Determinants of polycystic ovary syndrome: A matched case-control study

BACKGROUND

Polycystic ovarian syndrome (PCOS) is a multifaceted endocrine disorder of women of reproductive age with a multifactorial aetiology. Despite much research, there is still inconclusive data on the impact of dietary, lifestyle and socio-economic factors on PCOS aetiology. Thus, the present study explored the association of PCOS with diet, eating behaviour, other lifestyle and socio-economic factors.

METHODS

A matched-pair case-control study was conducted on 150 women with PCOS and 150 healthy controls. Information on diet, eating behaviour and physical activity, and also anthropometric and socio-economic data were collected through standard questionnaires. The adjusted odds ratios (AmOR) were calculated and reported using conditional multivariable logistic regression.

RESULTS

The results showed low education level (AmOR = 8.44; 95% confidence interval [CI] = 1.63-43.68), high sugar consumption (AmOR = 11.61; 95% CI = 2.05-65.72) along with higher body mass index (BMI) and inactivity to be significantly associated with PCOS. Also, a significant protective effect was found for cognitive dietary restraint (AmOR = 0.79; 95% CI = 0.66-0.93), crude fibre (AmOR = 0.61; 95% CI = 0.45-0.82) and protein intake.

CONCLUSIONS

Low education status may contribute to higher receptiveness to choosing unhealthy diets and lifestyles, resulting in adiposity and an increased risk of PCOS.

Association Between Abnormal Metabolic Parameters and Receiving Subsequent Interventions in Children and Adolescents Initiating Second-Generation Antipsychotics

Objectives: This study aimed to examine the association between abnormal readings of metabolic parameters detected during second-generation antipsychotic (SGA) treatment and the likelihood of receiving subsequent adverse drug event interventions. Methods: This was a nested case-control study conducted on patients 1-17 years of age with at least two prescriptions of SGAs between January 2010 and January 2019 using TriNetX EMR data. Following an incident density sampling procedure, patients who received the SGA metabolic adverse event intervention (mAEI) (case) were matched with three nonrecipients (controls). The abnormal readings of metabolic parameters within 30 days before the initiation of mAIEs were further identified. These metabolic parameters include body mass index (BMI) and laboratory parameters such as cholesterol, high-density lipoprotein, low-density lipoprotein, triglycerides, blood glucose, HbA1c, thyroid hormones, liver enzymes, and prolactin. The association of abnormal metabolic parameters with subsequent mAEIs was assessed using a conditional logistic regression model, after adjusting for demographic and other clinical risk factors. Results: One thousand eight hundred eighty-four children and adolescents met the inclusion criteria and were prescribed SGA mAEIs. The most common types of mAEIs prescribed were weight management pharmacotherapy (40.6%), switching from a high or medium metabolic risk profile SGA to a low-risk one (30.9%), nonpharmacological treatment (25.4%), and switching from SGA polytherapy to monotherapy (11.7%). The conditional logistic regression analysis on matched mAEI recipients and nonrecipients showed that patients with an abnormal BMI had 43% higher odds of receiving mAEI (odds ratio [95% confidence interval]: 1.43 [1.13-1.79]). However, the presence of an abnormal laboratory reading was not associated with the initiation of mAEIs. Conclusions: The prescribing of mAEIs were associated with the presence of obesity, but not with abnormal readings of other metabolic parameters, suggesting that additional data are needed to clarify the long-term implication of SGA metabolic adverse events other than weight gain and to inform the appropriate timing for interventions.

Sporadic Use of Antibiotics in Older Adults and the Risk of Dementia: A Nested Case-Control Study Based on German Health Claims Data

BACKGROUND

Antibiotics for systemic use may increase the risk of neurodegeneration, yet antibiotic therapy may be able to halt or mitigate an episode of neurodegenerative decline.

OBJECTIVE

To investigate the association of sporadic use of antibiotics and subsequent dementia risk (including Alzheimer's disease).

METHODS

We used data from the largest public health insurance fund in Germany, the Allgemeine Ortskrankenkasse (AOK). Each of the 35,072 dementia cases aged 60 years and older with a new dementia diagnosis during the observation period from 2006 to 2018 was matched with two control-patients by age, sex, and time since 2006. We ran conditional logistic regression models for dementia risk in terms of odds ratios (OR) as a function of antibiotic use for the entire antibiotic group and for each antibiotic subgroup. We controlled for comorbidities, need for long-term care, hospitalizations, and nursing home placement.

RESULTS

Antibiotic use was positively associated with dementia (OR = 1.18, 95% confidence interval (95% CI):1.14-1.22), which became negative after adjustment for comorbidities, at least one diagnosis of bacterial infection or disease, and covariates (OR = 0.93, 95% CI:0.90-0.96). Subgroups of antibiotics were also negatively associated with dementia after controlling for covariates: tetracyclines (OR = 0.94, 95% CI:0.90-0.98), beta-lactam antibacterials, penicillins (OR = 0.93, 95% CI:0.90-0.97), other beta-lactam antibacterials (OR = 0.92, 95% CI:0.88-0.95), macrolides, lincosamides, and streptogramins (OR = 0.88, 95% CI:0.85-0.92), and quinolone antibacterials (OR = 0.96, 95% CI:0.92-0.99).

CONCLUSION

Our results suggest that there was a decreased likelihood of dementia for preceding antibiotic use. The benefits of antibiotics in reducing inflammation and thus the risk of dementia need to be carefully weighed against the increase in antibiotic resistance.

Conditional concordance-assisted learning under matched case-control design for combining biomarkers for population screening

Incorporating promising biomarkers into cancer screening practices for early-detection is increasingly appealing because of the unsatisfactory performance of current cancer screening strategies. The matched case-control design is commonly adopted in biomarker development studies to evaluate the discriminative power of biomarker candidates, with an intention to eliminate confounding effects. Data from matched case-control studies have been routinely analyzed by the conditional logistic regression, although the assumed logit link between biomarker combinations and disease risk may not always hold. We propose a conditional concordance-assisted learning method, which is distribution-free, for identifying an optimal combination of biomarkers to discriminate cases and controls. We are particularly interested in combinations with a clinically and practically meaningful specificity to prevent disease-free subjects from unnecessary and possibly intrusive diagnostic procedures, which is a top priority for cancer population screening. We establish asymptotic properties for the derived combination and confirm its favorable finite sample performance in simulations. We apply the proposed method to the prostate cancer data from the carotene and retinol efficacy trial (CARET).

