Traditional Case-control Study Research Articles

Challenges and approaches to calibrating patient phenotype as evidence for cancer gene variant classification under ACMG/AMP guidelines.

Since first publication of the American College of Medical Genetics and Genomics/Association for Medical Pathology (ACMG/AMP) variant classification guidelines, additional recommendations for application of certain criteria have been released (https://clinicalgenome.org/docs/), to improve their application in the diagnostic setting. However, none have addressed use of the PS4 and PP4 criteria, capturing patient presentation as evidence towards pathogenicity. Application of PS4 can be done through traditional case-control studies, or "proband counting" within or across clinical testing cohorts. Review of the existing PS4 and PP4 specifications for Hereditary Cancer Gene Variant Curation Expert Panels revealed substantial differences in the approach to defining specifications. Using BRCA1, BRCA2 and TP53 as exemplar genes, we calibrated different methods proposed for applying the "PS4 proband counting" criterion. For each approach, we considered limitations, non-independence with other ACMG/AMP criteria, broader applicability, and variability in results for different datasets. Our findings highlight inherent overlap of proband-counting methods with ACMG/AMP frequency codes, and the importance of calibration to derive dataset-specific code weights that can account for potential between-dataset differences in ascertainment and other factors. Our work emphasizes the advantages and generalizability of logistic regression analysis over simple proband-counting approaches to empirically determine the relative predictive capacity and weight of various personal clinical features in the context of multigene panel testing, for improved variant interpretation. We also provide a general protocol, including instructions for data formatting and a web-server for analysis of personal history parameters, to facilitate dataset-specific calibration analyses required to use such data for germline variant classification.

Evaluating risk of bias using ROBINS-I tool in nonrandomized studies of adjuvanted influenza vaccine

Seasonal variation in influenza vaccine effectiveness (VE) makes real-world evidence (RWE) useful in supplementing the clinical-evidence base from randomized clinical trials. Adjuvanted inactivated influenza vaccine (aIIV) VE has been evaluated in multiple nonrandomized RWE studies. A systematic literature review of RWE studies evaluating the absolute or relative VE of aIIV was conducted. Identified studies were assessed by evaluators for risk of bias (RoB) by means of the ROBINS-I (Reduction of Bias In Non-randomized Studies of Interventions) tool to inform evidence-based medicine deliberations. Differences in evaluator assessments were resolved by consensus. The literature review yielded 14 follow-up studies, seven test-negative case-control (TNCC) studies, five traditional case-control studies, and one cluster-randomized clinical trial. Most follow-up studies and three TNCC studies were judged at low RoB. Issues increasing RoB included inadequate control of confounding, selection of controls, and reliance on recall of vaccination. The concerns identified in any of the designs could be mitigated with straightforward revisions to design or implementation. 17 of 27 nonrandomized studies of adjuvanted influenza-vaccine effectiveness, some from each of four study designs, were judged at low risk of material bias. These studies merit credence in assessing aIIV effectiveness relative to other influenza vaccines.

Resolving heterogeneity in obsessive-compulsive disorder through individualized differential structural covariance network analysis.

The high heterogeneity of obsessive-compulsive disorder (OCD) denies attempts of traditional case-control studies to derive neuroimaging biomarkers indicative of precision diagnosis and treatment. To handle the heterogeneity, we uncovered subject-level altered structural covariance by adopting individualized differential structural covariance network (IDSCN) analysis. The IDSCN measures how structural covariance edges in a patient deviated from those in matched healthy controls (HCs) yielding subject-level differential edges. One hundred patients with OCD and 106 HCs were recruited and whose T1-weighted anatomical images were acquired. We obtained individualized differential edges and then clustered patients into subtypes based on these edges. Patients presented tremendously low overlapped altered edges while frequently shared altered edges within subcortical-cerebellum network. Two robust neuroanatomical subtypes were identified. Subtype 1 presented distributed altered edges while subtype 2 presented decreased edges between default mode network and motor network compared with HCs. Altered edges in subtype 1 predicted the total Yale-Brown Obsessive Compulsive Scale score while that in subtype 2 could not. We depict individualized structural covariance aberrance and identify that altered connections within subcortical-cerebellum network are shared by most patients with OCD. These 2 subtypes provide new insights into taxonomy and facilitate potential clues to precision diagnosis and treatment of OCD.

