining directed acyclic graphs and the change-in-estimate procedure as a novel approach to adjustment-variable selection in epidemiology

The Impact of Pregnancy Readiness on Lupus Activity, Maternal Mental Health, and Pregnancy Outcomes

OBJECTIVE

Among individuals with systemic lupus erythematosus (SLE) who became pregnant, we explored the impact of medical readiness for pregnancy and personal readiness for pregnancy on the following aspects of maternal health: (1) provider-reported disease activity, (2) patient-perceived disease activity, (3) mood symptoms, (4) pregnancy-related health behaviors, and (5) pregnancy outcomes.

METHODS

All study participants were enrolled in a prospective registry, met Systemic Lupus Collaborating Clinics (SLICC) criteria for SLE, and had at least one pregnancy. Patient-reported outcomes were collected at the first rheumatology visit during pregnancy. "Medically ready" for pregnancy was defined as (1) <1 g of proteinuria, (2) no rheumatic teratogens at conception, and (3) continuing pregnancy-compatible SLE medications after conception. "Personally ready" was defined as planned pregnancy based on a London Measure of Unplanned Pregnancy score ≥10. Multivariable logistic regression models estimated the association of pregnancy readiness with each outcome of interest.

RESULTS

Among the 111 individuals enrolled, lack of medical readiness for pregnancy was associated with significantly higher rates of active disease and worse pregnancy outcomes; however, these patients did not perceive themselves as having higher disease activity. Lack of personal readiness for pregnancy was associated with significantly higher patient-perceived disease activity. Although medical readiness did not impact depressive symptoms substantially, lack of personal readiness for pregnancy was associated with much higher maternal depressive symptoms.

CONCLUSION

To improve pregnancy outcomes among individuals with SLE, greater focus is needed on improving medical optimization before conception. For maternal mental health and quality of life, greater focus is needed on decreasing the incidence of unplanned pregnancy.

Investigating the uses of machine learning algorithms to inform risk factor analyses: The example of avian infectious bronchitis virus (IBV) in broiler chickens

Infectious bronchitis virus (IBV) is a contagious coronavirus causing respiratory and urogenital disease in chickens and is responsible for significant economic losses for both the broiler and table egg layer industries. Despite IBV being regularly monitored using standard epidemiologic surveillance practices, knowledge and evidence of risk factors associated with IBV transmission remain limited. The study objective was to compare risk factor modeling outcomes between a traditional stepwise variable selection approach and a machine learning-based random forest Boruta algorithm using routinely collected IBV antibody titer data from broiler flocks. IBV antibody sampling events (n = 1111) from 166 broiler sites between 2016 and 2021 were accessed. Ninety-two geospatial-related and poultry-density variables were obtained using a geographic information system and data sets from publicly available sources. Seventeen and 27 candidate variables were screened to potentially have an association with elevated IBV antibody titers according to the manual selection and machine learning algorithm, respectively. Selected variables from both methods were further investigated by construction of multivariable generalized mixed logistic regression models. Six variables were shortlisted by both screening methods, which included year, distance to urban areas, main roads, landcover, density of layer sites and year, however, final models for both approaches only shared year as an important predictor. Despite limited significance of clinical outcomes, this work showcases the potential of a novel explorative modeling approach in combination with often unutilized resources such as publicly available geospatial data, surveillance health data and machine learning as potential supplementary tools to investigate risk factors related to infectious diseases.

Is social deprivation associated with the peritoneal dialysis outcomes? A cohort study with REIN registry data

BACKGROUND

Social deprivation is associated with lower peritoneal dialysis (PD) uptake. This study was carried out to evaluate the role of social deprivation on the outcome of PD.

METHODS

This was a retrospective study of data extracted from the Renal Epidemiology and Information Network registry for patients older than 18 years who started PD in metropolitan France between 1 January 2017 and 30 June 2018. The end of the observation period was 31 December 2020. The exposure was the European Deprivation Index calculated using the patient's address. The events of interest were death, transfer to haemodialysis (HD), transplantation and the composite event of death or transfer to HD. A Cox model and Fine and Gray model were used for the analysis.

RESULTS

A total of 1581 patients were included, of whom 418 (26.5%) belonged to Quintile 5 of the European Deprivation Index (the most deprived patients). In the Cox model, the most deprived subjects did not have a greater risk of death (cause-specific hazard ratio (cs-HR): 0.76 [95% confidence interval (CI): 0.53-1.10], transfer to HD (cs-HR 1.37 [95% CI: 0.95-1.98]) or the composite event of death or transfer to HD (cs-HR: 1.08 [95% CI: 0.84-1.38]) or a lower risk of kidney transplantation (cs-HR: 0.73 [95% CI: 0.48-1.10]). In the competing risk analysis, the most deprived subjects had a higher risk of transfer to HD (subdistribution hazard ratio (sd-HR): 1.54 [95% CI: 1.08-2.19]) and lower access to kidney transplantation (sd-HR: 0.68 [0.46-0.99]).

CONCLUSION

In PD patients, social deprivation was not associated with death or the composite event of death or transfer to HD. Socially deprived individuals had a greater risk of transfer to HD and lower access to kidney transplantation in the competing risk analysis.

Progression of C-reactive protein from birth through preadolescence varies by mode of delivery

Introduction

Delivery via caesarean section (C-section) has been associated with an increased risk of childhood chronic diseases such as obesity and asthma, which may be due to underlying systemic inflammation. However, the impact of specific C-section types may be differential, as emergency C-sections typically involve partial labor and/or membrane rupture. Our objectives were to determine if mode of delivery associates with longitudinal profiles of high sensitivity CRP (hs-CRP) -a marker of systemic inflammation-from birth through preadolescence, and to examine if CRP mediates the association between mode of delivery and preadolescent body mass index (BMI).

Methods

Data from the WHEALS birth cohort (N = 1,258) were analyzed; 564 of the 1,258 children in the cohort had data available for analysis. Longitudinal plasma samples (birth through 10-years of age) from 564 children from were assayed for hs-CRP levels. Maternal medical records were abstracted to obtain mode of delivery. Growth mixture models (GMMs) were used to determine classes of hs-CRP trajectories. Poisson regression with robust error variance was used to calculate risk ratios (RRs).

Results

Two hs-CRP trajectory classes were identified: class 1 (76% of children) was characterized by low hs-CRP, while class 2 (24% of children) was characterized by high and steadily increasing hs-CRP. In multivariable models, children delivered via planned C-section had 1.15 times higher risk of being in hs-CRP class 2, compared to vaginal deliveries (p = 0.028), while no association was found for unplanned C-section deliveries [RR (95% CI) = 0.96 (0.84, 1.09); p = 0.49]. Further, the effect of planned C-section on BMI z-score at age 10 was significantly mediated by hs-CRP class (percent mediated = 43.4%).

Conclusions

These findings suggest potentially beneficial effects of experiencing partial or full labor, leading to a lower trajectory of systemic inflammation throughout childhood and decreased BMI during preadolescence. These findings may have implications for chronic disease development later in life.

Preliminary study of the interactive effects of coronary heart disease and lacunar infarction on renal function in patients with type 2 diabetes mellitus by gender

BACKGROUND

Coronary heart disease (CHD) and lacunar infarction (LI) are the most common cardio- cerebrovascular complications of type 2 diabetes mellitus (T2DM) and a recognized risk factor for renal injury. Although a unidirectional association of CHD or LI with T2DM or the kidney has been demonstrated, however, it remains unknown whether there is an interactive effect of the coexistence of CHD and LI on renal function in T2DM patients. The aim of our study was to investigate the interaction between CHD and LI on renal function in gender-specific patients with T2DM and the association between cardio-cerebrovascular disease-related conventional serum markers and the estimated glomerular filtration rate (eGFR).

METHODS

We conducted a cross-sectional study in Beijing and Tianjin from April 2019 to August 2021. Participants with T2DM aged ≥18 years were asked to complete a one-to-one questionnaire and physical examination.

RESULTS

In this study, 389 eligible patients with T2DM were included, with a mean age of 63.04 ± 9.41 years, of whom 200 (51.41 %) were male. The proportions of patients with CHD, LI, and both CHD and LI were 28.53 %, 24.42 %, and 11.05 %, respectively. Compared to T2DM patients without either CHD or LI, those with both CHD and LI were found to have a significantly greater risk of reduced eGFR (OR: 12.82, 95 % CI 5.06-32.52, P < 0.001) than those with CHD alone (OR: 2.42, 95 % CI 1.37-3.00, P = 0.004) or LI alone (OR: 1.15, 95 % CI 0.61-2.18, P = 0.664). The combined presence of CHD and LI is associated with a significantly greater risk of decreased eGFR in female T2DM patients compared to their male counterparts. We found both multiplicative and additive effects in all T2DM patients; however, when stratified by sex, only multiplicative effects were observed. After controlling for interference from CHD, LI, and age, we found that total cholesterol (TC) was negatively correlated with eGFR in females (r = -0.156, P = 0.034), and low-density lipoprotein cholesterol (LDL-C) was negatively correlated with eGFR in males (r = -0.229, P = 0.001).

CONCLUSION

This study provides novel evidence that the synergistic effect of CHD and LI on renal injury in patients with T2DM is significantly greater than their individual effects. Women with T2DM who have both CHD and LI are at a 4.85-fold higher risk of decreased eGFR than men. Therefore, increased clinical attention should be given to preventing and treating vascular complications in T2DM patients, as well as aggressively reducing lipid levels, particularly TC and LDL-C, to delay or prevent renal dysfunction in T2DM patients.