A Description of the Statistical Methods for the Vaccine Impact on Diarrhea in Africa (VIDA) Study

BACKGROUND

Diarrheal diseases remain a health threat to children in low- and middle-income countries. The Vaccine Impact on Diarrhea in Africa (VIDA) study was a 36-month, prospective, matched case-control study designed to estimate the etiology, incidence, and adverse clinical consequences of moderate-to-severe diarrhea (MSD) in children aged 0-59 months. VIDA was conducted following rotavirus vaccine introduction at 3 censused sites in sub-Saharan Africa that participated in the Global Enteric Multicenter Study (GEMS) ∼10 years earlier. We describe the study design and statistical methods of VIDA and where they differ from GEMS.

METHODS

We aimed to enroll 8-9 MSD cases every 2 weeks from sentinel health centers in 3 age strata (0-11, 12-23, 24-59 months) and 1 to 3 controls matched by age, sex, date of case enrollment, and village. Clinical, epidemiological, and anthropometric data were collected at enrollment and ∼60 days later. A stool specimen collected at enrollment was analyzed by both conventional methods and quantitative PCR for enteric pathogens. For the matched case-control study, we estimated the population-based, pathogen-specific attributable fraction (AF) and attributable incidence adjusted for age, site, and other pathogens, and identified episodes attributable to a specific pathogen for additional analyses. A prospective cohort study nested within the original matched case-control study allowed assessment of (1) the association between potential risk factors and outcomes other than MSD status and (2) the impact of MSD on linear growth.

CONCLUSIONS

GEMS and VIDA together comprise the largest and most comprehensive assessment of MSD conducted to date in sub-Saharan Africa populations at highest risk for morbidity and mortality from diarrhea. The statistical methods used in VIDA have endeavored to maximize the use of available data to produce more robust estimates of the pathogen-specific disease burden that might be prevented by effective interventions.

A tree-based modeling approach for matched case-control studies

Conditional logistic regression (CLR) is the indisputable standard method for the analysis of matched case-control studies. However, CLR is strongly restricted with respect to the inclusion of non-linear effects and interactions of confounding variables. A novel tree-based modeling method is proposed which accounts for this issue and provides a flexible framework allowing for a more complex confounding structure. The proposed machine learning model is fitted within the framework of CLR and, therefore, allows to account for the matched strata in the data. A simulation study demonstrates the efficacy of the method. Furthermore, for illustration the method is applied to a matched case-control study on cervical cancer.

SARS-CoV-2 Vaccine Breakthrough by Omicron and Delta Variants, New York, USA

Recently emerged SARS-CoV-2 variants have greater potential than earlier variants to cause vaccine breakthrough infections. During emergence of the Delta and Omicron variants, a matched case-control analysis used a viral genomic sequence dataset linked with demographic and vaccination information from New York, USA, to examine associations between virus lineage and patient vaccination status, patient age, vaccine type, and time since vaccination. Case-patients were persons infected with the emerging virus lineage, and controls were persons infected with any other virus lineage. Infections in fully vaccinated and boosted persons were significantly associated with the Omicron lineage. Odds of infection with Omicron relative to Delta generally decreased with increasing patient age. A similar pattern was observed with vaccination status during Delta emergence but was not significant. Vaccines offered less protection against Omicron, thereby increasing the number of potential hosts for emerging variants.

Estimating COVID-19 Vaccine Effectiveness for Skilled Nursing Facility Healthcare Personnel, California, USA

We estimated real-world vaccine effectiveness among skilled nursing facility healthcare personnel who were regularly tested for SARS-CoV-2 infection in California, USA, during January‒March 2021. Vaccine effectiveness for fully vaccinated healthcare personnel was 73.3% (95% CI 57.5%-83.3%). We observed high real-world vaccine effectiveness in this population.

Predicting the Risk of Microtia From Prenatal Factors: A Hospital-Based Case-Control Study

Background

Although a wide range of risk factors for microtia were identified, the limitation of these studies, however, is that risk factors were not estimated in comparison with one another or from different domains. Our study aimed to uncover which factors should be prioritized for the prevention and intervention of non-syndromic microtia via tranditonal and meachine-learning statistical methods.

Methods

293 pairs of 1:1 matched non-syndromic microtia cases and controls who visited Shanghai Ninth People's Hospital were enrolled in the current study during 2017-2019. Thirty-nine risk factors across four domains were measured (i.e., parental sociodemographic characteristics, maternal pregnancy history, parental health conditions and lifestyles, and parental environmental and occupational exposures). Lasso regression model and multivariate conditional logistic regression model were performed to identify the leading predictors of microtia across the four domains. The area under the curve (AUC) was used to calculate the predictive probabilities.

Results

Eight predictors were identified by the lasso regression, including abnormal pregnancy history, genital system infection, teratogenic drugs usage, folic acid supplementation, paternal chronic conditions history, parental exposure to indoor decoration, paternal occupational exposure to noise and maternal acute respiratory infection. The additional predictors identified by the multivariate conditional logistic regression model were maternal age and maternal occupational exposure to heavy metal. Predictors selected from the conditional logistic regression and lasso regression both yielded AUCs (95% CIs) of 0.83 (0.79-0.86).

Conclusion

The findings from this study suggest some factors across multiple domains are key drivers of non-syndromic microtia regardless of the applied statistical methods. These factors could be used to generate hypotheses for further observational and clinical studies on microtia and guide the prevention and intervention strategies for microtia.