Case-Control Studies in Neurosurgery: The Issue of Effect Estimates

Clinical research questions are commonly answered using a case-control design. The decision to use this design is usually justified due to low cost, feasibility, and ease of execution. However, the case-control design presents challenges in execution, selection of cases/controls, and interpretation of effect measures (odds ratios, among others). In this paper, we clarify for a neurosurgical audience the design and appropriate effect size measures obtained from case-control studies. A narrative review was conducted of published literature on the topic. The future implementation of such studies was discussed and highlighted with several examples from neurosurgical practice. In a case-control design, participants are selected for a study based on their outcome status. Some participants have the outcome of interest (cases), whereas others do not (controls). Controls can be selected from a variety of sources, such as the general population, relatives/friends, or hospital patients without the disease under investigation. The most important criterion is that these controls come from the same study base as cases. Furthermore, it is essential to realize that measures of association obtained from a case-control study depend on the sampling strategy of the controls and, as such, have equivalent counterparts available from cohort studies. We delineate traditional case-control, case-cohort, and incidence density-sampled case-control studies and their applicability to common conditions encountered in daily neurosurgical practice (e.g., glioblastoma, aneurysms, and epilepsy). Neurosurgeons must understand the types of case-control studies and their associated effect measures to properly conduct research and incorporate research findings into clinical practice.

Current Approaches to Vaccine Safety Using Observational Data: A Rationale for the EUMAEUS (Evaluating Use of Methods for Adverse Events Under Surveillance-for Vaccines) Study Design.

Post-marketing vaccine safety surveillance aims to detect adverse events following immunization in a population. Whether certain methods of surveillance are more precise and unbiased in generating safety signals is unclear. Here, we synthesized information from existing literature to provide an overview of the strengths, weaknesses, and clinical applications of epidemiologic and analytical methods used in vaccine monitoring, focusing on cohort, case-control and self-controlled designs. These designs are proposed to be evaluated in the EUMAEUS (Evaluating Use of Methods for Adverse Event Under Surveillance–for vaccines) study because of their widespread use and potential utility. Over the past decades, there have been an increasing number of epidemiological study designs used for vaccine safety surveillance. While traditional cohort and case-control study designs remain widely used, newer, novel designs such as the self-controlled case series and self-controlled risk intervals have been developed. Each study design comes with its strengths and limitations, and the most appropriate study design will depend on availability of resources, access to records, number and distribution of cases, and availability of population coverage data. Several assumptions have to be made while using the various study designs, and while the goal is to mitigate any biases, violations of these assumptions are often still present to varying degrees. In our review, we discussed some of the potential biases (i.e., selection bias, misclassification bias and confounding bias), and ways to mitigate them. While the types of epidemiological study designs are well established, a comprehensive comparison of the analytical aspects (including method evaluation and performance metrics) of these study designs are relatively less well studied. We summarized the literature, reporting on two simulation studies, which compared the detection time, empirical power, error rate and risk estimate bias across the above-mentioned study designs. While these simulation studies provided insights on the analytic performance of each of the study designs, its applicability to real-world data remains unclear. To bridge that gap, we provided the rationale of the EUMAEUS study, with a brief description of the study design; and how the use of real-world multi-database networks can provide insights into better methods evaluation and vaccine safety surveillance.

Clinical risk factors and predictive score for the non-dipper profile in hypertensive patients: a case-control study