Maternal exposure to phthalates and total gestational weight gain in the LIFECODES birth cohort

Excessive gestational weight gain contributes to adverse maternal and neonatal outcomes. Environmental exposures such as phthalates may lead to metabolic dysregulation, and studies suggest possible associations between maternal phthalate exposure and altered gestational weight gain. We assessed the association between nine maternal phthalate metabolites and measures of total gestational weight gain (pre-pregnancy to median 35.1 weeks of gestation) in a case-control study nested within LIFECODES (N = 379), a prospective birth cohort from Boston, Massachusetts (2006-2008). Our primary outcome was total gestational weight gain z score, a measure independent of gestational age that can provide a less biased estimate of this association. Our secondary outcomes were total gestational weight gain, rate of gestational weight gain, and adequacy ratio. The results were stratified by pre-pregnancy body mass index category. We found that concentrations of mono-(3-carboxypropyl) phthalate (MCPP) and mono-n-butyl phthalate (MBP) were positively associated with total gestational weight gain z scores among participants with obesity: adjusted mean difference (95% Confidence Interval [CI]) = 0.242 (0.030 - 0.455) and 0.105 (-0.002 - 0.212) corresponding to an excess weight gain of 1.81 kg and 0.77 kg at 35 weeks of gestation per interquartile range-increase in MCPP and MBP, respectively. Also, among participants with obesity, MBP demonstrated a potential non-linear relationship with gestational weight gain in cubic spline models. These findings suggest that phthalates may be related to higher gestational weight gain, specifically, among individuals with pre-pregnancy obesity. Future research should investigate whether pregnant people with obesity represent a subpopulation with sensitivity to phthalate exposures.

Alcohol intake and the risk of epithelial ovarian cancer

PURPOSE

To investigate the association between alcohol intake over the lifetime and the risk of overall, borderline, and invasive ovarian cancer.

METHODS

In a population-based case-control study of 495 cases and 902 controls, conducted in Montreal, Canada, average alcohol intake over the lifetime and during specific age periods were computed from a detailed assessment of the intake of beer, red wine, white wine and spirits. Multivariable logistic regression was used to estimate odds ratios (ORs) with 95% confidence intervals (CIs) for the association between alcohol intake and ovarian cancer risk.

RESULTS

For each one drink/week increment in average alcohol intake over the lifetime, the adjusted OR (95% CI) was 1.06 (1.01-1.10) for ovarian cancer overall, 1.13 (1.06-1.20) for borderline ovarian cancers and 1.02 (0.97-1.08) for invasive ovarian cancers. This pattern of association was similarly observed for alcohol intake in early (15- < 25 years), mid (25- < 40 years) and late adulthood (≥ 40 years), as well as for the intake of specific alcohol beverages over the lifetime.

CONCLUSIONS

Our results support the hypothesis that a higher alcohol intake modestly increases the risk of overall ovarian cancer, and more specifically, borderline tumours.

Impact of Cesarean Delivery and Breastfeeding on Secretory Immunoglobulin A in the Infant Gut Is Mediated by Gut Microbiota and Metabolites

How gut immunity in early life is shaped by birth in relation to delivery mode, intrapartum antibiotic prophylaxis (IAP) and labor remains undetermined. We aimed to address this gap with a study of secretory Immunoglobulin A (SIgA) in the infant gut that also tested SIgA-stimulating pathways mediated by gut microbiota and metabolites. Among 1017 Canadian full-term infants, gut microbiota of fecal samples collected at 3 and 12 months were profiled using 16S rRNA sequencing; C. difficile was quantified by qPCR; fecal metabolites and SIgA levels were measured by NMR and SIgA enzyme-linked immunosorbent assay, respectively. We assessed the putative causal relationships from birth events to gut microbiota and metabolites, and ultimately to SIgA, in statistical sequential mediation models, adjusted for maternal gravida status in 551 infants. As birth mode influences the ability to breastfeed, the statistical mediating role of breastfeeding status and milk metabolites was also evaluated. Relative to vaginal birth without maternal IAP, cesarean section (CS) after labor was associated with reduced infant gut SIgA levels at 3 months (6.27 vs. 4.85 mg/g feces, p < 0.05); this association was sequentially mediated through gut microbiota and metabolites of microbial or milk origin. Mediating gut microbiota included Enterobacteriaceae, C. difficile, and Streptococcus. The milk or microbial metabolites in CS-SIgA mediating pathways were galactose, fucose, GABA, choline, lactate, pyruvate and 1,2-propanediol. This cohort study documented the impact of birth on infant gut mucosal SIgA. It is the first to characterize gut microbe-metabolite mediated pathways for early-life SIgA maturation, pathways that require experimental verification.

One-year follow-up of depression, anxiety, and quality of life of Peruvian patients who survived COVID-19

PURPOSE

To assess health-related quality of life (HRQoL) and its associated factors in patients who survived COVID-19 and to assess a prospective evaluation of the prevalence and severity of their depression and anxiety symptoms.

METHODS

We followed up a sample of hospitalized patients who survived COVID-19 at 3 and 12 months after discharge. We assessed HRQoL (Euroqol-5D-5L) through telephone interviews. Any problem in any dimension of Euroqol-5D-5L was considered as low HRQoL. The depression and anxiety symptoms were measured using the Patient Health Questionnaire-9 and Generalized Anxiety Disorder-7 tools, respectively. We estimated the adjusted prevalence ratios (aPR) to low HRQoL using Poisson regression and the changes on their depression and anxiety symptoms during the follow-up.

RESULTS

We included 119 patients with a mean follow-up time of 363.6 days. 74% of the participants had low HRQoL at one year after hospital discharge and were associated with being ≥ 41 years old (aPR: 1.95), having a previous history of psychiatric diagnoses before COVID-19 infection (aPR: 1.47), having any COVID-19 symptom during the follow-up at one year (aPR: 1.84), and having a family member who had died from COVID-19 during the first wave (aPR: 1.24). In addition, the clinically relevant depression symptoms were frequent, and they increased from 3 (14.3%) to 12 months (18.5%).

CONCLUSION

One year after COVID-19 hospitalization discharge, patients had low HRQoL, and their depression symptoms increased. These findings acknowledge the need to provide services that adequately address mental health sequels and HRQoL to reduce the burden of the COVID-19.

Dietary behaviours during COVID-19 among households at risk for food insecurity

The objective of the present study was to examine associations between variables of COVID-19-related concerns and changes in fruit and vegetable (FV) consumption among a sample of participants from the Brighter Bites program at risk for food insecurity. Cross-sectional data were collected during April-June 2020 using a rapid-response survey to understand social needs, COVID-19-related concerns and diet-related behaviours among families with children participating in Brighter Bites (n 1777) in the 2019-2020 school year at risk for food insecurity, within the surrounding Houston, Dallas, Austin, Texas area; Southwest Florida; Washington, D.C., United States. Of the 1777 respondents, 92 % of households reported being at risk for food insecurity. Among those from food insecure households, the majority were of Hispanic/Mexican-American/Latino (84⋅1 %) ethnic background, predominantly from Houston, Texas (71⋅4 %). During the pandemic, among individuals from food insecure households, 41 % (n 672) reported a decrease in FV intake, 32 % (n 527) reported an increase in FV intake, and 27 % (n 439) reported no change in FV intake. Those who reported concerns about financial stability had a 40 % greater risk of decreased FV intake compared to those not concerned about financial stability (RR 1⋅4; 95 % CI 1⋅0, 2⋅0; P = 0⋅03). The present study adds to this current body of sparse literature on how the initial phase of the pandemic impacted FV consumption behaviours among food insecure households with children. Effective interventions are needed to diminish the negative impact of COVID-19 on the population's health.

Using Causal Diagrams for Biomedical Research

Causal diagrams are used in biomedical research to develop and portray conceptual models that accurately and concisely convey assumptions about putative causal relations. Specifically, causal diagrams can be used for both observational studies and clinical trials to provide a scientific basis for some aspects of multivariable model selection. This methodology also provides an explicit framework for classifying potential sources of bias and enabling the identification of confounder, collider, and mediator variables for statistical analyses. We illustrate the potential serious miscalculation of effect estimates resulting from incorrect selection of variables for multivariable modeling without regard to their type and causal ordering as demonstrated by causal diagrams. Our objective is to improve researchers' understanding of the critical variable selection process to enhance their communication with collaborating statisticians regarding the scientific basis for intended study designs and multivariable statistical analyses. We introduce the concept of causal diagrams and their development as directed acyclic graphs to illustrate the usefulness of this methodology. We present numeric examples of effect estimate calculations and miscalculations based on analyses of the well-known Framingham Heart Study. Clinical researchers can use causal diagrams to improve their understanding of complex causation relations to determine accurate and valid multivariable models for statistical analyses. The Framingham Heart Study dataset and software codes for our analyses are provided in Appendix E1 (available online at http://www.annemergmed.com) to allow readers the opportunity to conduct their analyses.