Introduction to Matching in Case-Control and Cohort Studies

Matching is a technique through which patients with and without an outcome of interest (in case-control studies) or patients with and without an exposure of interest (in cohort studies) are sampled from an underlying cohort to have the same or similar distributions of some characteristics. This technique is used to increase the statistical efficiency and cost efficiency of studies. In case-control studies, besides time in risk set sampling, controls are often matched for each case with respect to important confounding factors, such as age and sex, and covariates with a large number of values or levels, such as area of residence (e.g., post code) and clinics/hospitals. In the statistical analysis of matched case-control studies, fixed-effect models such as the Mantel-Haenszel odds ratio estimator and conditional logistic regression model are needed to stratify matched case-control sets and remove selection bias artificially introduced by sampling controls. In cohort studies, exact matching is used to increase study efficiency and remove or reduce confounding effects of matching factors. Propensity score matching is another matching method whereby patients with and without exposure are matched based on estimated propensity scores to receive exposure. If appropriately used, matching can improve study efficiency without introducing bias and could also present results that are more intuitive for clinicians.

Conditional or unconditional logistic regression for frequency matched case-control design?

Frequency matching is commonly used in epidemiological case control studies to balance the distributions of the matching factors between the case and control groups and to improve the efficiency of case-control designs. Applied researchers have held a common opinion that unconditional logistic regression should be used to analyze frequency matched designs and conditional logistic regression is unnecessary. However, the justification of this view is unclear. To compare the performances of ULR and CLR in terms of simplicity, unbiasedness, and efficiency in a more intuitive way, we viewed frequency matching from the perspective of weighted sampling and derived the outcome models describing how the exposure and matching factors are associated with the outcome in the matched data separately in two scenarios: (1) only categorical variables are used for matching; (2) continuous variables are categorized for matching. In either scenario the derived outcome model is a logit model with stratum-specific intercepts. Correctly specified unconditional logistic regression can be more efficient than conditional logistic regression, particularly when continuous matching factors are used, whereas conditional logistic regression is a more practical approach because it is less dependent on modeling choices.

Risk factors associated with acute hepatopancreatic necrosis disease at shrimp farm level in Bac Lieu Province, Vietnam

Background and Aim:

Acute hepatopancreatic necrosis disease (AHPND) is a severe disease in shrimp farms and adversely affected the shrimp industry of Vietnam. So far, the study on risk factors associated with AHPND outbreaks is limited. The objective of this study was to determine the potential risk factors of AHPND at the shrimp farm level in Bac Lieu Province, Vietnam.

Materials and Methods:

Real-time-Polymerase chain reaction was used to analyze data collected from an active surveillance program of shrimp farms in 2017 in the Vinh Tien and Vinh Lac villages, Vinh Thinh commune, Hoa Binh district in Bac Lieu Province, Vietnam. The matched case-control study selected 20 cases and 20 control farms from 134 shrimp farms. In 2018, face-to-face interviews using structured questionnaires were conducted with the farmers of these selected farms.

Results:

Of the 59 studied variables, seven had p≤0.2 based on bivariate analyses. The results of multivariable analysis showed that the presence of fish-eating birds on shrimp farms was a significant association with AHPND (odds ratio=8, p=0.049).

Conclusion:

To reduce the effect of AHPND, farmers should apply effective methods to manage wild animals such as using a grid or net to cover the pond, combined with improved biosecurity.

Enterovirus Infections Are Associated With the Development of Celiac Disease in a Birth Cohort Study

Enterovirus and adenovirus infections have been linked to the development of celiac disease. We evaluated this association in children who developed biopsy-proven celiac disease (N = 41) during prospective observation starting from birth, and in control children (N = 53) matched for the calendar time of birth, sex, and HLA-DQ genotype. Enterovirus and adenovirus infections were diagnosed by seroconversions in virus antibodies in longitudinally collected sera using EIA. Enterovirus infections were more frequent in case children before the appearance of celiac disease-associated tissue transglutaminase autoantibodies compared to the corresponding period in control children (OR 6.3, 95% CI 1.8-22.3; p = 0.005). No difference was observed in the frequency of adenovirus infections. The findings suggest that enterovirus infections may contribute to the process leading to celiac disease.

Habitat selection by a threatened desert amphibian

Habitat degradation and fragmentation are major drivers of amphibian declines. The loss of environmental features that allow for movement between water sources may be particularly detrimental for amphibians in arid environments. Climate changes will increase the importance of microhabitats to amphibians. Enhancing areas to facilitate movement may be a necessary conservation strategy for many animal species that depend on wetlands, including federally threatened Chiricahua leopard frogs (Lithobates chiricahuensis). Habitat preferences of this frog species are not well understood. We sought to better understand fine-scale habitat selection, to inform conservation of Chiricahua leopard frogs. We conducted our study on the Ladder Ranch, a privately owned working bison ranch in New Mexico, USA that supports a large proportion of the remaining Chiricahua leopard frogs in the state. We attached radio transmitters to 44 frogs during summer 2014. We located each frog daily for up to 8 weeks (median = 30 days). We assessed fine-scale habitat selection by comparing characteristics at each frog location and a random location 5 m away using conditional logistic regression. Frogs preferred features that likely reduce desiccation, even after accounting for the presence of water. Frogs selected areas with more low-lying cover, especially aquatic vegetation and woody debris, a tree overstory, and a mud substrate. We recommend managing potential movement corridors for Chiricahua leopard frogs by ensuring the presence of muddy creek bottoms, woody debris, riparian overstory, low-lying ground cover, and pools. Microclimates created by these features seem especially valuable given warming temperatures and modified precipitation regimes, resulting in decreased surface water, soil moisture, and vegetation cover. Retaining or creating preferred habitat features and microclimates in areas between water sources may increase connectivity among isolated populations of Chiricahua leopard frogs and could improve persistence and recovery of other water-obligate species in arid landscapes.

Causal mediation analysis in nested case-control studies using conditional logistic regression

The paper proposes an approach to causal mediation analysis in nested case-control study designs, often incorporated with countermatching schemes using conditional likelihood, and we compare the method's performance to that of mediation analysis using the Cox model for the full cohort with a continuous or dichotomous mediator. Simulation studies are conducted to assess our proposed method and investigate the efficiency relative to the cohort. We illustrate the method using actual data from two studies of potential mediation of radiation risk conducted within the Adult Health Study cohort of atomic-bomb survivors. The performance becomes comparable to that based on the full cohort, illustrating the potential for valid mediation analysis based on the reduced data obtained through the nested case-control design.

