Associations of neighborhood greenspace, and active living environments with autism spectrum disorders: A matched case-control study in Ontario, Canada

BACKGROUND

Increasing evidence links early life residential exposure to natural urban environmental attributes and positive health outcomes in children. However, few studies have focused on their protective effects on the risk of autism spectrum disorder (ASD). The aim of this study was to investigate the associations of neighborhood greenspace, and active living environments during pregnancy with ASD in young children (≤6 years).

METHODS

We conducted a population-based matched case-control study of singleton term births in Ontario, Canada for 2012-2016. The ASD and environmental data was generated using the Ontario Autism Spectrum Profile, the Better Outcomes Registry & Network Ontario, and Canadian Urban Environmental Health Research Consortium. We employed conditional logistic regressions to estimate the odds ratio (OR) between ASD and environmental factors characterizing selected greenspace metrics and neighborhoods conducive to active living (i.e., green view index (GVI), normalized difference vegetation index (NDVI), tree canopy, park proximity and active living environments index (ALE)).

RESULTS

We linked 8643 mother-child pairs, including 1554 cases (18%). NDVI (OR 1.034, 0.944-1.024, per Inter Quartile Range [IQR] = 0.08), GVI (OR 1.025, 95% CI 0.953-1.087, per IQR = 9.45%), tree canopy (OR 0.992, 95% CI 0.903-1.089, per IQR = 6.24%) and the different categories of ALE were not associated with ASD in adjusted models for air pollution. In contrast, living closer to a park was protective (OR 0.888, 0.833-0.948, per 0.06 increase in park proximity index), when adjusted for air pollution.

CONCLUSIONS

This study reported mixed findings showing both null and beneficial effects of green spaces and active living environments on ASD. Further investigations are warranted to elucidate the role of exposure to greenspaces and active living environments on the development of ASD.

Adjusting for Berkson error in exposure in ordinary and conditional logistic regression and in Poisson regression

BACKGROUND

INTEROCC is a seven-country cohort study of occupational exposures and brain cancer risk, including occupational exposure to electromagnetic fields (EMF). In the absence of data on individual exposures, a Job Exposure Matrix (JEM) may be used to construct likely exposure scenarios in occupational settings. This tool was constructed using statistical summaries of exposure to EMF for various occupational categories for a comparable group of workers.

METHODS

In this study, we use the Canadian data from INTEROCC to determine the best EMF exposure surrogate/estimate from three appropriately chosen surrogates from the JEM, along with a fourth surrogate based on Berkson error adjustments obtained via numerical approximation of the likelihood function. In this article, we examine the case in which exposures are gamma-distributed for each occupation in the JEM, as an alternative to the log-normal exposure distribution considered in a previous study conducted by our research team. We also study using those surrogates and the Berkson error adjustment in Poisson regression and conditional logistic regression.

RESULTS

Simulations show that the introduced methods of Berkson error adjustment for non-stratified analyses provide accurate estimates of the risk of developing tumors in case of gamma exposure model. Alternatively, and under some technical assumptions, the arithmetic mean is the best surrogate when a gamma-distribution is used as an exposure model. Simulations also show that none of the present methods could provide an accurate estimate of the risk in case of stratified analyses.

CONCLUSION

While our previous study found the geometric mean to be the best exposure surrogate, the present study suggests that the best surrogate is dependent on the exposure model; the arithmetic means in case of gamma-exposure model and the geometric means in case of log-normal exposure model. However, we could present a better method of Berkson error adjustment for each of the two exposure models. Our results provide useful guidance on the application of JEMs for occupational exposure assessments, with adjustment for Berkson error.

Sparse estimation for case-control studies with multiple disease subtypes

The analysis of case-control studies with several disease subtypes is increasingly common, e.g. in cancer epidemiology. For matched designs, a natural strategy is based on a stratified conditional logistic regression model. Then, to account for the potential homogeneity among disease subtypes, we adapt the ideas of data shared lasso, which has been recently proposed for the estimation of stratified regression models. For unmatched designs, we compare two standard methods based on $L_1$-norm penalized multinomial logistic regression. We describe formal connections between these two approaches, from which practical guidance can be derived. We show that one of these approaches, which is based on a symmetric formulation of the multinomial logistic regression model, actually reduces to a data shared lasso version of the other. Consequently, the relative performance of the two approaches critically depends on the level of homogeneity that exists among disease subtypes: more precisely, when homogeneity is moderate to high, the non-symmetric formulation with controls as the reference is not recommended. Empirical results obtained from synthetic data are presented, which confirm the benefit of properly accounting for potential homogeneity under both matched and unmatched designs, in terms of estimation and prediction accuracy, variable selection and identification of heterogeneities. We also present preliminary results from the analysis of a case-control study nested within the EPIC (European Prospective Investigation into Cancer and nutrition) cohort, where the objective is to identify metabolites associated with the occurrence of subtypes of breast cancer.