Prenatal exposure to trans fatty acids and head growth in fetal life and childhood: triangulating confounder-adjustment and instrumental variable approaches

Dietary trans fatty acids (TFAs) are primarily industrially produced and remain abundant in processed food, particularly in low- and middle-income countries. Although TFAs are a cause of adverse cardiometabolic outcomes, little is known about exposure to TFAs in relation to brain development. We aimed to investigate the effect of maternal TFA concentration during pregnancy on offspring head growth in utero and during childhood. In a prospective population-based study in Rotterdam, the Netherlands, with 6900 mother–child dyads, maternal plasma TFA concentration was assessed using gas chromatography in mid-gestation. Offspring head circumference (HC) was measured in the second and third trimesters using ultrasonography; childhood brain morphology was assessed using magnetic resonance imaging at age 10 years. We performed regression analyses adjusting for sociodemographic and lifestyle confounders and instrumental variable (IV) analyses. Our IV analysis leveraged a national policy change that led to a substantial reduction in TFA and occurred mid-recruitment. After adjusting for covariates, maternal TFA concentration during pregnancy was inversely related to fetal HC in the third trimester (mean difference per 1% wt:wt increase: − 0.33, 95% CI − 0.51, − 0.15, cm) and to fetal HC growth from the second to the third trimester (− 0.04, 95% CI − 0.06, − 0.02, cm/week). Consistent findings were obtained with IV analyses, strengthening a causal interpretation. Association between prenatal TFA exposure and HC in the second trimester or global brain volume at age 10 years was inconclusive. Our findings are of important public health relevance as TFA levels in food remain high in many countries.

Critical window for the association between prenatal environmental tobacco smoke exposure and preterm birth

Although environmental tobacco smoke (ETS) exposure is considered to be a severe public health problem and a modifiable risk factor for preterm birth (PTB), we still lack a comprehensive understanding of the PTB risk associated with trimester-specific prenatal ETS exposure. This study aimed to examine the accumulation of risk across trimester ETS exposure and the critical window of the association between maternal ETS exposure during pregnancy and PTB. A total of 63,038 mother-child pairs were involved in the analysis of the 2017 survey of Longhua Child Cohort Study. Information about socio-demographic characteristics, prenatal ETS exposure, and birth outcomes were collected using a self-report questionnaire. A series of logistic regression models were employed to assess the associations between prenatal ETS exposure and PTB. We found that maternal ETS exposure during pregnancy significantly increased the risk of PTB and this association increased with both the average level of daily ETS exposure and the number of trimesters of ETS exposure. Moreover, mothers who were initially exposed to ETS in the 1st trimester of pregnancy had significant higher risk of PTB (OR = 1.34, 95% CI: 1.25-1.44). Furthermore, mothers exposed to ETS in the 1st trimester only (OR = 1.26, 95%CI: 1.04-1.50), in both 1st and 2nd trimester (OR = 1.35, 95%CI: 1.08-1.67) and throughout pregnancy (OR = 1.35, 95%CI: 1.24-1.46) experienced a significantly high risk of PTB. Prenatal maternal ETS exposure during only the 2nd trimester also resulted in a high risk of PTB with marginal significance (OR = 1.33, 95% CI:0.78-2.13). To conclude, the 1st and early 2nd trimester might be the critical window for prenatal ETS exposure causing PTB.

Association of peripubertal blood lead levels with reproductive hormones and semen parameters in a longitudinal cohort of Russian men

STUDY QUESTION

Are peripubertal blood lead levels (BLLs) associated with semen parameters and serum reproductive hormones among young Russian men?

SUMMARY ANSWER

We observed a suggestion of lower ejaculate volume with higher peripubertal BLL but no associations of BLLs with reproductive hormones measured throughout adolescence or with other sperm parameters measured at adulthood.

WHAT IS KNOWN ALREADY

Lead is a known reproductive toxicant and endocrine disruptor. Previous literature has shown associations between high lead exposure and poorer semen quality both in occupationally and environmentally exposed men. However, to our knowledge, no longitudinal studies have explored the association of childhood lead exposure with semen parameters and reproductive hormones in young men.

STUDY DESIGN, SIZE, DURATION

The Russian Children's Study is a prospective cohort study that enrolled 516 boys at age 8-9 years in 2003-2005 and followed them annually for 10 years. BLLs were measured at entry and lifestyle and health questionnaires were completed. Reproductive hormones were measured in blood samples collected every 2 years.

PARTICIPANTS/MATERIALS, SETTING, METHODS

Among the 516 boys enrolled, 481 had BLLs measured at entry. Of these, 453 had at least one measurement of serum testosterone, follicle stimulating hormone (FSH) or luteinizing hormone (LH) (median = 5 samples per boy) and 223 had semen samples collected ∼10 years after enrolment. Semen assessment included ejaculated volume, sperm concentration, progressive motility and total sperm count, and parameters were categorized using published andrology standards for low semen quality based on sperm count and motility. Linear mixed models were used to examine the associations of log-transformed BLLs (and BLL categories) with reproductive hormones and semen parameters, adjusting for potential confounders.

MAIN RESULTS AND THE ROLE OF CHANCE

Among the 223 young men with peripubertal BLLs and at least one semen sample (total samples = 438), the median (interquartile range) BLL was 3 (2, 5) µg/dl and 27% had BLL ≥5 µg/dl. Overall, 49% of the semen samples fell below reference levels for sperm count and/or motility. Men with peripubertal BLL ≥5 µg/dl had significantly lower ejaculated volume than those with BLL <5 µg/dl (mean = 2.42 vs 2.89 ml, P = 0.02), but this difference was attenuated in adjusted models (mean = 2.60 vs 2.83 ml, P = 0.25). No associations were observed between BLL measured at age 8-9 years and reproductive hormone levels or sperm parameters, including sperm concentration, total count, progressive motility and total progressive motile sperm count, or with the probability of having low semen quality based on sperm count/motility.

LIMITATIONS, REASONS FOR CAUTION

Only a subset of the original cohort participated in the semen quality portion of the study, although inverse probability weighting was used to account for possible selection bias. BLLs were only measured at a single time in peripuberty, and other exposure time periods, including later or longer-term childhood exposure, may be more predictive of semen quality. The young men were also exposed to other chemical contaminants before and during pubertal development.

WIDER IMPLICATIONS OF THE FINDINGS

While semen volume often receives less attention than other sperm parameters, it is an important component of male fertility. Additional prospective studies covering different exposure windows and including other seminal plasma biomarkers are warranted to explore our finding of potentially lower ejaculated volume with higher BLLs and to confirm the lack of associations for other semen parameters among youth exposed to environmental BLLs.

STUDY FUNDING/COMPETING INTEREST(S)

Funding was provided through grants R01ES0014370 and P30ES000002 from the National Institute of Environmental Health Sciences, grant R82943701 from the U.S. Environmental Protection Agency, and grant 18-15-00202 from the Russian Science Foundation (O.S and Y.D.). All authors report no competing interests.

TRIAL REGISTRATION NUMBER

N/A.

Association of Alcohol Use Diagnostic Codes in Pregnancy and Offspring Conotruncal and Endocardial Cushion Heart Defects

Background

The pathogenesis of congenital heart disease (CHD) remains largely unknown, with only a small percentage explained solely by genetic causes. Modifiable environmental risk factors, such as alcohol, are suggested to play an important role in CHD pathogenesis. We sought to evaluate the association between prenatal alcohol exposure and CHD to gain insight into which components of cardiac development may be most vulnerable to the teratogenic effects of alcohol.

Methods and Results

This was a retrospective analysis of hospital discharge records from the California Office of Statewide Health Planning and Development and linked birth certificate records restricted to singleton, live‐born infants from 2005 to 2017. Of the 5 820 961 births included, 16 953 had an alcohol‐related International Classification of Diseases, Ninth and Tenth Revisions (ICD‐9; ICD‐10) code during pregnancy. Log linear regression was used to calculate risk ratios (RR) for CHD among individuals with an alcohol‐related ICD‐9 and ICD10 code during pregnancy versus those without. Three models were created: (1) unadjusted, (2) adjusted for maternal demographic factors, and (3) adjusted for maternal demographic factors and comorbidities. Maternal alcohol‐related code was associated with an increased risk for CHD in all models (RR, 1.33 to 1.84); conotruncal (RR, 1.62 to 2.11) and endocardial cushion (RR, 2.71 to 3.59) defects were individually associated with elevated risk in all models.

Conclusions

Alcohol‐related diagnostic codes in pregnancy were associated with an increased risk of an offspring with a CHD, with a particular risk for endocardial cushion and conotruncal defects. The mechanistic basis for this phenotypic enrichment requires further investigation.

Prenatal and infant antibiotic exposure and childhood growth, obesity and cardiovascular risk factors: The Rhea mother-child cohort study, Crete, Greece

BACKGROUND

Early-life antibiotic use has been hypothesized to promote weight gain and increase the risk of childhood obesity.

OBJECTIVES

To examine the associations of prenatal and infant antibiotics with childhood growth, adiposity and cardiometabolic traits in the Greek Rhea cohort.

METHODS

We used data from 747 mother-child pairs with anthropometric measurements drawn from medical records or measured at 4 and 6 years of age. Antibiotic exposure was assessed by maternal report during pregnancy and at the first year of life. Children were classified as exposed to antibiotics prenatally if the mother received at least one course of oral antibiotics during pregnancy and postnatally if the mother reported that the child received at least one oral antibiotic treatment during the first year of life. Outcomes included repeated weight, body mass index (BMI), waist circumference, body fat (%), total cholesterol and blood pressure. We applied mixed effects, linear and log-binomial regression models after adjusting for important covariates.