Approximate Bayesian inference for case-crossover models

A case-crossover analysis is used as a simple but powerful tool for estimating the effect of short-term environmental factors such as extreme temperatures or poor air quality on mortality. The environment on the day of each death is compared to the one or more "control days" in previous weeks, and higher levels of exposure on death days than control days provide evidence of an effect. Current state-of-the-art methodology and software (integrated nested Laplace approximation [INLA]) cannot be used to fit the most flexible case-crossover models to large datasets, because the likelihood for case-crossover models cannot be expressed in a manner compatible with this methodology. In this paper, we develop a flexible and scalable modeling framework for case-crossover models with linear and semiparametric effects which retains the flexibility and computational advantages of INLA. We apply our method to quantify nonlinear associations between mortality and extreme temperatures in India. An R package implementing our methods is publicly available.

Optimising the case-crossover design for use in shared exposure settings

With a case-crossover design, a case's exposure during a risk period is compared to the case's exposures at referent periods. The selection of referents for this self-controlled design is determined by the referent selection strategy (RSS). Previous research mainly focused on systematic bias associated with the RSS. We additionally focused on how RSS determines the number of referents per risk, sensitivity to overdispersion and time-varying confounding.

We illustrated the consequences of different RSS using a simulation study informed by data on meteorological variables and Legionnaires’ disease. By randomising the events and exposure time series, we explored statistical power associated with time-stratified and fixed bidirectional RSS and their susceptibility to systematic bias and confounding bias. In addition, we investigated how a high number of events on the same date (e.g. outbreaks) affected coefficient estimation. As illustrated by our work, referent selection alone can be insufficient to control for a time-varying confounding bias. In contrast to systematic bias, confounding bias can be hard to detect. We studied potential solutions: varying the model parameters and link-function, outlier-removal and aggregating the input-data over smaller areas. Our simulation study offers a framework for researchers looking to detect and to avoid bias in case-crossover studies.

Gene-by-gene interactions associated with the risk of conotruncal heart defects

BACKGROUND

The development of conotruncal heart defects (CTDs) involves a complex relationship among genetic variants and maternal lifestyle factors. In this article, we focused on the interactions between 13 candidate genes within folate, homocysteine, and transsulfuration pathways for potential association with CTD risk.

METHODS

Targeted sequencing was used for 328 case-parental triads enrolled in the National Birth Defects Prevention Study (NBDPS). To evaluate the interaction of two genes, we applied a conditional logistic regression model for all possible SNP pairs within two respective genes by contrasting the affected infants with their pseudo-controls. The findings were replicated in an independent sample of 86 NBDPS case-parental triads genotyped by DNA microarrays. The results of two studies were further integrated by a fixed-effect meta-analysis.

RESULTS

One SNP pair (i.e., rs4764267 and rs6556883) located in gene MGST1 and GLRX, respectively, was found to be associated with CTD risk after multiple testing adjustment using simpleM, a modified Bonferroni correction approach (nominal p-value of 4.62e-06; adjusted p-value of .04). Another SNP pair (i.e., rs11892646 and rs56219526) located in gene DNMT3A and MTRR, respectively, achieved marginal significance after multiple testing adjustment (adjusted p-value of .06).

CONCLUSION

Further studies with larger sample sizes are needed to confirm and elucidate these potential interactions.

Doubly robust conditional logistic regression

Epidemiologic research often aims to estimate the association between a binary exposure and a binary outcome, while adjusting for a set of covariates (eg, confounders). When data are clustered, as in, for instance, matched case-control studies and co-twin-control studies, it is common to use conditional logistic regression. In this model, all cluster-constant covariates are absorbed into a cluster-specific intercept, whereas cluster-varying covariates are adjusted for by explicitly adding these as explanatory variables to the model. In this paper, we propose a doubly robust estimator of the exposure-outcome odds ratio in conditional logistic regression models. This estimator protects against bias in the odds ratio estimator due to misspecification of the part of the model that contains the cluster-varying covariates. The doubly robust estimator uses two conditional logistic regression models for the odds ratio, one prospective and one retrospective, and is consistent for the exposure-outcome odds ratio if at least one of these models is correctly specified, not necessarily both. We demonstrate the properties of the proposed method by simulations and by re-analyzing a publicly available dataset from a matched case-control study on induced abortion and infertility.

What factors affect the quality of medical students' doctoral theses? A comparative study in affiliated hospitals of a Chinese university

Background: The quality of doctoral theses decides whether medical PhD and MD candidates could get their doctoral degree successfully. Good quality theses could be rewarded by Peking University which is a great honor for both doctoral candidates and for mentors. The present study aims to determine factors affecting the quality of medical doctoral theses. Methods: Honored theses and nonhonored theses were matched 1:3 randomly by specialty and submission year. Conditional logistic regressions were utilized. Results: Five domains comprising 17 indicators were put forward to evaluate the quality of doctoral theses. 41 honored theses and 119 matched nonhonored theses from years 2012-2016 were analyzed by univariate and multivariate conditional logistic regression. Degree type (OR: 107.56, 95%CI: 1.20-9632.70, P=0.041), first author impact factor (OR:1.24, 95%CI: 1.01-1.53, P=0.040) and correctly reported statistic results (OR: 43.18, 95%CI: 1.88-991.61, P=0.019) are independent factors influencing the quality of a doctoral thesis. Conclusions: The present study indicates that there is a significant gap between PhD and MD students on quality of thesis. The rewarded theses have a feature of high first author impact factor. However, most medical students need more training on statistics to improve the quality of their doctoral theses.