Non-linear dose-response relation between urinary levels of nicotine and its metabolites and cognitive impairment among an elderly population in China

BACKGROUND

Active smoking and exposure to environmental tobacco smoke may be related to cognitive function decline. We assessed the associations of urinary levels of nicotine and its metabolites with cognitive function.

METHODS

A total of 553 elder adults at high risk of cognitive impairment and 2212 gender- and age-matched individuals at low risk of cognitive impairment were selected at a ratio of 1: 4 from the remained individuals (n = 6771) who completed the baseline survey of the Shenzhen Ageing-Related Disorder Cohort, after excluding those with either Alzheimer's disease, Parkinson's syndrome or stroke as well as those with missing data on variables (including active and passive smoking status, Mini-Cog score). Urinary levels of nicotine and its metabolites and cognitive function for all individuals were measured by high-performance liquid chromatography-tandem mass spectrometry (LC-MS/MS) and assessed using the Mini-Cog test, respectively. Associations of urinary levels of nicotine and its metabolites with cognitive function were analyzed by conditional logistic regression models.

RESULTS

Individuals in the highest tertile of urinary OHCotGluc (OR: 1.52, 95%CI: 1.19-1.93) or NNO (OR: 1.50, 95%CI: 1.16-1.93) levels as well as in the second tertile of urinary ∑Nic level (OR: 1.43, 95%CI: 1.13-1.82) were at higher risk of cognitive impairment compared with those in the corresponding lowest tertile. Restricted cubic spline models revealed the non-linear dose-response relationships between urinary levels of OHCotGluc, NNO or ∑Nic and the risk of cognitive impairment.

CONCLUSIONS

Urinary levels of OHCotGluc, NNO or ∑Nic exhibited a non-linear dose-response relationship with cognitive function in the urban elderly.

Association Between Carbon Monoxide Intoxication and Incidence of Ischemic Stroke: A Retrospective Nested Case-Control Study in South Korea

OBJECTIVES

Severe neurological sequelae occur in patients with carbon monoxide (CO) intoxication; however, whether the latter increases the long-term risk of developing ischemic stroke is unclear. We investigated the association between CO intoxication and ischemic stroke using data from the Korean National Health Information Database.

MATERIALS AND METHODS

We performed a retrospective, nested case-control study of 27,984 individuals treated for CO intoxication and 27,984 sex- and age-matched controls. Initially, we calculated the overall incidence and hazard ratio (HR) of ischemic stroke using conditional logistic regression. Thereafter, we calculated the incidences and HRs according to covariates and follow-up periods.

RESULTS

The CO intoxication group had a significantly higher risk of developing ischemic stroke than the control group (adjusted HR 2.31, 95% CI [confidence interval] = 2.01-2.65). Male sex (adjusted HR 2.73, 95% CI = 2.23-3.34), age <40 (adjusted HR 3.53, 95% CI = 2.15-5.82), low income (adjusted HR 2.55, 95% CI = 1.56-4.15), comorbidities (adjusted HR 2.59, 95% CI = 1.48-4.52), and current smokers (adjusted HR 3.55, 95% CI = 1.67-7.60) had a higher risk of ischemic stroke. The risk of ischemic stroke was highest within 2 years after CO intoxication (adjusted HR 7.47, 95% CI = 2.76-20.26), and even >6 years after, the risk remained significantly higher than in the control group (adjusted HR 1.84, 95% CI = 1.53-2.20).

CONCLUSIONS

CO intoxication and the long-term risk of ischemic stroke are associated.

Flexible derivative time-varying model in matched case-crossover studies for a small number of geographical locations among the participants

In matched case-crossover studies, any stratum effect is removed by conditioning on the fixed number of case-control sets in the stratum, and hence, the conditional logistic regression model is not able to detect any effects associated with matching covariates. However, some matching covariates such as time and location often modify the effect of covariates, making the estimations obtained by conditional logistic regression incorrect. Therefore, in this paper, we propose a flexible derivative time-varying coefficient model to evaluate effect modification by time and location, in order to make correct statistical inference, when the number of locations is small. Our proposed model is developed under the Bayesian hierarchical model framework and allows us to simultaneously detect relationships between the predictor and binary outcome and between the predictor and time. Inference is proposed based on the derivative function of the estimated function to determine whether there is an effect modification due to time and/or location, for a small number of locations among the participants. We demonstrate the accuracy of the estimation using a simulation study and an epidemiological example of a 1-4 bidirectional case-crossover study of childhood aseptic meningitis with drinking water turbidity.