RESULTS

Around 14.6% of the participating children were prenatally exposed to antibiotics and 32.4% received antibiotics during the first year of life. Prenatal exposure to antibiotics was associated with a twofold increase in the risk for obesity (risk ratio [RR]; 95% confidence interval [CI]: 2.09 [1.58, 2.76]) and abdominal obesity (RR [95% CI]: 2.56 [1.89, 3.47]) at 6 years. Postnatal exposure to antibiotics was associated with increased weight (beta [95% CI]: 00.25 [0.06, 0.44]) and BMI (beta [95% CI]: 0.23 [0.003, 0.45]) SD scores from 2 to 7 years of life.

CONCLUSION

Early-life antibiotic use was associated with accelerated childhood growth and higher adiposity.

Amplification of Bias Due to Exposure Measurement Error

Observational epidemiologic studies typically face challenges of exposure measurement error and confounding. Consider an observational study of the association between a continuous exposure and an outcome, where the exposure variable of primary interest suffers from classical measurement error (i.e., the measured exposures are distributed around the true exposure with independent error). In the absence of exposure measurement error, it is widely recognized that one should control for confounders of the association of interest to obtain an unbiased estimate of the effect of that exposure on the outcome of interest. However, here we show that, in the presence of classical exposure measurement error, the net bias in an estimate of the association of interest may increase upon adjustment for confounders. We offer an analytical expression for calculating the change in net bias in an estimate of the association of interest upon adjustment for a confounder in the presence of classical exposure measurement error, and we illustrate this problem using simulations.

Changes in Digital Media Use and Physical Activity in German Young Adults under the Covid-19 Pandemic - A Cross-Sectional Study

Many studies observed a reduction of physical activity (PA) and an increase in digital media use in young adults during the COVID-19 pandemic. However, few studies have been conducted in Europe or looked at changes in the association between both behaviors. Hence, this study aims at investigating the changes in digital media use/social media use and PA as well as in its association among young adults in Germany. Cross-sectional data of 884 German young adults (mean age 22.36 (±1.99), 76% female) collected via an online questionnaire between August 1 and September 30, 2020 were analyzed. Participants reported on digital media use (smartphone, television, computer, gaming console), social media use (Facebook, Instagram, Snapchat, Twitter, YouTube, TikTok) and PA (days/week of ≥30 min. PA) separately for the period of strict infection control measures in Germany (March - end of May 2020) and for normal times (before March 2020). Descriptive statistics of digital media use, social media use and PA were compared between both periods. Linear regression adjusted for sociodemographic and work-related characteristics were conducted for both periods with total media use, the various media devices and social media use, respectively, as independent and PA as dependent variables. Whereas PA did not differ between both periods, mean total digital media use increased by 1 hour during the period of strict infection control measures. Digital media use and social media use were negatively associated with PA in both time periods. Differences in these associations by sex could be found for some digital media devices. However, 60% of respondents did not comply with the WHO recommendations for PA. Under consideration of possible recall bias, young adults' digital media use, but not PA, seemed to have changed under the strict infection control measures. However, interventions are needed to increase PA and to prevent its reduction in the course of the pandemic.

Impact of Maternal Intrapartum Antibiotics, and Caesarean Section with and without Labour on Bifidobacterium and Other Infant Gut Microbiota

Background and Aims: Few studies consider the joint effect of multiple factors related to birth, delivery mode, intrapartum antibiotic prophylaxis and the onset of labour, on the abundance of Bifidobacterium and the quantity of this genus and its species Bifidobacterium longum subsp. infantis in the infant gut microbiota. We implemented such a study. Methods: Among 1654 Canadian full-term infants, the gut microbiota of faecal samples collected at 3 months were profiled by 16S rRNA sequencing; the genus Bifidobacterium and Bifidobacterium longum subsp. infantis were quantified by qPCR. Associations between Bifidobacterium and other gut microbiota were examined by Spearman’s rank correlation. Results: Following vaginal birth, maternal IAP exposure was associated with reduced absolute quantities of bifidobacteria among vaginally delivered infants (6.80 vs. 7.14 log₁₀ (gene-copies/g faeces), p < 0.05), as well as their lowered abundance relative to other gut microbiota. IAP differences in infant gut bifidobacterial quantity were independent of maternal pre-pregnancy body-mass-index (BMI), and remarkably, they were limited to breastfed infants. Pre-pregnancy BMI adjustment revealed negative associations between absolute quantities of bifidobacteria and CS with or without labour in non-breastfed infants, and CS with labour in exclusively breastfed infants. Significant correlations between Bifidobacterium abundance and other microbial taxa were observed. Conclusions: This study documented the impact of the birth mode and feeding status on the abundance of gut Bifidobacterium, and pointed to the important ecological role of the genus Bifidobacterium in gut microbiota due to its strong interaction with other gut microbiota in early infancy.

Role of locoregional surgery in treating FIGO 2009 stage IVB cervical cancer patients: a population-based study

OBJECTIVE

We aimed to analyse the clinical value of primary site surgery in improving the cancer-specific survival (CSS) and overall survival (OS) of initial metastatic cervical cancer patients.

DESIGN

A population-based retrospective study.

SETTING

National Cancer Institute's Surveillance, Epidemiology and End Results database.

PARTICIPANTS

We analysed 1390 patients with the International Federation of Gynecology and Obstetrics 2009 stage IVB cervical cancer with complete clinical data treated between 2010 and 2016.

INTERVENTIONS

Primary site surgery.

MEASURES

Propensity score matching (PSM) with a ratio of 1:2 was used to balance measure covariates of comparison groups. Survival time was calculated using Kaplan-Meier methods and compared by the log-rank test. To eliminate the bias of site-specific metastasis, clinicopathological factors and subsequent therapy on survival analysis, subgroup analyses stratified by metastasis type, clinicopathological factors and subsequent therapy were employed to evaluate the effect of cervical surgery on survival. Combination of directed acyclic graph and change-in-estimate procedures was performed to indentified confounders, and Cox regression was used to assess the survival benefit of cervical surgery for primary metastatic cervical cancer patients. The consistency of our findings was evaluated through sensitivity analysis.

RESULTS

Matching resulted in two comparison groups with minor differences in most variables. Pre-and-post-PSM, the median CSS and OS in the surgery group were 1.3 and 1.5, 1.1 and 1.2 times of those in the non-surgery group, respectively. Primary site surgery conferred prognosis superiority for patients with metastases to distant lymph node and other sites rather than organ metastases. After PSM and adjusting confounders, local surgery reduced the cancer related and overall mortality rates by 31% and 30%, respectively.

CONCLUSIONS

Surgical procedures could promote survival in patients with primary metastatic cervical cancer and should be considered a therapeutic option for carefully chosen patients.

Reflection on modern methods: building causal evidence within high-dimensional molecular epidemiological studies of moderate size

This commentary provides a practical perspective on epidemiological analysis within a single high-dimensional study of moderate size to consider a causal question. In this setting, non-causal confounding is important. This occurs when a factor is a determinant of outcome and the underlying association between exposure and the factor is non-causal. That is, the association arises due to chance, confounding or other bias rather than reflecting that exposure and the factor are causally related. In particular, the influence of technical processing factors must be accounted for by pre-processing measures to remove artefact or to control for these factors such as batch run. Work steps include the evaluation of alternative non-causal explanations for observed exposure-disease associations and strategies to obtain the highest level of causal inference possible within the study. A systematic approach is required to work through a question set and obtain insights on not only the exposure-disease association but also the multifactorial causal structure of the underlying data where possible. The appropriate inclusion of molecular findings will enhance the quest to better understand multifactorial disease causation in modern observational epidemiological studies.

Prevalence of diabetes and hypertension and their interaction effects on cardio-cerebrovascular diseases: a cross-sectional study

BACKGROUND

Hypertension and diabetes mellitus are two of the major risk factors for cardio-cerebrovascular diseases (CVDs). Although prior studies have confirmed that the coexistence of the two can markedly increase the risk of CVDs, few studies investigated whether potential interaction effects of hypertension and diabetes can result in greater cardio-cerebrovascular damage. We aimed to investigate the prevalence of hypertension and diabetes and whether they both affect synergistically the risk of CVDs.

METHODS

A cross-sectional study was conducted by using a multistage stratified random sampling among communities in Changsha City, Hunan Province. Study participants aged > = 18 years were asked to complete questionnaires and physical examinations. Multivariate logistic regression models were performed to evaluate the association of diabetes, hypertension, and their multiplicative interaction with CVDs with adjustment for potential confounders. We also evaluated additive interaction with the relative excess risk ratio (RERI), attribution percentage (AP), synergy index (SI).

RESULTS

A total of 14,422 participants aged 18-98 years were collected (men = 5827, 40.7%). The prevalence was 22.7% for hypertension, 7.0% for diabetes, and 3.8% for diabetes with hypertension complication, respectively. Older age, women, higher educational level, unmarried status, obesity (central obesity) were associated with increased risk of hypertension and diabetes. We did not find significant multiplicative interaction of diabetes and hypertension on CVDs, but observed a synergistic additive interaction on coronary heart disease (SI, 1.43; 95% CI, 1.03-1.97; RERI, 1.94; 95% CI, 0.05-3.83; AP, 0.26; 95% CI, 0.06-0.46).

CONCLUSIONS

Diabetes and hypertension were found to be associated with a significantly increased risk of CVDs and a significant synergistic additive interaction of diabetes and hypertension on coronary heart disease was observed. Participants who were old, women, highly educated, unmarried, obese (central obese) had increased risk of diabetes and hypertension.