Ozone pollution and asthma emergency department visits in the Central Valley, California, USA, during June to September of 2015: a time-stratified case-crossover analysis

Objective: The San Joaquin Valley (SJV) exceeds the state and national standards for ozone (O₃). This study investigates whether short-term exposure to O₃ is associated with asthma emergency department (ED) visits. Methods: We identified 1,101 ED visits in June-September of 2015 in SJV, California, who lived within 15 km of active air monitors. Conditional logistic regression models were used to obtain the odds ratio (OR) and 95% confidence interval (CI) associated with an interquartile (IQR) increase in ozone. We explored the potential effect modification by sex (female and male), race (White, Black and Hispanic), age (2-5, 6-18, 19-40, 41-64 and > = 65) and county (Merced, Madera, Kings, Fresno and Kern). Results: An IQR range (18.1 ppb) increase in O₃ exposure three days before an asthma attack (lag 3) was associated with a 6.6% [OR: 1.066 (95% CI: 1.032, 1.082)] increase in the odds of having an asthma ED visit. The overall ORs differed across age groups and races/ethnicities, with strongest for children aged 6-18 years [OR: 1.219 (95% CI: 1.159, 1.280)], adults 19-40 years [OR: 1.102 (95% CI: 1.053, 1.154)] and Blacks [OR: 1.159 (95% CI: 1.088, 1.236)], respectively. O₃ exposure was not positively associated with asthma ED visits for Whites, while it was for other underrepresented groups. Fresno had the highest number of asthma ED visits and positive association among all five counties. Conclusion: We found that O₃ exposure is associated with asthma ED visits in the SJV.

The role of conservative versus innovative nesting behavior on the 25-year population expansion of an avian predator

Species ranges often change in relation to multiple environmental and demographic factors. Innovative behaviors may affect these changes by facilitating the use of novel habitats, although this idea has been little explored. Here, we investigate the importance of behavior during range change, using a 25-year population expansion of Bonelli's eagle in southern Portugal. This unique population is almost exclusively tree nesting, while all other populations in western Europe are predominantly cliff nesting. During 1991-2014, we surveyed nest sites and estimated the year when each breeding territory was established. We approximated the boundaries of 84 territories using Dirichlet tessellation and mapped topography, land cover, and the density of human infrastructures in buffers (250, 500, and 1,000 m) around nest and random sites. We then compared environmental conditions at matching nest and random sites within territories using conditional logistic regression, and used quantile regression to estimate trends in nesting habitats in relation to the year of territory establishment. Most nests (>85%, n = 197) were in eucalypts, maritime pines, and cork oaks. Nest sites were farther from the nests of neighboring territories than random points, and they were in areas with higher terrain roughness, lower cover by agricultural and built-up areas, and lower road and powerline densities. Nesting habitat selection varied little with year of territory establishment, although nesting in eucalypts increased, while cliff nesting and cork oak nesting, and terrain roughness declined. Our results suggest that the observed expansion of Bonelli's eagles was facilitated by the tree nesting behavior, which allowed the colonization of areas without cliffs. However, all but a very few breeding pairs settled in habitats comparable to those of the initial population nucleus, suggesting that after an initial trigger possibly facilitated by tree nesting, the habitat selection remained largely conservative. Overall, our study supports recent calls to incorporate information on behavior for understanding and predicting species range shifts.

Bayesian Variable Selection Methods for Matched Case-Control Studies

Matched case-control designs are currently used in many biomedical applications. To ensure high efficiency and statistical power in identifying features that best discriminate cases from controls, it is important to account for the use of matched designs. However, in the setting of high dimensional data, few variable selection methods account for matching. Bayesian approaches to variable selection have several advantages, including the fact that such approaches visit a wider range of model subsets. In this paper, we propose a variable selection method to account for case-control matching in a Bayesian context and apply it using simulation studies, a matched brain imaging study conducted at Massachusetts General Hospital, and a matched cardiovascular biomarker study conducted by the High Risk Plaque Initiative.

Case-crossover design and its implementation in R

Case-crossover design is a variation of case-control design that it employs persons' history periods as controls. Case-crossover design can be viewed as the hybrid of case-control study and crossover design. Characteristic confounding that is constant within one person can be well controlled with this method. The relative risk and odds ratio, as well as their 95% confidence intervals (CIs), can be estimated using Cochran-Mantel-Haenszel method. R codes for the calculation are provided in the main text. Readers may adapt these codes to their own task. Conditional logistic regression model is another way to estimate odds ratio of the exposure. Furthermore, it allows for incorporation of other time-varying covariates that are not constant within subjects. The model fitting per se is not technically difficult because there is well developed statistical package. However, it is challenging to convert original dataset obtained from case report form to that suitable to be passed to clogit() function. R code for this task is provided and explained in the text.

The influence of caregivers and behavioral and psychological symptoms on nursing home placement of persons with Alzheimer's disease: A matched case-control study

OBJECTIVES

Behavioral and psychological symptoms of dementia in individuals with Alzheimer's disease and caregiver characteristics may influence the decision to provide care at home or in a nursing home, though few studies examine this association near the actual time of nursing home placement. Using a matched case-control design, this study investigates the association between (1) total Neuropsychiatric Inventory score, (2) the Neuropsychiatric Inventory-4 (an agitation/aggression subscale), and (3) individual domains of the Neuropsychiatric Inventory and nursing home placement.

METHODS

Data from the South Carolina Alzheimer's disease Registry provides an opportunity to expand the literature by looking at cases at the time of nursing home care eligibility/placement and allowing for propensity-score-matched controls. Cases (n = 352) entered a nursing home within 6 months of study initiation; controls (n = 289) remained in the community. Registry data were combined with caregiver survey data, including the Neuropsychiatric Inventory. Conditional logistic regression was applied.

RESULTS

A 10% increase in the Neuropsychiatric Inventory score implied a 30% increase in odds of nursing home admission (odds ratio: 1.30; 95% confidence interval: 1.14-1.50), having married or male caregivers predicted nursing home placement. Cases versus controls were significantly more likely to have behavioral and psychological symptoms of dementia related to agitation/aggression 1 month prior to nursing home admission.

CONCLUSION

Interventions targeting behavioral and psychological symptoms of dementia without available effective interventions in individuals with Alzheimer's disease and caregiver support services are necessary to prevent or delay nursing home admission.

A Conditional Logistic Regression Modeling Approach

INTRODUCTION

This article is part of the Focus Theme of Methods of Information in Medicine on "Big Data and Analytics in Healthcare".

BACKGROUND

Hospital readmissions raise healthcare costs and cause significant distress to providers and patients. It is, therefore, of great interest to healthcare organizations to predict what patients are at risk to be readmitted to their hospitals. However, current logistic regression based risk prediction models have limited prediction power when applied to hospital administrative data. Meanwhile, although decision trees and random forests have been applied, they tend to be too complex to understand among the hospital practitioners.