A matched case-control study of risk factors associated with multiple sclerosis in Kuwait

Background

Genetic and environmental factors seem to have etiologic roles in multiple sclerosis (MS). Kuwait is regarded as medium to high risk country for MS. However, there is a paucity of published data on the risk factors for MS in Kuwait. Therefore, this matched case-control study examined the association between various factors including family history, stressful life events, exposure to tobacco smoke, vaccination history, comorbidities and MS risk in Kuwait.

Methods

Confirmed 110 MS cases and age (± 5 years), gender and nationality matched controls (1:1) were enrolled. A pre-tested structured questionnaire was used to collect the data through face-to-face interviews both from cases and controls. Conditional logistic regression was used to analyze the data.

Results

Among both cases and controls, majority were Kuwaiti (82.7%), and female (76.4%). Multivariable model showed that cases compared to controls were significantly more likely to have had a family history of MS (adjusted matched odds ratio (mOR_adj) = 5.1; 95% CI: 2.1–12.4; p < 0.001) or less likely to have been vaccinated against influenza A and B viruses before MS onset (mOR_adj = 0.4; 95% CI: 0.2–0.8; p = 0.010). None of the other variables considered were significantly related to MS status in this study.

Conclusions

Family history of MS had significantly direct, whereas, vaccination against influenza A and B viruses had inverse associations with MS status. Future studies may contemplate to verify the observed results.

Statistical methods for biomarker data pooled from multiple nested case-control studies

Pooling biomarker data across multiple studies allows for examination of a wider exposure range than generally possible in individual studies, evaluation of population subgroups and disease subtypes with more statistical power, and more precise estimation of biomarker-disease associations. However, circulating biomarker measurements often require calibration to a single reference assay prior to pooling due to assay and laboratory variability across studies. We propose several methods for calibrating and combining biomarker data from nested case-control studies when reference assay data are obtained from a subset of controls in each contributing study. Specifically, we describe a two-stage calibration method and two aggregated calibration methods, named the internalized and full calibration methods, to evaluate the main effect of the biomarker exposure on disease risk and whether that association is modified by a potential covariate. The internalized method uses the reference laboratory measurement in the analysis when available and otherwise uses the estimated value derived from calibration models. The full calibration method uses calibrated biomarker measurements for all subjects, including those with reference laboratory measurements. Under the two-stage method, investigators complete study-specific analyses in the first stage followed by meta-analysis in the second stage. Our results demonstrate that the full calibration method is the preferred aggregated approach to minimize bias in point estimates. We also observe that the two-stage and full calibration methods provide similar effect and variance estimates but that their variance estimates are slightly larger than those from the internalized approach. As an illustrative example, we apply the three methods in a pooling project of nested case-control studies to evaluate (i) the association between circulating vitamin D levels and risk of stroke and (ii) how body mass index modifies the association between circulating vitamin D levels and risk of cardiovascular disease.

Assessing crash risk considering vehicle interactions with trucks using point detector data

Trucks have distinct driving characteristics in general traffic streams such as lower speeds and limitations in acceleration and deceleration. As a consequence, vehicles keep longer headways or frequently change lane when they follow a truck, which is expected to increase crash risk. This study introduces several traffic measures at the individual vehicle level to capture vehicle interactions between trucks and non-trucks and analyzed how the measures affect crash risk under different traffic conditions. The traffic measures were developed using headways obtained from Inductive Loop Detectors (ILDs). In addition, a truck detection algorithm using a Gaussian Mixture (GM) model was developed to identify trucks and to estimate truck exposure from ILD data. Using the identified vehicle types from the GM model, vehicle interaction metrics were categorized into three groups based on the combination of leading and following vehicle types. The effects of the proposed traffic measures on crash risk were modeled in two different cases of prior- and non-crash using a case-control approach utilizing a conditional logistic regression. Results showed that the vehicle interactions between the leading and following vehicle types were highly associated with crash risk, and further showed different impacts on crash risk by traffic conditions. Specifically, crashes were more likely to occur when a truck following a non-truck had shorter average headway but greater headway variance in heavy traffic while a non-truck following a truck had greater headway variance in light traffic. This study obtained meaningful conclusions that vehicle interactions involved with trucks were significantly related to the crash likelihood rather than the measures that estimate average traffic condition such as total volume or average headway of the traffic stream.