BackgroundNight-time BP, especially non-dipper, is a stronger predictor of adverse cardiovascular outcomes. Ambulatory blood pressure monitoring (ABPM) is a gold standard for the detection of non-dippers but it often is unavailable and expensive. This study aims to determine clinical risk factors that predict non-dipper.MethodsAn exploratory traditional case-control study, exclusive sampling of control was conducted from January 2013 to September 2018 to explore clinical risk factors associated with non-dippers in hypertensive patients. Subgroup analysis was performed in each treated and untreated hypertensive patient. The parsimonious predictive score for non-dippers was constructed.ResultsThe study included 208 hypertensive patients receiving 24 h ABPM. There were 104 dippers and 104 non-dippers. Significant clinical risk factors associated with non-dippers were the age of > 65 years, average office diastolic blood pressure (DBP), and fasting plasma glucose of > 5.6 mmol/L. Results of subgroup analysis showed that dyslipidemia, history of coronary artery disease, use of angiotensin-converting enzyme inhibitors (ACEIs) and direct vasodilators, average office DBP, and serum uric acid were associated with non-dippers in treated hypertensive patients, however, there were no risk factors associated with non-dippers in the untreated group. The predictive score for non-dippers in treated group included average office DBP, dyslipidemia, serum uric acid, male, calcium channel blockers and ACEIs use. The area under Receiver Operating Characteristic (AuROC) was 0.723. A cut-off point which was > 0.0701 and prevalence of non-dippers of 46%, this score had a sensitivity of 77.4%, specificity of 65.6%, positive predictive value (PPV) of 66.1%, and negative predictive value (NPV) of 79.6%. For untreated group, age, hemoglobin and body mass index were included in the predictive model. AuROC was 0.74. There was a sensitivity of 51.9%, specificity of 91.2%, PPV of 82.4%, and NPV of 70.5% at the cut-off point of > 0.357, and prevalence of 44%.ConclusionThere were several significant clinical risk factors associated with non-dippers in treated hypertensive patients. The predictive score might be useful for the detection of non-dippers; however, it cannot replace ABPM.

Theoretical Framework for Retrospective Studies of the Effectiveness of SARS-CoV-2 Vaccines.

Observational studies of the effectiveness of vaccines to prevent COVID-19 are needed to inform real-world use. Such studies are now underway amid the ongoing rollout of SARS-CoV-2 vaccines globally. Although traditional case-control and test-negative design studies feature prominently among strategies used to assess vaccine effectiveness, such studies may encounter important threats to validity. Here, we review the theoretical basis for estimation of vaccine direct effects under traditional case-control and test-negative design frameworks, addressing specific natural history parameters of SARS-CoV-2 infection and COVID-19 relevant to these designs. Bias may be introduced by misclassification of cases and controls, particularly when clinical case criteria include common, nonspecific indicators of COVID-19. When using diagnostic assays with high analytical sensitivity for SARS-CoV-2 detection, individuals testing positive may be counted as cases even if their symptoms are due to other causes. The traditional case-control design may be particularly prone to confounding due to associations of vaccination with healthcare-seeking behavior or risk of infection. The test-negative design reduces but may not eliminate this confounding, for instance, if individuals who receive vaccination seek care or testing for less-severe illness. These circumstances indicate the two study designs cannot be applied naively to datasets gathered through public health surveillance or administrative sources. We suggest practical strategies to reduce bias in vaccine effectiveness estimates at the study design and analysis stages.

The utility of hierarchical models of psychopathology in genetics and biomarker research.

The utility of hierarchical models of psychopathology in genetics and biomarker research.

Sex-specific pathways among tri-allelic serotonin transporter polymorphism, trait neuroticism and generalized anxiety disorder.

Neuroticism personality trait is recognized as an important endophenotypic predictor of generalized anxiety disorder (GAD). Furthermore, endophenotype-based pathway approaches have recently been shown to have greater advantages for gene-finding strategies than traditional case-control studies. In the present study, in addition to conventional case-control methods, we used pathway analyses to test whether the tri-allelic serotonin transporter promoter polymorphism (combining 5-HTTLPR and rs25531) is associated with risk of GAD through its effects on trait neuroticism. We included 2236 Han Chinese adults in this study, including 736 patients with GAD and 1500 healthy participants. We genotyped the 5-HTTLPR and rs25531 polymorphisms using the polymerase chain reaction restriction fragment length polymorphism method. We used the Neuroticism scale of the Maudsley Personality Inventory (MPI) short version (MPI-Neuroticism) to measure participants' tendency toward neuroticism. Using endophenotype-based path analyses, we found significant indirect effects of the tri-allelic genotype on risk of GAD, mediated by MPI-Neuroticism in both men and women. Compared to women carrying the S'S' genotype, women carrying the L' allele had higher levels of MPI-Neuroticism, which in turn were associated with higher risk of GAD. Men, however, showed the opposite pattern. Using traditional case-control comparisons, we observed that the effect of tri-allelic genotype on GAD was significant, but only in women. Participants were restricted to Han Chinese, and we used only 1 questionnaire to assess neuroticism. These findings are the first to show that the triallelic 5-HTTLPR polymorphism is associated with elevated risk of GAD, and that this effect is mediated via increased trait neuroticism, a sex-dependent risk pathway.