OBJECTIVES

Explore the use of conditional logistic regression to increase the prediction accuracy.

METHODS

We analyzed an HCUP statewide inpatient discharge record dataset, which includes patient demographics, clinical and care utilization data from California. We extracted records of heart failure Medicare beneficiaries who had inpatient experience during an 11-month period. We corrected the data imbalance issue with under-sampling. In our study, we first applied standard logistic regression and decision tree to obtain influential variables and derive practically meaning decision rules. We then stratified the original data set accordingly and applied logistic regression on each data stratum. We further explored the effect of interacting variables in the logistic regression modeling. We conducted cross validation to assess the overall prediction performance of conditional logistic regression (CLR) and compared it with standard classification models.

RESULTS

The developed CLR models outperformed several standard classification models (e.g., straightforward logistic regression, stepwise logistic regression, random forest, support vector machine). For example, the best CLR model improved the classification accuracy by nearly 20% over the straightforward logistic regression model. Furthermore, the developed CLR models tend to achieve better sensitivity of more than 10% over the standard classification models, which can be translated to correct labeling of additional 400 - 500 readmissions for heart failure patients in the state of California over a year. Lastly, several key predictor identified from the HCUP data include the disposition location from discharge, the number of chronic conditions, and the number of acute procedures.

CONCLUSIONS

It would be beneficial to apply simple decision rules obtained from the decision tree in an ad-hoc manner to guide the cohort stratification. It could be potentially beneficial to explore the effect of pairwise interactions between influential predictors when building the logistic regression models for different data strata. Judicious use of the ad-hoc CLR models developed offers insights into future development of prediction models for hospital readmissions, which can lead to better intuition in identifying high-risk patients and developing effective post-discharge care strategies. Lastly, this paper is expected to raise the awareness of collecting data on additional markers and developing necessary database infrastructure for larger-scale exploratory studies on readmission risk prediction.

Computationally simple analysis of matched, outcome-based studies of ordinal disease states

Outcome-based sampling is an efficient study design for rare conditions, such as glioblastoma. It is often used in conjunction with matching, for increased efficiency and to potentially avoid bias due to confounding. A study was conducted at the Massachusetts General Hospital that involved retrospective sampling of glioblastoma patients with respect to multiple-ordered disease states, as defined by three categories of overall survival time. To analyze such studies, we posit an adjacent categories logit model and exploit its allowance for prospective analysis of a retrospectively sampled study and its advantageous removal of set and level specific nuisance parameters through conditioning on sufficient statistics. This framework allows for any sampling design and is not limited to one level of disease within each set, such as in previous publications. We describe how this ordinal conditional model can be fit using standard conditional logistic regression procedures. We consider an alternative pseudo-likelihood approach that potentially offers robustness under partial model misspecification at the expense of slight loss of efficiency under correct model specification for small sample sizes. We apply our methods to the Massachusetts General Hospital glioblastoma study.

Conditional entropy in variation-adjusted windows detects selection signatures associated with expression quantitative trait loci (eQTLs)

BACKGROUND

Over the past 50,000 years, shifts in human-environmental or human-human interactions shaped genetic differences within and among human populations, including variants under positive selection. Shaped by environmental factors, such variants influence the genetics of modern health, disease, and treatment outcome. Because evolutionary processes tend to act on gene regulation, we test whether regulatory variants are under positive selection. We introduce a new approach to enhance detection of genetic markers undergoing positive selection, using conditional entropy to capture recent local selection signals.

RESULTS

We use conditional logistic regression to compare our Adjusted Haplotype Conditional Entropy (H|H) measure of positive selection to existing positive selection measures. H|H and existing measures were applied to published regulatory variants acting in cis (cis-eQTLs), with conditional logistic regression testing whether regulatory variants undergo stronger positive selection than the surrounding gene. These cis-eQTLs were drawn from six independent studies of genotype and RNA expression. The conditional logistic regression shows that, overall, H|H is substantially more powerful than existing positive-selection methods in identifying cis-eQTLs against other Single Nucleotide Polymorphisms (SNPs) in the same genes. When broken down by Gene Ontology, H|H predictions are particularly strong in some biological process categories, where regulatory variants are under strong positive selection compared to the bulk of the gene, distinct from those GO categories under overall positive selection. . However, cis-eQTLs in a second group of genes lack positive selection signatures detectable by H|H, consistent with ancient short haplotypes compared to the surrounding gene (for example, in innate immunity GO:0042742); under such other modes of selection, H|H would not be expected to be a strong predictor.. These conditional logistic regression models are adjusted for Minor allele frequency(MAF); otherwise, ascertainment bias is a huge factor in all eQTL data sets. Relationships between Gene Ontology categories, positive selection and eQTL specificity were replicated with H|H in a single larger data set. Our measure, Adjusted Haplotype Conditional Entropy (H|H), was essential in generating all of the results above because it: 1) is a stronger overall predictor for eQTLs than comparable existing approaches, and 2) shows low sequential auto-correlation, overcoming problems with convergence of these conditional regression statistical models.

CONCLUSIONS

Our new method, H|H, provides a consistently more robust signal associated with cis-eQTLs compared to existing methods. We interpret this to indicate that some cis-eQTLs are under positive selection compared to their surrounding genes. Conditional entropy indicative of a selective sweep is an especially strong predictor of eQTLs for genes in several biological processes of medical interest. Where conditional entropy is a weak or negative predictor of eQTLs, such as innate immune genes, this would be consistent with balancing selection acting on such eQTLs over long time periods. Different measures of selection may be needed for variant prioritization under other modes of evolutionary selection.