Diet, obesity and colorectal carcinoma risk: results from a national cancer registry-based middle-eastern study

BACKGROUND

Cancer of colon and rectum (colorectal) is one of the most common cancers worldwide. There is a scarcity of published data on the risk factors for colorectal cancer (CRC) from the Middle-Eastern countries specifically in Kuwait. Therefore, this matched case-control study sought to examine the risk factors associated with CRC in Kuwait.

METHODS

One hundred and three histopathologically confirmed colorectal cancer cases were recruited from Kuwait Cancer Control Centre Registry. Two hundred and six controls matched with cases (2:1 ratio) on age, gender and nationality were selected from medical, ophthalmology, orthopedic and/ or surgical out-patient clinics at three main general hospitals in Kuwait. A structured questionnaire was used to collect the data from cases and controls through face-to-face interview. Adjusted matched odds ratios (mOR_adj) and their 95% confidence intervals (CI) were estimated using a multivariable conditional logistic regression model.

RESULTS

Multivariable conditional logistic regression model showed that cases were 4.3 times more likely to have had attainted obesity (BMI ≥ 30) in their lifetime compared to controls (mOR_adj = 4.3; 95% CI: 1.6-11.4). Compared to controls, cases rarely consumed fruits and vegetable (mOR_adj = 20.8; 95% CI: 4.4-99.5), tended to consume red meat 2-3 times a week (mOR_adj = 3.8; 95% CI: 1.6-8.7) or more than 4 times a week (mOR_adj = 9.4; 95% CI: 2.5-35.4). Reportedly cases compared to controls frequently (nearly every week) suffered from constipation (mOR_adj = 5.6; 95% CI: 1.9-16.5). However, CRC cases were less likely than controls to have been diagnosed in the past with hypercholesterolemia (mOR_adj = 0.3; 95% CI: 0.2-0.7) or diabetes mellitus type II (mOR_adj = 0.4; 95% CI: 0.2-0.8).

CONCLUSIONS

Obesity, excessive red meat consumption and infrequent fruits/vegetables intake were associated with an increased CRC risk. Overcoming identified pitfalls in dietary pattern and maintenance of healthy weight may help minimize CRC risk in Kuwait and perhaps other countries in the region. Further studies on genetic basis in conjunction with life styles and dietary factors may unravel their joint contributions to CRC risk and furnish tools for curtailing CRC risk in this and other similar populations.

Valid and efficient subgroup analyses using nested case-control data

BACKGROUND

It is not uncommon for investigators to conduct further analyses of subgroups, using data collected in a nested case-control design. Since the sampling of the participants is related to the outcome of interest, the data at hand are not a representative sample of the population, and subgroup analyses need to be carefully considered for their validity and interpretation.

METHODS

We performed simulation studies, generating cohorts within the proportional hazards model framework and with covariate coefficients chosen to mimic realistic data and more extreme situations. From the cohorts we sampled nested case-control data and analysed the effect of a binary exposure on a time-to-event outcome in subgroups defined by a covariate (an independent risk factor, a confounder or an effect modifier) and compared the estimates with the corresponding subcohort estimates. Cohort analyses were performed with Cox regression, and nested case-control samples or restricted subsamples were analysed with both conditional logistic regression and weighted Cox regression.

RESULTS

For all studied scenarios, the subgroup analyses provided unbiased estimates of the exposure coefficients, with conditional logistic regression being less efficient than the weighted Cox regression.

CONCLUSIONS

For the study of a subpopulation, analysis of the corresponding subgroup of individuals sampled in a nested case-control design provides an unbiased estimate of the effect of exposure, regardless of whether the variable used to define the subgroup is a confounder, effect modifier or independent risk factor. Weighted Cox regression provides more efficient estimates than conditional logistic regression.

Comments on: blood product transfusion in emergency department patients: a case control study of practice patterns and impact on outcome

Clinical decision makings according studies result require the valid and correct data collection, andanalysis. However, there are some common methodological and statistical issues which may ignore by authors. In individual matched case- control design bias arising from the unconditional analysis instead of conditional analysis. Using an unconditional logistic for matched data causes the imposition of a large number of nuisance parameters which may result in seriously biased estimates.

Addressing data privacy in matched studies via virtual pooling

BACKGROUND

Data confidentiality and shared use of research data are two desirable but sometimes conflicting goals in research with multi-center studies and distributed data. While ideal for straightforward analysis, confidentiality restrictions forbid creation of a single dataset that includes covariate information of all participants. Current approaches such as aggregate data sharing, distributed regression, meta-analysis and score-based methods can have important limitations.