Where Does CLTS Work Best? Quantifying Predictors of CLTS Performance in Four Countries

Improving the effectiveness of rural sanitation interventions is critical for meeting the United Nations' Sustainable Development Goals and improving public health. Community-led total sanitation (CLTS) is the most widely used rural sanitation intervention globally; however, evidence shows that CLTS does not work equally well everywhere. Contextual factors outside the control of implementers may partially determine CLTS outcomes, although the extent of these influences is poorly understood. In this study, we investigate the extent to which 18 contextual factors from readily available datasets can help predict the achievement and sustainability of open-defecation-free (ODF) status in Cambodia, Ghana, Liberia, and Zambia. Using multilevel logistic regressions, we found that the predictors of CLTS performance varied between countries, with the exception of small community size. Accessibility and literacy levels were correlated with CLTS outcomes, but the direction of correlation differed between countries. To translate findings into practical guidance for CLTS implementers, we used classification and regression trees to identify a "split point" for each contextual factor significantly associated with ODF achievement. We also identified the combinations of factors conducive to a minimum of 50% ODF achievement. This study demonstrates that publicly available, high-resolution datasets on accessibility, socioeconomic, and environmental factors can be leveraged to target CLTS activities to the most favorable contexts.

A Clinical Rationale for Assessing the Impact of Childhood Sexual Abuse on Adjunctive Subcutaneous Esketamine for Treatment-Resistant Depression

Background: A history of child sexual abuse (CSA) is related to higher suicide rates and poor treatment outcomes in depressed adult patients. Twenty years after the first study investigating the effects of ketamine/esketamine on depression and suicide, there is a lack of data on the CSA effects on this emerging treatment. Here, we assess the impact of CSA on adjunctive subcutaneous (SC) esketamine for treatment-resistant depression (TRD). Methods: A directed acyclic graphic (DAG) was designed to identify clinical confounders between CSA and esketamine predictors of response. The confounders were applied in a statistical model to predict depression symptom trajectory in a sample of 67 TRD outpatients. Results: The patient sample had a relatively high prevalence rate of CSA (35.82%). Positive family history of first-degree relatives with alcohol use disorder and sex were clinical mediators of the effects of esketamine in a CSA adult population. Overall, the presence of at least one CSA event was unrelated to esketamine symptom reduction. Conclusions: Unlike responses to conventional antidepressants and psychotherapy, CSA does not appear to predict poor response to esketamine.

What Environmental Factors Influence the Concentration of Fecal Indicator Bacteria in Groundwater? Insights from Explanatory Modeling in Uganda and Bangladesh

Information about microbial water quality is critical for managing water safety and protecting public health. In low-income countries, monitoring all drinking water supplies is impractical because financial resources and capacity are insufficient. Data sets derived from satellite imagery, census, and hydrological models provide an opportunity to examine relationships between a suite of environmental risk factors and microbial water quality over large geographical scales. We investigated the relationships between groundwater fecal contamination and different environmental parameters in Uganda and Bangladesh. In Uganda, groundwater contamination was associated with high population density (p < 0.001; OR = 1.27), high cropland coverage (p < 0.001; OR = 1.47), high average monthly precipitation (p < 0.001; OR = 1.14), and high surface runoff (p < 0.001; OR = 1.37), while low groundwater contamination was more likely in areas further from cities (p < 0.001; OR = 0.66) and with higher forest coverage (p < 0.001; OR = 0.70). In Bangladesh, contamination was associated with higher weekly precipitation (p < 0.001; OR = 1.44) and higher livestock density (p = 0.05; OR = 1.11), while low contamination was associated with low forest coverage (p < 0.001; OR = 1.23) and high cropland coverage (p < 0.001; OR = 0.80). We developed a groundwater contamination index for each country to help decision-makers identify areas where groundwater is most prone to fecal contamination and prioritize monitoring activities. Our approach demonstrates how to harness satellite-derived data to guide water safety management.

Kidney damage and associated risk factors in rural and urban sub-Saharan Africa (AWI-Gen): a cross-sectional population study

BACKGROUND

Rapid epidemiological health transitions occurring in vulnerable populations in Africa that have an existing burden of infectious and non-communicable diseases predict an increased risk and consequent prevalence of kidney disease. However, few studies have characterised the true burden of kidney damage and associated risk factors in Africans. We investigated the prevalence of markers for kidney damage and known risk factors in rural and urban settings in sub-Saharan Africa.

METHODS

In this cross-sectional population study (Africa Wits-International Network for the Demographic Evaluation of Populations and their Health Partnership for Genomic Studies [AWI-Gen]), we recruited unrelated adult participants aged 40-60 years from four rural community research sites (Nanoro, Burkina Faso; Navrongo, Ghana; Agincourt and Dikgale, South Africa), and two urban community research sites (Nairobi, Kenya; and Soweto, South Africa). Participants were identified and selected using random sampling frames already in use at each site. Participants completed a lifestyle and medical history questionnaire, had anthropometric and blood pressure measurements taken, and blood and urine samples were collected. Markers of kidney damage were defined as low estimated glomerular filtration rate (eGFR; <60 mL/min per 1·73 m²), presence of albuminuria (urine albumin creatinine ratio >3 mg/mmol); or chronic kidney disease (low eGFR or albuminuria, or both). We calculated age-adjusted prevalence of chronic kidney disease, low eGFR, and albuminuria by site and sex and used logistic regression models to assess risk factors of kidney damage.

FINDINGS

Between August, 2013, and August, 2016, we recruited 10 702 participants, of whom 8110 were analysable. 4120 (50·8%) of analysable participants were male, with a mean age of 49·9 years (SD 5·8). Age-standardised population prevalence was 2·4% (95% CI 2·1-2·8) for low eGFR, 9·2% (8·4-10·0) for albuminuria, and 10·7% (9·9-11·7) for chronic kidney disease, with higher prevalences in South African sites than in west African sites (14·0% [11·9-16·4] in Agincourt vs 6·6% [5·5-7·9] in Nanoro). Women had a higher prevalence of chronic kidney disease (12·0% [10·8-13·2] vs 9·5% [8·3-10·8]) and low eGFR (3·0% [2·6-3·6] vs 1·7% [1·3-2·3]) than did men, with no sex-specific differences for albuminuria (9·9% [8·8-11·0] vs 8·4% [7·3-9·7]). Risk factors for kidney damage were older age (relative risk 1·04, 95% CI 1·03-1·05; p<0·0001), hypertension (1·97, 1·68-2·30; p<0·0001), diabetes (2·22, 1·76-2·78; p<0·0001), and HIV (1·65, 1·36-1·99; p<0·0001); whereas male sex was protective (0·85, 0·73-0·98; p=0·02).

INTERPRETATION

Regional differences in prevalence and risks of chronic kidney disease in sub-Saharan Africa relate in part to varying stages of sociodemographic and epidemiological health transitions across the area. Public health policy should focus on integrated strategies for screening, prevention, and risk factor management in the broader non-communicable disease and infectious diseases framework.

FUNDING

National Human Genome Research Institute, Office of the Director, Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institute of Environmental Health Sciences, the Office of AIDS Research, and National Institute of Diabetes and Digestive and Kidney Diseases, all of the National Institutes of Health, and the South African Department of Science and Technology.

Lifetime recreational moderate-to-vigorous physical activity and ovarian cancer risk: A case-control study

Results of epidemiologic studies of physical activity and ovarian cancer risk are inconsistent. Few have attempted to measure physical activity over the lifetime or in specific age windows, which may better capture etiologically relevant exposures. We examined participation in moderate-to-vigorous recreational physical activity (MVPA) in relation to ovarian cancer risk. In a population-based case-control study conducted in Montreal, Canada from 2011 to 2016 (485 cases and 887 controls), information was collected on lifetime participation in various recreational physical activities, which was used to estimate MVPA for each participant. MVPA was represented as average energy expenditure over the lifetime and in specific age-periods in units of metabolic equivalents (METs)-hours per week. Odds ratios (OR) and 95% confidence intervals (CI) for the relation between average MVPA and ovarian cancer risk were estimated using multivariable logistic regression models. Confounding was assessed using directed acyclic graphs combined with a change-in-estimate approach. The adjusted OR (95% CI) for each 28.5 MET-hr/week increment of lifetime recreational MVPA was 1.11 (0.99-1.24) for ovarian cancer overall. ORs for individual age-periods were weaker. When examined by menopausal status, the OR (95% CI) for lifetime MVPA was 1.21 (1.00-1.45) for those diagnosed before menopause and 1.04 (0.89-1.21) for those diagnosed postmenopausally. The suggestive positive associations were stronger for invasive ovarian cancers and more specifically for high-grade serous carcinomas. These results do not support a reduced ovarian cancer risk associated with MVPA.

Maternal Distress During Pregnancy and Recurrence in Early Childhood Predicts Atopic Dermatitis and Asthma in Childhood

BACKGROUND

Early-life stress is becoming an important determinant of immune system programming. Maternal prenatal distress is found to be associated with atopic disease in offspring but the separate effects of postnatal distress are not well-studied.

RESEARCH QUESTION

Does the likelihood of asthma and atopic dermatitis in children increase when they are exposed to maternal distress pre- and postnatally in a sex-specific manner?

STUDY DESIGN AND METHODS

Using data from a provincial newborn screen and health-care database for 12,587 children born in 2004, maternal distress (depression or anxiety) was defined as prenatal, self-limiting, recurrent, or late-onset postpartum. Atopic dermatitis (AD) and asthma at ages 5 years and 7 years of age were diagnosed by using hospitalization, physician visit, or prescription records. Associations between maternal distress and childhood asthma and AD were determined by using multiple logistic regression.