Feasibility of case-control and test-negative designs to evaluate dengue vaccine effectiveness in Malaysia

BackgroundThe world’s first dengue vaccine [Dengvaxia; Sanofi Pasteur] was licensed in 2015 and others are in development. Real-world evaluations of dengue vaccines will therefore soon be needed. We assessed feasibility of case control (CC) and test-negative (TN) design studies for dengue vaccine effectiveness by measuring associations between socio-demographic risk factors, and hospitalized dengue outcomes, in Malaysia. MethodsFollowing ethical approval, we conducted hospital-based dengue surveillance for one year in three referral hospitals. Suspected cases aged 9–25 years underwent dengue virological confirmation by RT-PCR and/or NS1 Ag ELISA at a central laboratory. Two age- and geography-matched hospitalized non-dengue case-controls were recruited for a traditional CC study. Suspected cases testing negative were test-negative controls. Socio-demographic, risk factor and routine laboratory data were collected. Logistic regression models were used to estimate associations between confirmed dengue and risk factors. ResultsWe recruited 327 subjects; 155 were suspected of dengue. The planned sample size was not met. 124 (80%) of suspected cases were dengue-confirmed; seven were assessed as severe. Three had missing RT-PCR results; the study recruited 28 test-negative controls. Only 172 matched controls could be recruited; 90 cases were matched with ≥1 controls. Characteristics of cases and controls were mostly similar. By CC design, two variables were significant risk factors for hospitalized dengue: recent household dengue contact (OR: 54, 95% CI: 7.3–397) and recent neighbourhood insecticidal fogging (OR: 2.1; 95% CI: 1.3–3.6). In the TN design, no risk factors were identified. In comparison with gold-standard diagnostics, routine tests performed poorly. ConclusionsThe CC design may be more appropriate than the TN design for hospitalized dengue vaccine effectiveness studies. Selection bias in case control selection could be minimized by protocol changes more easily than increasing TN design control numbers, because early-stage dengue diagnosis in endemic countries is highly specific. MREC study approval: (39)KKM/NIHSEC/P16-1334.

Network analysis of cases with methicillin-resistant Staphylococcus aureus and controls in a large tertiary care facility

Despite increased awareness of infection control precautions, methicillin-resistant Staphylococcus aureus (MRSA) still spreads through patients and contaminated objects, causing a substantial burden of illness and cost. Our objective was to define risk factors for contracting MRSA in a tertiary health care facility using a historic case-control study and to validate health care network changes during pre-outbreak and outbreak periods. We conducted a case-control study using secondary data on hospitalizations where infected or colonized cases were compared with matched controls who tested negative by age, sex, and campus over 1 year. Social networks of all cases and controls were built from links joining patients to rooms, roommates, and health care providers over time. Matched controls were similar to cases in comorbidity, lengths of stay, mortality, and number of roommates, rooms, and health care providers. As expected, the number of rooms and roommates increased in the outbreak by more than 50%. A timed animation of the network at one campus identified potential source patients linked to 2 rooms and many roommates, after which cases connected to those same rooms proliferated. Only the network animation over time revealed possible transmission of MRSA through the network, rather than attributes measured in the traditional case control study.

Concurrent use of benzodiazepines, antidepressants, and opioid analgesics with zolpidem and risk for suicide: a case-control and case-crossover study.