METHODS

We propose a novel application of an existing epidemiologic tool, specimen pooling, to enable confidentiality-preserving analysis of data arising from a matched case-control, multi-center design. Instead of pooling specimens prior to assay, we apply the methodology to virtually pool (aggregate) covariates within nodes. Such virtual pooling retains most of the information used in an analysis with individual data and since individual participant data is not shared externally, within-node virtual pooling preserves data confidentiality. We show that aggregated covariate levels can be used in a conditional logistic regression model to estimate individual-level odds ratios of interest.

RESULTS

The parameter estimates from the standard conditional logistic regression are compared to the estimates based on a conditional logistic regression model with aggregated data. The parameter estimates are shown to be similar to those without pooling and to have comparable standard errors and confidence interval coverage.

CONCLUSIONS

Virtual data pooling can be used to maintain confidentiality of data from multi-center study and can be particularly useful in research with large-scale distributed data.

Air pollution by particulate matter PM10 may trigger multiple sclerosis relapses

BACKGROUND

Seasonal variation of relapses in multiple sclerosis (MS) suggests that season-dependent factors, such as ambient air pollution, may trigger them. However, only few studies have considered possible role of air pollutants as relapse's risk factor.

OBJECTIVE

We investigated the effect of particulate matter of aerodynamic diameter smaller than 10µm (PM₁₀) on MS relapses.

METHODS

In total, 536 relapsing MS patients from Strasbourg city (France) were included, accounting for 2052 relapses over 2000-2009 period. A case-crossover design was used with cases defined as the days of relapse and controls being selected in the same patient at plus and minus 35 days. Different lags from 0 to 30 days were considered. Conditional logistic regressions, adjusted on meteorological parameters, school and public holidays, were used and exposure was considered first as a quantitative variable and second, as a binary variable.

RESULTS

The natural logarithm of the average PM₁₀ concentration lagged from 1 to 3 days before relapse onset was significantly associated with relapse risk (OR =1.40 [95% confidence interval 1.08-1.81]) in cold season. Consistent results were observed when considering PM₁₀ as a binary variable, even if not significant.

CONCLUSION

With an appropriate study design and robust ascertainment of neurological events and exposure, the present study highlights the effect of PM₁₀ on the risk of relapse in MS patients, probably through oxidative stress mechanisms.

Methodologic considerations in the design and analysis of nested case-control studies: association between cytokines and postoperative delirium

Background

The nested case-control study (NCC) design within a prospective cohort study is used when outcome data are available for all subjects, but the exposure of interest has not been collected, and is difficult or prohibitively expensive to obtain for all subjects. A NCC analysis with good matching procedures yields estimates that are as efficient and unbiased as estimates from the full cohort study. We present methodological considerations in a matched NCC design and analysis, which include the choice of match algorithms, analysis methods to evaluate the association of exposures of interest with outcomes, and consideration of overmatching.

Methods

Matched, NCC design within a longitudinal observational prospective cohort study in the setting of two academic hospitals. Study participants are patients aged over 70 years who underwent scheduled major non-cardiac surgery. The primary outcome was postoperative delirium from in-hospital interviews and medical record review. The main exposure was IL-6 concentration (pg/ml) from blood sampled at three time points before delirium occurred. We used nonparametric signed ranked test to test for the median of the paired differences. We used conditional logistic regression to model the risk of IL-6 on delirium incidence. Simulation was used to generate a sample of cohort data on which unconditional multivariable logistic regression was used, and the results were compared to those of the conditional logistic regression. Partial R-square was used to assess the level of overmatching.

Results

We found that the optimal match algorithm yielded more matched pairs than the greedy algorithm. The choice of analytic strategy—whether to consider measured cytokine levels as the predictor or outcome-- yielded inferences that have different clinical interpretations but similar levels of statistical significance. Estimation results from NCC design using conditional logistic regression, and from simulated cohort design using unconditional logistic regression, were similar. We found minimal evidence for overmatching.

Conclusions

Using a matched NCC approach introduces methodological challenges into the study design and data analysis. Nonetheless, with careful selection of the match algorithm, match factors, and analysis methods, this design is cost effective and, for our study, yields estimates that are similar to those from a prospective cohort study design.