Venous thromboembolism among HIV-positive patients and anticoagulation clinic outcomes integrated within the HIV primary care setting

The purpose of this study was to explore factors associated with venous thromboembolism (VTE) among a cohort of HIV-infected patients and to describe early outcomes of warfarin anticoagulation therapy treated in a pharmacist-based anticoagulation clinic (ACC). A nested case-control study was conducted using the University of Alabama at Birmingham 1917 HIV Clinic Cohort. Conditional logistic regression was used to estimate factors associated with VTE. Among HIV-infected VTE cases, ACC-managed patients were compared to primary care provider (PCP)-managed patients to determine Time within Therapeutic INR Range (TTR). CD4 < 200 cells/µl (OR = 4.50; 95% CI = 1.52, 13.37; p = 0.007) and prior surgical procedures (13.20; 1.56; 111.4; p = 0.018) demonstrated positive associations with VTE, whereas longer HIV duration demonstrated a negative association (0.87; 0.78, 0.98; p = 0.019). TTR was 56.2% among ACC-managed patients compared to 30.5% of PCP-managed patients (p = 0.174). Overall, prior surgical procedures and low CD4 count were associated with an increased risk of VTE among HIV-infected patients. Despite small sample size, patients managed in ACC tend to achieve greater proportion of TTR compared to those managed by PCPs, suggesting that this model of therapy may provide additional benefits to HIV-infected patients.

Effect of driver, roadway, collision, and vehicle characteristics on crash severity: a conditional logistic regression approach

The aim of the study is to examine the factors that appear to have a higher potential for serious injury or death of drivers in traffic accidents in Turkey, such as collision type, roadway surface, vehicle speed, alcohol/drug use, and restraint use. Driver crash severity is the dependent variable of this study with two categories, fatal and non-fatal. Due to the binary nature of the dependent variable, a conditional logistic regression analysis was found suitable. Of the 16 independent variables obtained from Turkish police accident reports, 11 variables were found most significantly associated with driver crash severity. They are age, education level, restraint use, roadway condition, roadway type, time of day, collision location, collision type, number and direction of vehicles, vehicle speed, and alcohol/drug use. This study found that belted drivers aged 18-25 years involving two vehicles travelling in the same direction, in an urban area, during the daytime, and on an avenue or a street have better chances of survival in traffic accidents.

Regularization Paths for Conditional Logistic Regression: The clogitL1 Package

We apply the cyclic coordinate descent algorithm of Friedman, Hastie, and Tibshirani (2010) to the fitting of a conditional logistic regression model with lasso [Formula: see text] and elastic net penalties. The sequential strong rules of Tibshirani, Bien, Hastie, Friedman, Taylor, Simon, and Tibshirani (2012) are also used in the algorithm and it is shown that these offer a considerable speed up over the standard coordinate descent algorithm with warm starts. Once implemented, the algorithm is used in simulation studies to compare the variable selection and prediction performance of the conditional logistic regression model against that of its unconditional (standard) counterpart. We find that the conditional model performs admirably on datasets drawn from a suitable conditional distribution, outperforming its unconditional counterpart at variable selection. The conditional model is also fit to a small real world dataset, demonstrating how we obtain regularization paths for the parameters of the model and how we apply cross validation for this method where natural unconditional prediction rules are hard to come by.

Methods of estimating or accounting for neighborhood associations with health using complex survey data

Reasons for health disparities may include neighborhood-level factors, such as availability of health services, social norms, and environmental determinants, as well as individual-level factors. Investigating health inequalities using nationally or locally representative data often requires an approach that can accommodate a complex sampling design, in which individuals have unequal probabilities of selection into the study. The goal of the present article is to review and compare methods of estimating or accounting for neighborhood influences with complex survey data. We considered 3 types of methods, each generalized for use with complex survey data: ordinary regression, conditional likelihood regression, and generalized linear mixed-model regression. The relative strengths and weaknesses of each method differ from one study to another; we provide an overview of the advantages and disadvantages of each method theoretically, in terms of the nature of the estimable associations and the plausibility of the assumptions required for validity, and also practically, via a simulation study and 2 epidemiologic data analyses. The first analysis addresses determinants of repeat mammography screening use using data from the 2005 National Health Interview Survey. The second analysis addresses disparities in preventive oral health care using data from the 2008 Florida Behavioral Risk Factor Surveillance System Survey.

Determinants of pelvic organ prolapse among the women of the western part of Nepal: a case-control study

AIM

This study was carried out in order to identify the determinants of pelvic organ prolapse (POP) among the women of the western part of Nepal.

METHODS

This was a matched case-control study. Cases were defined as women diagnosed with third and fourth degree POP, and the control group consisted of women who were screened and confirmed of not having any degree of POP. Cases (n = 183) and controls (n = 183) were randomly selected and for each case, one control was selected, matched by age and place of residence. A semistructured interview was carried out to obtain the information for both groups. Bivariate analysis along with conditional logistic regression analysis was carried out to identify the association between selected variables with POP.

RESULTS

The results showed the significant association between number of vaginal deliveries, tear of vagina in the last childbirth, sphincter damage in the last childbirth and duration of labor in the last childbirth with POP, after adjusting for educational status of the women, carrying heavy loads, type of usual work and incidence of diarrhea.

CONCLUSION

The results suggest that prolonged labor along with vaginal delivery having sphincter and vaginal tear are the determinants of POP and program managers should strongly consider these factors to develop interventions targeting the prevention of POP.

Causal inference in paired two-arm experimental studies under noncompliance with application to prognosis of myocardial infarction

Motivated by a study about prompt coronary angiography in myocardial infarction, we propose a method to estimate the causal effect of a treatment in two-arm experimental studies with possible noncompliance in both treatment and control arms. We base the method on a causal model for repeated binary outcomes (before and after the treatment), which includes individual covariates and latent variables for the unobserved heterogeneity between subjects. Moreover, given the type of noncompliance, the model assumes the existence of three subpopulations of subjects: compliers, never-takers, and always-takers. We estimate the model using a two-step estimator: at the first step, we estimate the probability that a subject belongs to one of the three subpopulations on the basis of the available covariates; at the second step, we estimate the causal effects through a conditional logistic method, the implementation of which depends on the results from the first step. The estimator is approximately consistent and, under certain circumstances, exactly consistent. We provide evidence that the bias is negligible in relevant situations. We compute standard errors on the basis of a sandwich formula. The application shows that prompt coronary angiography in patients with myocardial infarction may significantly decrease the risk of other events within the next 2 years, with a log-odds of about  - 2. Given that noncompliance is significant for patients being given the treatment because of high-risk conditions, classical estimators fail to detect, or at least underestimate, this effect.