RESULTS

After adjusting for risk factors, a significant association between maternal prenatal (OR, 1.27; 95% CI, 1.11-1.46), recurrent postpartum (OR, 1.28; 95% CI, 1.11-1.48), and late-onset postpartum (OR, 1.19, 95% CI, 1.06-1.34) distress was found with AD at age 5 years. Asthma at age 7 years was also associated with maternal prenatal distress (OR, 1.57; 95% CI, 1.29-1.91) and late-onset postnatal distress (OR, 1.22; 95% CI, 1.01-1.46). Self-limiting postnatal distress was not found to be a risk factor for either atopic condition. Associations with AD or asthma were of a similar magnitude in boys and girls; the exception was recurrent postnatal distress, which increased risk for asthma in boys only.

INTERPRETATION

This population-based study provides evidence for sex-specific associations between maternal prenatal and postnatal distress, as well as the development of AD and asthma. The findings support recommendations for greater psychosocial support of mothers during pregnancy and early childhood to prevent childhood atopic disease.

Exposure to Persistent Organic Pollutants (POPs) and Their Relationship to Hepatic Fat and Insulin Insensitivity among Asian Indian Immigrants in the United States

Persistent organic pollutants (POPs), such as dichlorodiphenyltrichloroethane (DDT) and other organochlorine compounds, are abundant in the environment and in foodstuffs from the Indian subcontinent. These environmental contaminants have been associated with a higher risk of diabetes in numerous studies. Asian Indians are well known to have a high risk of diabetes compared with other populations, and this risk is also found in migrant populations of Asian Indians in the United States, Europe, and elsewhere. We hypothesized that high plasma concentrations of POPs in Asian Indian migrants are linked to a variety of diabetes-related pathologies and explored the mechanism for the induction of these effects. We measured 30 environmental pollutants in plasma samples obtained from 147 participants in the Metabolic syndrome and Atherosclerosis in South Asians Living in America pilot study using a gas chromatography-tandem mass spectrometry analytical method that uses less than 0.5 mL of plasma. We found that plasma levels of o,p'-DDT and p,p'-DDT were independently associated with both body mass index (BMI) and waist circumference. Doubling the levels of the sums of these DDTs was associated with insulin insensitivity (-0.38 Matsuda index, p = 0.001), increased adiposity (1.26 kg/m2 BMI and 3.58 cm waist circumference increase, p < 0.0001), circulating insulin (12.9 mIU/L, p = 0.002), hepatic fat (-0.051 HU, p = 0.001), as well as increased odds of obesity (OR = 2.17, p < 0.001, BMI-based; OR = 2.37, p = 0.001, waist-based), prediabetes (OR = 1.55, p = 0.02), diabetes (OR = 1.72, p = 0.01), and fatty liver (OR = 1.66, p = 0.01) in multivariable models accounting for confounding by age, sex, years in the US, education, and fish protein. Furthermore, levels of DDTs were associated with increased hepatic fat and circulating insulin, independent of obesity and confounders. These findings suggest that exposure to DDTs may contribute to the risk of metabolic disease among Asian Indians by affecting hepatic fat levels independent of obesity.

Cord Leptin is Associated with Neuropsychomotor Development in Childhood

OBJECTIVE

Leptin is critical for central nervous system development and maturation. This study aimed to evaluate the potential regulatory role of cord leptin in the neuropsychomotor development of children ages 18 months to 6 years.

METHODS

This study included 424 children from a prospective mother-child cohort (Rhea Study; Crete, Greece) with available cord leptin levels and data on neurodevelopmental outcomes at 18 months (Bayley Scales of Infant and Toddler Development, Third Edition), 4 years (McCarthy Scales of Children's Abilities), and 6 years (Raven's Coloured Progressive Matrices and Trail Making Test). Multivariable linear regression models were used to explore the associations.

RESULTS

Each 10-ng/mL increase in the cord leptin level was associated with increased scores on the gross motor scale at 18 months (β coefficient: 3.8; 95% CI: 0.0-7.5), with decreased scores in the general cognitive performance (β coefficient: -3.0; 95% CI: -5.5 to -0.4), perceptual performance (β coefficient: -3.4; 95% CI: -6.0 to -9.9), working memory (β coefficient: -3.1; 95% CI: -5.7 to -0.4), executive function (β coefficient -3.1; 95% CI: -5.7 to -0.5), and functions of the posterior cortex (β coefficient: -2.7; 95% CI: -5.2 to -0.1) scales at 4 years, and with a 3.7-unit decrease in the Raven's Coloured Progressive Matrices score at 6 years (β coefficient: -3.7; 95% CI: -6.9 to -0.5).

CONCLUSIONS

Increased cord leptin levels are associated with enhanced gross motor development at 18 months but decreased cognitive performance in early and middle childhood.

Pregnant Inuit Women's Exposure to Metals and Association with Fetal Growth Outcomes: ACCEPT 2010⁻2015

Environmental contaminants such as heavy metals are transported to the Arctic regions via atmospheric and ocean currents and enter the Arctic food web. Exposure is an important risk factor for health and can lead to increased risk of a variety of diseases. This study investigated the association between pregnant women's levels of heavy and essential metals and the birth outcomes of the newborn child. This cross-sectional study is part of the ACCEPT birth cohort (Adaption to Climate Change, Environmental Pollution, and dietary Transition) and included 509 pregnant Inuit women ≥18 years of age. Data were collected in five Greenlandic regions during 2010⁻2015. Population characteristics and birth outcomes were obtained from medical records and midwives, respectively, and blood samples were analyzed for 13 metals. Statistical analysis included one-way ANOVA, Spearman's rho, and multiple linear and logistic regression analyses. The proportion of current smokers was 35.8%. The levels of cadmium, chromium, and nickel were higher compared to reported normal ranges. Significant regional differences were observed for several metals, smoking, and parity. Cadmium and copper were significantly inversely related to birth outcomes. Heavy metals in maternal blood can adversely influence fetal development and growth in a dose⁻response relationship. Diet and lifestyle factors are important sources of toxic heavy metals and deviant levels of essential metals. The high frequency of smokers in early pregnancy is of concern, and prenatal exposure to heavy metals and other environmental contaminants in the Greenlandic Inuit needs further research.

Shift Work Patterns, Chronotype, and Epithelial Ovarian Cancer Risk

BACKGROUND

Shift work causing circadian disruption is classified as a "probable carcinogen" and may contribute to the pathogenesis of hormone-sensitive cancers. This study investigated shift work exposure in relation to epithelial ovarian cancer (EOC) risk.

METHODS

In a population-based case-control study with 496 EOC cases and 906 controls, lifetime occupational histories were collected and used to calculate cumulative years of shift work exposure, average number of night shifts per month, and average number of consecutive night shifts per month. ORs and 95% confidence intervals (CI) for associations with EOC risk were estimated using logistic regression. Associations were also examined according to chronotype and menopausal status.

RESULTS

More than half of the cases (53.4%) and controls (51.7%) worked evening and/or night shifts. There was no clear pattern of increasing EOC risk with increasing years of shift work; the adjusted OR of EOC comparing the highest shift work category versus never working shift work was 1.20 (95% CI, 0.89-1.63). This association was more pronounced among those self-identified as having a "morning" chronotype (OR, 1.64; 95% CI, 1.01-2.65). Associations did not greatly differ by menopausal status.

CONCLUSIONS

These results do not strongly demonstrate a relationship between shift work and EOC risk.

IMPACT

This study collected detailed shift work information and examined shift work patterns according to shift times and schedules. The findings highlight that chronotype should be considered in studies of shift work as an exposure.

Participation in a clinical trial of a text messaging intervention is associated with increased infant HIV testing: A parallel-cohort randomized controlled trial

Objective

Text messages significantly improve uptake of infant HIV testing in clinical trial contexts. Women who were excluded from a randomized trial in Kenya were followed to create a comparison between women who were enrolled and did not receive the study SMS intervention and women who were screened but not enrolled.

Design

Parallel-cohort randomized controlled trial analysis.

Methods

We compared time to infant HIV testing between women in three groups: the Trial SMS group, the Trial Control group, and the Comparison Cohort comprised of women who were screened but not enrolled.

Results

Of the 1,115 women screened, 388 (35%) were eligible for trial enrollment, and were randomized to receive either intervention text messages (Trial SMS; N = 195) or continue usual care (Trial Control; N = 193). Among 727 women not enrolled in the study (Comparison Cohort), we obtained infant HIV testing data from clinic records for 510 (70%). The cumulative probability of infant HIV testing was highest in the Trial SMS group (92.0%; 95% CI 87.5–95.3), followed by the Trial Control group (85.1%; 95% CI 79.5–89.8), and lowest among women in the Comparison Cohort (43.4%; 95% CI 39.2–47.8).

Conclusions

Both the Trial SMS group and the Trial Control group were significantly more likely to have their infants tested for HIV compared to the Comparison Cohort, providing evidence of a “clinical trial effect.” This analysis suggests that SMS interventions should be implemented as an adjunct to consistent and engaged delivery of basic health services.

Covariate selection strategies for causal inference: Classification and comparison

When causal effects are to be estimated from observational data, we have to adjust for confounding. A central aim of covariate selection for causal inference is therefore to determine a set that is sufficient for confounding adjustment, but other aims such as efficiency or robustness can be important as well. In this paper, we review six general approaches to covariate selection that differ in the targeted type of adjustment set. We discuss and illustrate their advantages and disadvantages using causal diagrams. Moreover, the approaches and different ways of implementing them are compared empirically in an extensive simulation study. We conclude that there are considerable differences between the approaches but none of them is uniformly best, with performance depending on the chosen adjustment method as well as the true confounding structure. Any prior structural knowledge on the causal relations is helpful to choose the most appropriate method.