Environmental and individual attributes associated with child maltreatment resulting in hospitalization or death

Maltreatment continues to be a leading cause of death for young children. Researchers are beginning to uncover which neighborhood attributes may be associated with maltreatment outcomes. However, few studies have been able to explore these influences while controlling for individual family attributes, and none have been able to parse out the most severe outcomes-injuries resulting in hospitalization or death. This study utilizes a retrospective, case-control design on a dataset containing both individual and environmental level attributes of children who have been hospitalized or died due to maltreatment to explore the relative influence of attributes inside and outside the household walls. Binary conditional logistic regression was used to model the outcome as a function of the individual and environmental level predictors. Separate analyses also separated the outcome by manner of maltreatment: abuse or neglect. Finally, a sub-analysis included protective predictors representing access to supportive resources. Findings indicate that neighborhood attributes were similar for both cases and controls, except in the neglect only model, wherein impoverishment was associated with higher odds of serious maltreatment. Dense housing increased risk in all models except the neglect only model. In a sub-analysis, distance to Family Resource Centers was inversely related to serious maltreatment. In all models, variables representing more extreme intervention and/or removal of the victim and/or perpetrator from the home (foster care or criminal court involvement) were negatively associated with the risk of becoming a case. Medi-Cal insurance eligibility of a child was also negatively associated with becoming a case. Government interventions may be playing a critical role in child protection. More research is needed to ascertain how these interventions assert their influence.

Detecting Familial Aggregation

In addition to characterizing the distribution of genetic features of populations (mutation and allele frequencies; measures of Hardy-Weinberg equilibrium), genetic epidemiology and statistical genetics aim to explore and define the role of genomic variation in risk of disease or variation in traits of interest. To facilitate this kind of exploration, genetic epidemiology and statistical genetics address a series of questions: 1. Does the disease tend to cluster in families more than expected by chance alone? 2. Does the disease appear to follow a particular genetic model of transmission in families? 3. Does variation at a particular genomic position tend to cosegregate with disease in families? 4. Do specific genetic variants tend to be carried more frequently by those with disease than by those without these variants in a given population (or across families)? The first question can be examined using studies of familial aggregation or correlation. An ancillary question: "how much of the susceptibility to disease (or variation in disease-related traits) might be accounted for by genetic factors?" is typically answered by estimating heritability, the proportion of variance in a trait or in risk to a disease attributable to genetics. The second question can be formally tested using pedigrees for which disease affection status or trait values are available through a modeling approach known as segregation analysis. The third question can be answered with data on genomic markers in pedigrees with affected members informative for linkage, where meiotic cross-over events are estimated or assessed. The fourth question is answerable using genotype data on genomic markers on unrelated affected and unaffected individuals and/or families with affected members and unaffected members. All of these questions can also be explored for quantitative (or continuously distributed) traits by examining variation in trait values between family members or between unrelated individuals. While each of these questions and the analytical approaches for answering them is explored extensively in subsequent chapters (heritability in Chapters 8 and 9 ; segregation in Chapter 12 ; linkage in Chapters 13 - 17 ; and association in Chapters 18 - 20 ), this chapter focuses on statistical methods to address questions of familial aggregation of qualitative phenotypes (e.g., disease status) or quantitative phenotypes.While studies exploring genotype-phenotype correlations are arguably the most important and common type of statistical genetic study performed, these studies are performed under the assumption that genetic contributors at least partially explain risk of a disease or a trait of interest. This may not always be the case, especially with diseases or traits known to be strongly influenced by environmental factors. For this reason, before any of the last three questions described above can be answered, it is important to ask first whether the disease clusters among family members more than unrelated persons, as this constitutes evidence of a possible heritable contribution to disease, justifying the pursuit of studies answering the other questions. In this chapter, the underlying principles of familial aggregation studies are addressed to provide an understanding and set of analytical tools to help answer the question if diseases or traits of interest are likely to be heritable and therefore justify subsequent statistical genetic studies to identify specific genetic causes.

Association between flood and the morbidity of bacillary dysentery in Zibo City, China: a symmetric bidirectional case-crossover study

This study examined the relationship between daily morbidity of bacillary dysentery and flood in 2007 in Zibo City, China, using a symmetric bidirectional case-crossover study. Odds ratios (ORs) and 95 % confidence intervals (CIs) on the basis of multivariate model and stratified analysis at different lagged days were calculated to estimate the risk of flood on bacillary dysentery. A total of 902 notified bacillary dysentery cases were identified during the study period. The median of case distribution was 7-year-old and biased to children. Multivariable analysis showed that flood was associated with an increased risk of bacillary dysentery, with the largest OR of 1.849 (95 % CI 1.229-2.780) at 2-day lag. Gender-specific analysis showed that there was a significant association between flood and bacillary dysentery among males only (ORs >1 from lag 1 to lag 5), with the strongest lagged effect at 2-day lag (OR = 2.820, 95 % CI 1.629-4.881), and the result of age-specific indicated that youngsters had a slightly larger risk to develop flood-related bacillary dysentery than older people at one shorter lagged day (OR = 2.000, 95 % CI 1.128-3.546 in youngsters at lag 2; OR = 1.879, 95 % CI 1.069-3.305 in older people at lag 3). Our study has confirmed that there is a positive association between flood and the risk of bacillary dysentery in selected study area. Males and youngsters may be the vulnerable and high-risk populations to develop the flood-related bacillary dysentery. Results from this study will provide recommendations to make available strategies for government to deal with negative health outcomes due to floods.