Glucose-6-phosphate dehydrogenase deficiency, chlorproguanil-dapsone with artesunate and post-treatment haemolysis in African children treated for uncomplicated malaria

BACKGROUND

Malaria is a leading cause of mortality, particularly in sub-Saharan African children. Prompt and efficacious treatment is important as patients may progress within a few hours to severe and possibly fatal disease. Chlorproguanil-dapsone-artesunate (CDA) was a promising artemisinin-based combination therapy (ACT), but its development was prematurely stopped because of safety concerns secondary to its associated risk of haemolytic anaemia in glucose-6-phosphate dehydrogenase (G6PD)-deficient individuals. The objective of the study was to assess whether CDA treatment and G6PD deficiency are risk factors for a post-treatment haemoglobin drop in African children<5 years of age with uncomplicated malaria.

METHODS

This case-control study was performed in the context of a larger multicentre randomized clinical trial comparing safety and efficacy of four different ACT in children with uncomplicated malaria. Children, who after treatment experienced a haemoglobin drop≥2 g/dl (cases) within the first four days (days 0, 1, 2, and 3), were compared with those without an Hb drop (controls). Cases and controls were matched for study site, sex, age and baseline haemoglobin measurements. Data were analysed using a conditional logistic regression model.

RESULTS

G6PD deficiency prevalence, homo- or hemizygous, was 8.5% (10/117) in cases and 6.8% (16/234) in controls (p=0.56). The risk of a Hb drop≥2 g/dl was not associated with either G6PD deficiency (adjusted odds ratio (AOR): 0.81; p=0.76) or CDA treatment (AOR: 1.28; p=0.37) alone. However, patients having both risk factors tended to have higher odds (AOR: 11.13; p=0.25) of experiencing a Hb drop≥2 g/dl within the first four days after treatment, however this finding was not statistically significant, mainly because G6PD deficient patients treated with CDA were very few. In non-G6PD deficient individuals, the proportion of cases was similar between treatment groups while in G6PD-deficient individuals, haemolytic anaemia occurred more frequently in children treated with CDA (56%) than in those treated with other ACT (29%), though the difference was not significant (p=0.49).

CONCLUSION

The use of CDA for treating uncomplicated malaria may increase the risk of haemolytic anaemia in G6PD-deficient children.

Hospital admissions for asthma and acute bronchitis in El Paso, Texas: do age, sex, and insurance status modify the effects of dust and low wind events?

BACKGROUND

El Paso County (Texas) is prone to still air inversions and is one of the dust "hot spots" in North America. In this context, we examined the sub-lethal effects of airborne dust and low wind events on human respiratory health (i.e., asthma and acute bronchitis) between 2000 and 2003, when 110 dust and 157 low wind events occurred. Because environmental conditions may not affect everyone the same, we explored the effects of dust and low wind within three age groups (children, adults, and the elderly), testing for effect modifications by sex and insurance status, while controlling for weather and air pollutants.

METHODS

We used a case-crossover design using events matched with referent days on the same day-of-the-week, month, and year with conditional logistic regression to estimate the probability of hospital admission, while controlling for apparent temperature (lag 1), nitrogen dioxide, and particulate matter of 2.5μm or less.

RESULTS

Children (aged 1-17) were 1.19 (95% confidence interval: 1.00-1.41) times more likely to be hospitalized for asthma three days after a low wind event, and 1.33 (95% CI: 1.01-1.75) times more likely to be hospitalized for acute bronchitis one day after a dust event than on a clear day. Girls were more sensitive to acute bronchitis hospitalizations after dust events (1.83, 95% CI: 1.09-3.08) than boys, but less sensitive than boys to acute bronchitis hospitalizations after low wind events (0.68, 95% CI: 0.46-1.00). We found general trends with regard to dust and low wind events being associated with increased odds of hospitalization for asthma and bronchitis amongst all ages and adults (aged 18-64). Adults covered by Medicaid and adults without health insurance had higher risks of hospitalization for asthma and acute bronchitis after both low wind and dust events.

CONCLUSIONS

Results suggest that there were respiratory health effects associated with dust and low wind events in El Paso, with stronger impacts among children and poor adults. Girls and boys with acute bronchitis were differentially sensitive to dust and low wind events.

Fatal injuries in the slums of Nairobi and their risk factors: results from a matched case-control study

Injuries contribute significantly to the rising morbidity and mortality attributable to non-communicable diseases in the developing world. Unfortunately, active injury surveillance is lacking in many developing countries, including Kenya. This study aims to describe and identify causes of and risk factors for fatal injuries in two slums in Nairobi city using a demographic surveillance system framework. The causes of death are determined using verbal autopsies. We used a nested case-control study design with all deaths from injuries between 2003 and 2005 as cases. Two controls were randomly selected from the non-injury deaths over the same period and individually matched to each case on age and sex. We used conditional logistic regression modeling to identity individual- and community-level factors associated with fatal injuries. Intentional injuries accounted for about 51% and unintentional injuries accounted for 49% of all injuries. Homicides accounted for 91% of intentional injuries and 47% of all injury-related deaths. Firearms (23%) and road traffic crashes (22%) were the leading single causes of deaths due to injuries. About 15% of injuries were due to substance intoxication, particularly alcohol, which in this community comes from illicit brews and is at times contaminated with methanol. Results suggest that in the pervasively unsafe and insecure environment that characterizes the urban slums, ethnicity, residence, and area level factors contribute significantly to the risk of injury-related mortality.

An Evaluation of Multiple Behavioral Risk Factors for Cancer in a Working Class, Multi-Ethnic Population

Behavioral risk factors for cancer tend to cluster within individuals, which can compound risk beyond that associated with the individual risk factors alone. There has been increasing attention paid to the prevalence of multiple risk factors (MRF) for cancer, and to the importance of designing interventions that help individuals reduce their risks across multiple behaviors simultaneously. The purpose of this paper is to develop methodology to identify an optimal linear combination of multiple risk factors (score function) which would facilitate evaluation of cancer interventions.