Maternal age at delivery, lung function and asthma in offspring: a population-based survey

There is limited information about potential impact of maternal age on the respiratory health of offspring. We investigated the association of maternal age at delivery with adult offspring's lung function, respiratory symptoms and asthma, and potential differences according to offspring sex.10 692 adults from 13 countries participating in the European Community Respiratory Health Survey (ECRHS) II responded to standardised interviews and provided lung function measurements and serum for IgE measurements at age 25-55 years. In logistic and linear multilevel mixed models we adjusted for participants' characteristics (age, education, centre, number of older siblings) and maternal characteristics (smoking in pregnancy, education) while investigating for differential effects by sex. Maternal age was validated in a subsample using data from the Norwegian birth registry.Increasing maternal age was associated with increasing forced expiratory volume in 1 s (2.33 mL per year, 95% CI 0.34-4.32 mL per year), more consistent in females (p_trend 0.025) than in males (p_trend 0.14). Asthma (OR 0.85, 95% CI 0.79-0.92) and respiratory symptoms (OR 0.87, 95% CI 0.82-0.92) decreased with increasing maternal age (per 5 years) in females, but not in males (p_interaction 0.05 and 0.001, respectively). The results were consistent across centres and not explained by confounding factors.Maternal ageing was related to higher adult lung function and less asthma/symptoms in females. Biological characteristics in offspring related to maternal ageing are plausible and need further investigation.

Randomized clinical trials and observational studies in the assessment of drug safety

Randomized clinical trials are considered as the preferred design to assess the potential causal relationships between drugs or other medical interventions and intended effects. For this reason, randomized clinical trials are generally the basis of development programs in the life cycle of drugs and the cornerstone of evidence-based medicine. Instead, randomized clinical trials are not the design of choice for the detection and assessment of rare, delayed and/or unexpected effects related to drug safety. Moreover, the highly homogeneous populations resulting from restrictive eligibility criteria make randomized clinical trials inappropriate to describe comprehensively the safety profile of drugs. In that context, observational studies have a key added value when evaluating the benefit-risk balance of the drugs. However, observational studies are more prone to bias than randomized clinical trials and they have to be designed, conducted and reported judiciously. In this article, we discuss the strengths and limitations of randomized clinical trials and of observational studies, more particularly regarding their contribution to the knowledge of medicines' safety profile. In addition, we present general recommendations for the sensible use of observational data.

Neurocognitive and physical functioning in the Seveso Women's Health Study

2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) is neurotoxic in animals but few studies have investigated its effects on the human brain. Related dioxin-like compounds have been linked to poorer cognitive and motor function in older adults, with effects more pronounced in women, perhaps due to the loss of neuro-protective estrogen in menopause. On 10 July 1976, a chemical explosion in Seveso, Italy, resulted in one of the highest known residential exposures to TCDD. In 1996, we initiated the Seveso Women's Health Study, a retrospective cohort study of the health of the women who were newborn to 40 years old in 1976. Here, we investigate whether TCDD exposure is associated with physical functioning and working memory more than 20 years later. Individual TCDD concentration (ppt) was measured in archived serum collected soon after the explosion. In 1996 and 2008, we measured physical functioning (n=154) and working memory (n=459), respectively. We examined associations between serum TCDD and motor and cognitive outcomes with multivariate linear regression and semi-parametric estimators. A 10-fold increase in serum TCDD was not associated with walking speed (adjusted β=0.0006ft/s, 95% Confidence Interval (CI): -0.13, 0.13), upper body mobility (adjusted β=-0.06, 95% CI: -0.36, 0.23), or manual dexterity (adjusted β=0.34, 95% CI: -0.65, 1.33). We observed an inverted U-shaped association in grip strength, with poorer strength in the lowest and highest TCDD exposure levels. There was no association between TCDD and the Wechsler digit and spatial span tests. Neither menopause status at assessment nor developmental timing of exposure modified associations between TCDD and working memory. Our findings, in one of the only studies of TCDD's effects on neuropsychological and physical functioning in women, do not indicate an adverse effect on these domains, with the exception of a U-shaped relationship with grip strength. Given the limited assessment and relative youth of the women at this follow-up, future work examining additional neuropsychological outcomes is warranted.

Surrenderers' Relationships with Cats Admitted to four Australian Animal Shelters

Simple Summary

The surrender of cats to animal shelters results in financial, social and moral burdens for the community. Human caretaking of cats was explored in a sample of people surrendering cats to shelters in Australia. At the shelters surrenderers classified themselves as owners or non-owners and a questionnaire identified that this was related to their method of acquisition of the cat, their association time with the cat, the closeness of their relationship with the cat and their degree of responsibility for the cat’s care. A model of ownership perception was developed to provide a better understanding of factors influencing ownership perception. Understanding ownership perceptions in cats surrendered to shelters is important as these can inform the development of more targeted and effective intervention strategies to reduce numbers of unwanted cats.

Abstract

The surrender of cats to animal shelters results in financial, social and moral burdens for the community. Correlations of caretaking and interactions with surrendered cats were calculated, to understand more about humans’ relationships with surrendered cats and the contribution of semi-owned cats to shelter intakes. A questionnaire was used to collect detailed information about 100 surrenderers’ relationships with cats they surrendered to four animal shelters in Australia, with each surrenderer classifying themselves as being either the owner or a non-owner of the surrendered cat (ownership perception). Method of acquisition of the cat, association time, closeness of the relationship with the cat and degree of responsibility for the cat’s care were all associated with ownership perception. Many non-owners (59%) fed and interacted with the cat they surrendered but rarely displayed other caretaking behaviours. However, most surrenderers of owned and unowned cats were attached to and felt responsible for the cat. Based on these results and other evidence, a causal model of ownership perception was proposed to provide a better understanding of factors influencing ownership perception. This model consisted of a set of variables proposed as directly or indirectly influencing ownership perception, with connecting arrows to indicate proposed causal relationships. Understanding ownership perception and the contribution of semi-owned cats to shelter intake is important as these can inform the development of more targeted and effective intervention strategies to reduce numbers of unwanted cats.

Surrenderers' Relationships with Cats Admitted to four Australian Animal Shelters

The surrender of cats to animal shelters results in financial, social and moral burdens for the community. Correlations of caretaking and interactions with surrendered cats were calculated, to understand more about humans' relationships with surrendered cats and the contribution of semi-owned cats to shelter intakes. A questionnaire was used to collect detailed information about 100 surrenderers' relationships with cats they surrendered to four animal shelters in Australia, with each surrenderer classifying themselves as being either the owner or a non-owner of the surrendered cat (ownership perception). Method of acquisition of the cat, association time, closeness of the relationship with the cat and degree of responsibility for the cat's care were all associated with ownership perception. Many non-owners (59%) fed and interacted with the cat they surrendered but rarely displayed other caretaking behaviours. However, most surrenderers of owned and unowned cats were attached to and felt responsible for the cat. Based on these results and other evidence, a causal model of ownership perception was proposed to provide a better understanding of factors influencing ownership perception. This model consisted of a set of variables proposed as directly or indirectly influencing ownership perception, with connecting arrows to indicate proposed causal relationships. Understanding ownership perception and the contribution of semi-owned cats to shelter intake is important as these can inform the development of more targeted and effective intervention strategies to reduce numbers of unwanted cats.

Variable selection - A review and recommendations for the practicing statistician

Statistical models support medical research by facilitating individualized outcome prognostication conditional on independent variables or by estimating effects of risk factors adjusted for covariates. Theory of statistical models is well-established if the set of independent variables to consider is fixed and small. Hence, we can assume that effect estimates are unbiased and the usual methods for confidence interval estimation are valid. In routine work, however, it is not known a priori which covariates should be included in a model, and often we are confronted with the number of candidate variables in the range 10-30. This number is often too large to be considered in a statistical model. We provide an overview of various available variable selection methods that are based on significance or information criteria, penalized likelihood, the change-in-estimate criterion, background knowledge, or combinations thereof. These methods were usually developed in the context of a linear regression model and then transferred to more generalized linear models or models for censored survival data. Variable selection, in particular if used in explanatory modeling where effect estimates are of central interest, can compromise stability of a final model, unbiasedness of regression coefficients, and validity of p-values or confidence intervals. Therefore, we give pragmatic recommendations for the practicing statistician on application of variable selection methods in general (low-dimensional) modeling problems and on performing stability investigations and inference. We also propose some quantities based on resampling the entire variable selection process to be routinely reported by software packages offering automated variable selection algorithms.

How Confounder Strength Can Affect Allocation of Resources in Electronic Health Records

When electronic health record (EHR) data are used, multiple approaches may be available for measuring the same variable, introducing potentially confounding factors. While additional information may be gleaned and residual confounding reduced through resource-intensive assessment methods such as natural language processing (NLP), whether the added benefits offset the added cost of the additional resources is not straightforward. We evaluated the implications of misclassification of a confounder when using EHRs. Using a combination of simulations and real data surrounding hospital readmission, we considered smoking as a potential confounder. We compared ICD-9 diagnostic code assignment, which is an easily available measure but has the possibility of substantial misclassification of smoking status, with NLP, a method of determining smoking status that more expensive and time-consuming than ICD-9 code assignment but has less potential for misclassification. Classification of smoking status with NLP consistently produced less residual confounding than the use of ICD-9 codes; however, when minimal confounding was present, differences between the approaches were small. When considerable confounding is present, investing in a superior measurement tool becomes advantageous.