A case-base sampling method for estimating recurrent event intensities

Case-base sampling provides an alternative to risk set sampling based methods to estimate hazard regression models, in particular when absolute hazards are also of interest in addition to hazard ratios. The case-base sampling approach results in a likelihood expression of the logistic regression form, but instead of categorized time, such an expression is obtained through sampling of a discrete set of person-time coordinates from all follow-up data. In this paper, in the context of a time-dependent exposure such as vaccination, and a potentially recurrent adverse event outcome, we show that the resulting partial likelihood for the outcome event intensity has the asymptotic properties of a likelihood. We contrast this approach to self-matched case-base sampling, which involves only within-individual comparisons. The efficiency of the case-base methods is compared to that of standard methods through simulations, suggesting that the information loss due to sampling is minimal.

Short-term effects of floods on Japanese encephalitis in Nanchong, China, 2007-2012: A time-stratified case-crossover study

This time-stratified case-crossover study aimed to quantify the impact of floods on daily Japanese encephalitis (JE) cases from 2007 to 2012 in Nanchong city of Sichuan Province, China. Using conditional logistic regression analysis, we calculated the odds ratios (ORs) and 95% confidence intervals (CIs) at different lagged days, adjusting for daily average temperature (AT) and daily average relative humidity (ARH). A total of 370 JE cases were notified during the study period, with the median patient age being 4.2years. The seasonal pattern of JE cases clustered in July and August during the study period. Floods were significantly associated with an increased number of JE cases from lag 23 to lag 24, with the strongest lag effect at lag 23 (OR=2.00, 95% CI: 1.14-3.52). Similarly, AT and ARH were positively associated with daily JE cases from lag 0 to lag 8 and from lag 0 to lag 9, respectively. Floods, with AT and ARH, can be used to forecast JE outbreaks in the study area. Based on the results of this study, recommendations include undertaking control measures before the number of cases increases, especially for regions with similar geographic, climatic, and socio-economic conditions as those in the study area.

The case-crossover design via penalized regression

BACKGROUND

The case-crossover design is an attractive alternative to the classical case-control design which can be used to study the onset of acute events if the risk factors of interest vary in time. By comparing exposures within cases at different time periods, the case-crossover design does not rely on control subjects which can be difficult to acquire. However, using the standard method of maximum likelihood, resulting risk estimates can be heavily biased when the prevalence to risk factors is very low (or very high).

METHODS

To overcome the problem of low risk factor prevalences, penalized conditional logistic regression via the lasso (least absolute shrinkage and selection operator) has been proposed in the literature as well as related methods such as the Firth correction. We apply and compare several penalized regression approaches in the context of a case-crossover analysis of the European Study of Severe Cutaneous Adverse Reactions (EuroSCAR; 1997-2001).

RESULTS

Out of 30 drugs, standard methods only correctly classified 17 drugs (including some highly implausible risk estimates), while penalized methods correctly classified 22 drugs.

CONCLUSION

Penalized methods generally yield better risk classifications and much more plausible risk estimates for the EuroSCAR study than standard methods. As these novel techniques can be easily implemented using available R packages, we encourage routine use of penalized conditional logistic regression for case-crossover data.

Quantitative analysis of burden of bacillary dysentery associated with floods in Hunan, China

BACKGROUND

Jishou and Huaihua, two cities in the west of Hunan Province, had suffered from severe floods because of long-lasting and heavy rainfall during the end of June and July 2012. However, the Disability Adjusted of Life Years (DALYs) of bacillary dysentery caused by the floods have not been examined before. The study aimed to quantify the impact of the floods on the burden of bacillary dysentery in Hunan, China.

METHODS

A unidirectional case-crossover study was firstly conducted to determine the relationship between daily cases of bacillary dysentery and the floods in Jishou and Huaihua of Hunan Province in 2012. Odds ratios (ORs) estimated by conditional logistic regression were used to quantify the risk of the floods on the disease. The years lived with disability (YLDs) of bacillary dysentery attributable to floods were then estimated based on the WHO framework to calculate potential impact fraction in the Burden of Disease study.