To evaluate whether the concurrent use of benzodiazepines, antidepressants, and opioid analgesics with zolpidem increases the risk of suicide or triggers suicide compared with the use of zolpidem alone. We conducted a case-control and case-crossover study using the Korean National Health Insurance Service-National Sample Cohort database. Cases were older than 20years with a suicide record (International Codes of Disease 10th Revision codes: X-60-X84 and Y87.0 intentional self-harm) between January 1, 2004, and December 31, 2013. For case-control design, ten controls were matched to each case by age, sex, index year, region, income, and health insurance type. For case-crossover analysis, we set hazard period to 60days and assigned five corresponding sets of control periods of equal length. Exposure was assessed during 60days before suicide for combinations of benzodiazepines, antidepressants, opioid analgesics with zolpidem against zolpidem alone. We conducted a conditional logistic regression to estimate odds ratios (ORs) and their 95% confidence intervals (CIs). In the case-control study, the risk of suicide was 2.80-fold higher in cases taking benzodiazepines and antidepressants with zolpidem than in those taking zolpidem alone (adjusted OR [aOR], 2.80; 95% CI, 1.38-5.70). However, in the case-crossover study, suicide risk showed no significant difference (crude OR [cOR], 0.92; 95% CI, 0.55-1.52) and was underpowered. The results of the traditional case-control study confirmed that the concurrent use of benzodiazepines and antidepressants with zolpidem was associated with an increased risk of suicide compared with the use of zolpidem alone. However, there was no significant difference in the magnitude of risk in the within-person comparison design.

A rare missense variant of CASP7 is associated with familial late-onset Alzheimer's disease

IntroductionThe genetic architecture of Alzheimer's disease (AD) is only partially understood. MethodsWe conducted an association study for AD using whole sequence data from 507 genetically enriched AD cases (i.e., cases having close relatives affected by AD) and 4917 cognitively healthy controls of European ancestry (EA) and 172 enriched cases and 179 controls of Caribbean Hispanic ancestry. Confirmation of top findings from stage 1 was sought in two family-based genome-wide association study data sets and in a whole genome–sequencing data set comprising members from 42 EA and 115 Caribbean Hispanic families. ResultsWe identified associations in EAs with variants in 12 novel loci. The most robust finding is a rare CASP7 missense variant (rs116437863; P = 2.44 × 10−10) which improved when combined with results from stage 2 data sets (P = 1.92 × 10−10). DiscussionOur study demonstrated that an enriched case design can strengthen genetic signals, thus allowing detection of associations that would otherwise be missed in a traditional case-control study.

A Dynamic Model for Evaluation of the Bias of Influenza Vaccine Effectiveness Estimates From Observational Studies.

Given that influenza vaccination is now widely recommended in the United States, observational studies based on patients with acute respiratory illness (ARI) remain as the only option to estimate influenza vaccine effectiveness (VE). We developed a dynamic probability model to evaluate bias of VE estimates from passive surveillance cohort, test-negative, and traditional case-control studies. The model includes 2 covariates (health status and health awareness) that might affect the probabilities of vaccination, developing ARI, and seeking medical care. Our results suggest that test-negative studies produce unbiased estimates of VE against medically attended influenza when: 1) Vaccination does not affect the probability of noninfluenza ARI; and 2) health status has the same effect on the probability of influenza and noninfluenza ARIs. The same estimate might be severely biased (i.e., estimated VE - true VE ≥ 0.20) for estimating VE against symptomatic influenza if the vaccine affects the probability of seeking care against influenza ARI. VE estimates from test-negative studies might also be severely biased for both outcomes of interest when vaccination affects the probability of noninfluenza ARI, but estimates from passive surveillance cohort studies are unbiased in this case. Finally, VE estimates from traditional case-control studies suffer from bias regardless of the source of bias.

A comparison of the test-negative and traditional case-control study designs with respect to the bias of estimates of rotavirus vaccine effectiveness

Estimation of the effectiveness of rotavirus vaccines via the test-negative control study design has gained popularity over the past few years. In this study design, children with severe diarrhea who test positive for rotavirus infection are considered as cases, while children who test negative serve as controls. We use a simple probability model to evaluate and compare the test-negative control and the traditional case-control designs with respect to the bias of resulting estimates of rotavirus vaccine effectiveness (VE). Comparisons are performed under two scenarios, corresponding to studies performed in high-income and low-income countries. We consider two potential sources of bias: (a) misclassification bias resulting from imperfect sensitivity and specificity of the test used to diagnose rotavirus infection, and (b) selection bias associated with possible effect of rotavirus vaccination on the probability of contracting severe non-rotavirus diarrhea.Our results suggest that both sources of bias may produce VE estimates with substantial bias. Particularly, lack of perfect specificity is associated with severe negative bias. For example, if the specificity of the diagnostic test is 90% then VE estimates from both types of case-control studies may under-estimate the true VE by more than 20%. If the vaccine protects children against non-rotavirus diarrhea then VE estimates from test-negative control studies may be close to zero even though the true VE is 50%. However, the sensitivity and specificity of the enzyme immunoassay test currently used to diagnose rotavirus infections are both over 99%, and there is no solid evidence that the existing rotavirus vaccines affect the rates of non-rotavirus diarrhea. We therefore conclude that the test-negative control study design is a convenient and reliable alternative for estimation of rotavirus VE.