The alarming problems of confounding equivalence using logistic regression models in the perspective of causal diagrams

Background

Confounders can produce spurious associations between exposure and outcome in observational studies. For majority of epidemiologists, adjusting for confounders using logistic regression model is their habitual method, though it has some problems in accuracy and precision. It is, therefore, important to highlight the problems of logistic regression and search the alternative method.

Methods

Four causal diagram models were defined to summarize confounding equivalence. Both theoretical proofs and simulation studies were performed to verify whether conditioning on different confounding equivalence sets had the same bias-reducing potential and then to select the optimum adjusting strategy, in which logistic regression model and inverse probability weighting based marginal structural model (IPW-based-MSM) were compared. The “do-calculus” was used to calculate the true causal effect of exposure on outcome, then the bias and standard error were used to evaluate the performances of different strategies.

Results

Adjusting for different sets of confounding equivalence, as judged by identical Markov boundaries, produced different bias-reducing potential in the logistic regression model. For the sets satisfied G-admissibility, adjusting for the set including all the confounders reduced the equivalent bias to the one containing the parent nodes of the outcome, while the bias after adjusting for the parent nodes of exposure was not equivalent to them. In addition, all causal effect estimations through logistic regression were biased, although the estimation after adjusting for the parent nodes of exposure was nearest to the true causal effect. However, conditioning on different confounding equivalence sets had the same bias-reducing potential under IPW-based-MSM. Compared with logistic regression, the IPW-based-MSM could obtain unbiased causal effect estimation when the adjusted confounders satisfied G-admissibility and the optimal strategy was to adjust for the parent nodes of outcome, which obtained the highest precision.

Conclusions

All adjustment strategies through logistic regression were biased for causal effect estimation, while IPW-based-MSM could always obtain unbiased estimation when the adjusted set satisfied G-admissibility. Thus, IPW-based-MSM was recommended to adjust for confounders set.

Electronic supplementary material

The online version of this article (10.1186/s12874-017-0449-7) contains supplementary material, which is available to authorized users.

An analysis of blood donation barriers experienced by North American and Caribbean university students in Grenada, West Indies

OBJECTIVES

To estimate the associations of nationality, university program, donation history and gender, with blood donation barriers experienced by non-donating students on the day of a campus blood drive. This project focused particularly on nationality and the effect of the different blood donation cultures in the students' countries of origin.

METHODS

A retrospective cohort study of 398 North American and Caribbean university students was conducted at St. George's University, Grenada, in 2010. Data were collected from non-donating students on campus while a blood drive was taking place. Log-binomial regression was used to estimate associations between the exposures of interest and donation barriers experienced by the students.

RESULTS

North American (voluntary blood donation culture) students were more likely than Caribbean (replacement blood donation culture) students to experience "Lack of Time" (relative risk (RR) = 1.57; 95% confidence interval (CI): 1.19-2.07) and "Lack of Eligibility" (RR = 1.55; 95% CI: 1.08-2.22) as barriers to donation. Conversely, Caribbean students were a third as likely to state "Lack of Incentive" (RR = 0.32; 95% CI: 0.20-0.50), "Fear of Infection" (RR = 0.35; 95% CI: 0.21-0.58), and "Fear of Needles" (RR = 0.32; 95% CI: 0.21-0.48) were barriers than North American students.

CONCLUSIONS

University students from voluntary blood donation cultures are likely to experience different barriers to donation than those from replacement cultures. Knowledge of barriers that students from contrasting blood donation systems face provides valuable information for blood drive promotion in university student populations that contain multiple nationalities.

Chronic PM2.5 exposure and risk of infant bronchiolitis and otitis media clinical encounters

Chronic particulate matter less than 2.5μm in diameter (PM_2.5) exposure can leave infants more susceptible to illness. Our objective is to estimate associations of the chronic PM_2.5 exposure with infant bronchiolitis and otitis media (OM) clinical encounters. We obtained all first time bronchiolitis (n=18,029) and OM (n=40,042) clinical encounters among children less than 12 and 36 months of age, respectively, diagnosed from 2001 to 2009 and two controls per case matched on birthdate and gestational age from the Pregnancy to Early Life Longitudinal data linkage system in Massachusetts. We applied conditional logistic regression to estimate odds ratios (OR) and confidence intervals (CI) per 2-μg/m³ increase in lifetime average satellite based PM_2.5 exposure_. Effect modification was assessed by age, gestational age, frequency of clinical encounter, and income. We examined associations between residential distance to roadways, traffic density, and infant bronchiolitis and OM risk. PM_2.5 was not associated with infant bronchiolitis (OR=1.02, 95% CI=1.00, 1.04) and inversely associated with OM (OR=0.97, 95% CI=0.95, 0.99). There was no evidence of effect modification. Compared to infants living near low traffic density, infants residing in high traffic density had elevated risk of bronchiolitis (OR=1.23, 95% CI=1.14, 1.31) but not OM (OR=0.98, 95% CI=0.93, 1.02) clinical encounter. We did not find strong evidence to support an association between early-life long-term PM_2.5 exposure and infant bronchiolitis or OM. Bronchiolitis risk was increased among infants living near high traffic density.

[Use of causal diagrams in Epidemiology: application to a situation with confounding]

Epidemiological research still rarely uses causal diagrams, despite growing recognition of their explanatory potential. One possible reason is that many research programs involve themes in which there is a certain degree of uncertainty as to mechanisms in the processes that generate the data. In this study, the relationship between occupational stress and obesity is used as an example of the application of causal diagrams to questions related to confounding. The article presents the selection stages for variables in statistical adjustment and the derivation of a causal diagram's statistical implications. The main advantage of causal diagrams is that they explicitly reveal the respective model's underlying hypotheses, allowing critical analysis of the implications and thereby facilitating identification of sources of bias and uncertainty in the epidemiological study's results.

Use of Causal Diagrams to Inform the Design and Interpretation of Observational Studies: An Example from the Study of Heart and Renal Protection (SHARP)

Observational studies often seek to estimate the causal relevance of an exposure to an outcome of interest. However, many possible biases can arise when estimating such relationships, in particular bias because of confounding. To control for confounding properly, careful consideration of the nature of the assumed relationships between the exposure, the outcome, and other characteristics is required. Causal diagrams provide a simple graphic means of displaying such relationships, describing the assumptions made, and allowing for the identification of a set of characteristics that should be taken into account (i.e., adjusted for) in any analysis. Furthermore, causal diagrams can be used to identify other possible sources of bias (such as selection bias), which if understood from the outset, can inform the planning of appropriate analyses. In this article, we review the basic theory of causal diagrams and describe some of the methods available to identify which characteristics need to be taken into account when estimating the total effect of an exposure on an outcome. In doing so, we review the concept of collider bias and show how it is inappropriate to adjust for characteristics that may be influenced, directly or indirectly, by both the exposure and the outcome of interest. A motivating example is taken from the Study of Heart and Renal Protection, in which the relevance of smoking to progression to ESRD is considered.

Physical activity and lung cancer risk in men and women

PURPOSE

Although evidence has accumulated that recreational physical activities (PA) may reduce lung cancer risk, there is little evidence concerning the possible role of a potentially more important source of PA, namely occupational PA. We investigated both recreational and lifetime occupational PA in relation to lung cancer risk in a population-based case-control study in Montreal, Canada (N_CASES = 727; N_CONTROLS = 1,351).

METHODS

Unconditional logistic regression was used to estimate odds ratios (OR), separately for men and women, adjusting for smoking, exposure to occupational carcinogens, and sociodemographic and lifestyle factors.

RESULTS

In both sexes, increasing recreational PA was associated with a lower lung cancer risk (OR_MEN = 0.66, 95% confidence interval (CI) 0.47-0.92; OR_WOMEN = 0.55, 95% CI 0.34-0.88, comparing the highest versus lowest tertiles). For occupational PA, no association was observed among women, while increasing occupational PA was associated with increased risk among men (OR_MEN = 1.96, 95% CI 1.27-3.01). ORs were not modified by occupational lung carcinogen exposure, body mass index, and smoking level; results were similar across lung cancer histological types.

CONCLUSIONS

Our results support the previous findings for recreational PA and lung cancer risk. Unexpectedly, our findings suggest a positive association for occupational PA; this requires replication and more detailed investigation.

Explanation and Elaboration Document for the STROBE-Vet Statement: Strengthening the Reporting of Observational Studies in Epidemiology-Veterinary Extension

The STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) statement was first published in 2007 and again in 2014. The purpose of the original STROBE was to provide guidance for authors, reviewers, and editors to improve the comprehensiveness of reporting; however, STROBE has a unique focus on observational studies. Although much of the guidance provided by the original STROBE document is directly applicable, it was deemed useful to map those statements to veterinary concepts, provide veterinary examples, and highlight unique aspects of reporting in veterinary observational studies. Here, we present the examples and explanations for the checklist items included in the STROBE-Vet statement. Thus, this is a companion document to the STROBE-Vet statement methods and process document (JVIM_14575 "Methods and Processes of Developing the Strengthening the Reporting of Observational Studies in Epidemiology-Veterinary (STROBE-Vet) Statement" undergoing proofing), which describes the checklist and how it was developed.