RESULTS

Multivariable analysis showed that floods were significantly associated with an increased risk of the number of cases of bacillary dysentery (OR=3.270, 95% CI: 1.299-8.228 in Jishou; OR=2.212, 95% CI: 1.052-4.650 in Huaihua). The strongest effect was shown with a 1-day lag in Jishou and a 4-day lag in Huaihua. Attributable YLD per 1000 of bacillary dysentery due to the floods was 0.0296 in Jishou and 0.0157 in Huaihua.

CONCLUSIONS

Our study confirms that floods have significantly increased the risks of bacillary dysentery in the study areas. In addition, a sudden and severe flooding with a shorter duration may cause more burdens of bacillary dysentery than a persistent and moderate flooding. Public health preparation and intervention programs should be taken to reduce and prevent a potential risk of bacillary dysentery epidemics after floods.

Analytic power and sample size calculation for the genotypic transmission/disequilibrium test in case-parent trio studies

Case-parent trio studies considering genotype data from children affected by a disease and their parents are frequently used to detect single nucleotide polymorphisms (SNPs) associated with disease. The most popular statistical tests for this study design are transmission/disequilibrium tests (TDTs). Several types of these tests have been developed, for example, procedures based on alleles or genotypes. Therefore, it is of great interest to examine which of these tests have the highest statistical power to detect SNPs associated with disease. Comparisons of the allelic and the genotypic TDT for individual SNPs have so far been conducted based on simulation studies, since the test statistic of the genotypic TDT was determined numerically. Recently, however, it has been shown that this test statistic can be presented in closed form. In this article, we employ this analytic solution to derive equations for calculating the statistical power and the required sample size for different types of the genotypic TDT. The power of this test is then compared with the one of the corresponding score test assuming the same mode of inheritance as well as the allelic TDT based on a multiplicative mode of inheritance, which is equivalent to the score test assuming an additive mode of inheritance. This is, thus, the first time the power of these tests are compared based on equations, yielding instant results and omitting the need for time-consuming simulation studies. This comparison reveals that these tests have almost the same power, with the score test being slightly more powerful.

Variable selection and prediction using a nested, matched case-control study: Application to hospital acquired pneumonia in stroke patients

Matched case-control designs are commonly used in epidemiologic studies for increased efficiency. These designs have recently been introduced to the setting of modern imaging and genomic studies, which are characterized by high-dimensional covariates. However, appropriate statistical analyses that adjust for the matching have not been widely adopted. A matched case-control study of 430 acute ischemic stroke patients was conducted at Massachusetts General Hospital (MGH) in order to identify specific brain regions of acute infarction that are associated with hospital acquired pneumonia (HAP) in these patients. There are 138 brain regions in which infarction was measured, which introduce nearly 10,000 two-way interactions, and challenge the statistical analysis. We investigate penalized conditional and unconditional logistic regression approaches to this variable selection problem that properly differentiate between selection of main effects and of interactions, and that acknowledge the matching. This neuroimaging study was nested within a larger prospective study of HAP in 1915 stroke patients at MGH, which recorded clinical variables, but did not include neuroimaging. We demonstrate how the larger study, in conjunction with the nested, matched study, affords us the capability to derive a score for prediction of HAP in future stroke patients based on imaging and clinical features. We evaluate the proposed methods in simulation studies and we apply them to the MGH HAP study.

Nonparametric Diagnostic Test for Conditional Logistic Regression

The use of conditional logistic regression models to analyze matched case-control data has become standard in statistical analysis. However, methods to test the fit of these models has primarily focused on influential observations and the presence of outliers, while little attention has been given to the functional form of the covariates. In this paper we present methods to test the functional form of the covariates in the conditional logistic regression model, these methods are based on nonparametric smoothers. We assess the performance of the proposed methods via simulation studies and illustrate an example of their use on data from a community based intervention.

Correlation between dietary factors and the risk of pancreatic cancer

AIM: To explore the association between dietary factors and the risk of pancreatic cancer to provide a scientific basis for prevention of pancreatic cancer through diet and lifestyle changes.

METHODS: A case-control study involving 97 patients with pancreatic cancer and 194 controls was conducted. Controls were matched to cases for age and sex. All of them were interviewed with uniform questionnaires. Conditional logistic regression was used for univariate and multivariate analysis.

RESULTS: The development of pancreatic cancer was positively associated with intake of desserts (OR = 4.706), but negatively with intake of onion (OR = 0.068), yam (OR = 0.301), sweet potato (OR = 0.178), and fruit (OR = 0.299).

CONCLUSION: Dietary factors may play an important role in the development of pancreatic cancer.