Subtypes of Benign Breast Disease as a Risk Factor of Breast Cancer: A Systematic Review and Meta Analyses

Researchers suggest that benign breast disease (BBD) is a key risk factor for breast cancer. The present study aimed to determinate the risk level of breast cancer in terms of various BBD subgroups. A meta-analysis was performed to determinate the risk of breast cancer associated with BBD. Observational studies (traditional case-control studies, nested case-control studies, and cohort studies) published from January 2000 to June 2015 were assessed to evaluate the risk of developing breast cancer related to BBD. Various databases such as Medline (PubMed), Web of Science (ISI), Scopus, and Google Scholar were searched. The additional search included the Journal of Breast Cancer Research and Treatment and the Journal of Cancer Research. Twenty studies out of 21 were used to estimate the risk of developing breast cancer related to proliferative disease without atypia versus non-proliferative disease and the reported risk ranged from 1.04 to 1.83. The reported risk of developing breast cancer related to proliferative disease with atypia versus non-proliferative disease in 21 studies ranged from 1.59 to 4.74. Based on 20 studies, the pooled risk estimates for developing breast cancer related to proliferative disease without atypia versus non-proliferative disease was 1.58 (95% CI: 1.51-1.66). Based on 21 studies, the pooled risk estimates for developing breast cancer related to proliferative disease with atypia versus non-proliferative disease was 3.49 (95% CI: 3.23-3.77). The overall result of this review showed an elevated risk for breast cancer related to BBD subtypes. We propose better strategies for screening recommendations for such women.

Effectiveness of inactivated quadrivalent influenza vaccine in the 2015/2016 season as assessed in both a test-negative case-control study design and a traditional case-control study design.

Both studies confirmed significant VE against influenza A, significant two-dose VE against influenza B, and better two-dose VE than one-dose VE. What is Known: • Influenza vaccine effectiveness (VE) varies from year to year. • Observational studies are conventionally used for VE assessment. However, they are inherently susceptible to bias and confounding. What is New: • This is the first report of influenza VE assessment using more than one observational study and performed in a specific area during the same season. • VE estimates obtained in our traditional case-control study were lower than those in our test-negative case-control study, but both studies found significant VE against influenza.

A comparison of the test-negative and the traditional case-control study designs for estimation of influenza vaccine effectiveness under nonrandom vaccination

BackgroundAs annual influenza vaccination is recommended for all U.S. persons aged 6 months or older, it is unethical to conduct randomized clinical trials to estimate influenza vaccine effectiveness (VE). Observational studies are being increasingly used to estimate VE. We developed a probability model for comparing the bias and the precision of VE estimates from two case-control designs: the traditional case-control (TCC) design and the test-negative (TN) design. In both study designs, acute respiratory illness (ARI) patients seeking medical care testing positive for influenza infection are considered cases. In the TN design, ARI patients seeking medical care who test negative serve as controls, while in the TCC design, controls are randomly selected individuals from the community who did not contract an ARI.MethodsOur model assigns each study participant a covariate corresponding to the person’s health status. The probabilities of vaccination and of contracting influenza and non-influenza ARI depend on health status. Hence, our model allows non-random vaccination and confounding. In addition, the probability of seeking care for ARI may depend on vaccination and health status. We consider two outcomes of interest: symptomatic influenza (SI) and medically-attended influenza (MAI).ResultsIf vaccination does not affect the probability of non-influenza ARI, then VE estimates from TN studies usually have smaller bias than estimates from TCC studies. We also found that if vaccinated influenza ARI patients are less likely to seek medical care than unvaccinated patients because the vaccine reduces symptoms’ severity, then estimates of VE from both types of studies may be severely biased when the outcome of interest is SI. The bias is not present when the outcome of interest is MAI.ConclusionsThe TN design produces valid estimates of VE if (a) vaccination does not affect the probabilities of non-influenza ARI and of seeking care against influenza ARI, and (b) the confounding effects resulting from non-random vaccination are similar for influenza and non-influenza ARI. Since the bias of VE estimates depends on the outcome against which the vaccine is supposed to protect, it is important to specify the outcome of interest when evaluating the bias.

Predicting Road Accidents: A Rare-events Modeling Approach

Modeling road accident occurrence has gained increasing attention over the years. So far, considerable efforts have been made from researchers and policy makers in order to explain road accidents and improve road safety performance of highways. In reality, road accidents are rare events. In such cases, the binary dependent variable is characterized by dozens to thousands of times fewer events (accidents) than non-events (non-accidents). Instead of using traditional logistic regression methods, this paper considers accidents as rare events and proposes a series of rare-events logit models which are applied in order to model road accident occurrence by utilizing real-time traffic data. This statistical procedure was initially proposed by King and Zeng (2001) when scholars study rare events such as wars, massive economic crises and so on. Rare-events logit models basically estimate the same models as traditional logistic regression, but the estimates as well as the probabilities are corrected for the bias that occurs when the sample is small or the observed events are very rare. Consequently, the basic problem of underestimating the event probabilities is avoided as stated by King and Zeng (2001). To the best of our knowledge, this is the first time that this approach is followed when road accident data are analyzed. Instead of applying a traditional case-control study, the complete dataset of hourly aggregated traffic data such as flow, occupancy, mean time speed and percentage of trucks, were collected from three random loop detectors in the Attica Tollway (“Attiki Odos”) located in Greater Athens Area in Greece for the 2008–2011 period. The modeling results showed an adequate statistical fit and reveal a negative relationship between accident occurrence and the natural logarithm of speed in the accident location. This study attempts to contribute to the understanding of accident occurrence in motorways by developing novel models such as the rare-events logit for the first time in safety evaluation of motorways.

Allowing for population stratification in case-only studies of gene-environment interaction, using genomic control.

Gene-environment interactions (G × E) have attracted considerable research interest in the past owing to their scientific and public health implications, but powerful statistical methods are required to successfully track down G × E, particularly at a genome-wide level. Previously, a case-only (CO) design has been proposed as a means to identify G × E with greater efficiency than traditional case-control or cohort studies. However, as with genotype-phenotype association studies themselves, hidden population stratification (PS) can impact the validity of G × E studies using a CO design. Since this problem has been subject to little research to date, we used comprehensive simulation to systematically assess the type I error rate, power and effect size bias of CO studies of G × E in the presence of PS. Three types of PS were considered, namely genetic-only (PSG), environment-only (PSE), and joint genetic and environmental stratification (PSGE). Our results reveal that the type I error rate of an unadjusted Wald test, appropriate for the CO design, would be close to its nominal level (0.05 in our study) as long as PS involves only one interaction partner (i.e., either PSG or PSE). In contrast, if the study population is stratified with respect to both G and E (i.e., if there is PSGE), then the type I error rate is seriously inflated and estimates of the underlying G × E interaction are biased. Comparison of CO to a family-based case-parents design confirmed that the latter is more robust against PSGE, as expected. However, case-parent trios may be particularly unsuitable for G × E studies in view of the fact that they require genotype data from parents and that many diseases with an environmental component are likely to be of late onset. An alternative approach to adjusting for PS is principal component analysis (PCA), which has been widely used for this very purpose in past genome-wide association studies (GWAS). However, resolving genetic PS properly by PCA requires genetic data at the population level, the availability of which would conflict with the basic idea of the CO design. Therefore, we explored three modified Wald test statistics, inspired by the genomic control (GC) approach to GWAS, as an alternative means to allow for PSGE. The modified statistics were benchmarked against a stratified Wald test assuming known population affiliation, which should provide maximum power under PS. Our results confirm that GC is capable of successfully and efficiently correcting the PS-induced inflation of the type I error rate in CO studies of G × E.