A haplotype-based test of association using data from cohort and nested case-control epidemiologic studies

Haplotype-based risk models can lead to powerful methods for detecting the association of a disease with a genomic region of interest. In population-based studies of unrelated individuals, however, the haplotype status of some subjects may not be discernible without ambiguity from available locus-specific genotype data. A score test for detecting haplotype-based association using genotype data has been developed in the context of generalized linear models for analysis of data from cross-sectional and retrospective studies. In this article, we develop a test for association using genotype data from cohort and nested case-control studies where subjects are prospectively followed until disease incidence or censoring (end of follow-up) occurs. Assuming a proportional hazard model for the haplotype effects, we derive an induced hazard function of the disease given the genotype data, and hence propose a test statistic based on the associated partial likelihood. The proposed test procedure can account for differential follow-up of subjects, can adjust for possibly time-dependent environmental co-factors and can make efficient use of valuable age-at-onset information that is available on cases. We provide an algorithm for computing the test statistic using readily available statistical software. Utilizing simulated data in the context of two genomic regions GPX1 and GPX3, we evaluate the validity of the proposed test for small sample sizes and study its power in the presence and absence of missing genotype data.     

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.     

Robust trend tests for genetic association in case-control studies using family data

We studied a trend test for genetic association between disease and the number of risk alleles using case-control data. When the data are sampled from families, this trend test can be adjusted to take into account the correlations among family members in complex pedigrees. However, the test depends on the scores based on the underlying genetic model and thus it may have substantial loss of power when the model is misspecified. Since the mode of inheritance will be unknown for complex diseases, we have developed two robust trend tests for case-control studies using family data. These robust tests have relatively good power for a class of possible genetic models. The trend tests and robust trend tests were applied to a dataset of Genetic Analysis Workshop 14 from the Collaborative Study on the Genetics of Alcoholism.     

Hormonal and reproductive risk factors of papillary thyroid cancer: A population-based case-control study in France

The three times higher incidence of thyroid cancer in women compared to men points to a role of female sex hormones in its etiology. However the effects of these factors are poorly understood. We analyzed the association between thyroid cancer and hormonal and reproductive factors among women enrolled in CATHY, a population-based case-control study conducted in France. The study included 430 cases of papillary thyroid cancer and 505 controls frequency-matched on age and area of residence. The odds ratios for thyroid cancer increased with age at menarche (p trend 0.05). Postmenopausal women were at increased risk, as compared to premenopausal women, particularly if menopause followed an ovariectomy, and for women with age at menopause < 55 years. In addition, use of oral contraceptives and menopausal hormone therapy reduced the association with thyroid cancer by about one third, and breastfeeding by 27%. Overall, these findings provide evidence that the risk of thyroid cancer increases with later age at menarche and after menopause, and decreases with use of oral contraceptives and menopausal hormone therapy. These findings confirm an implication of hormonal factors in papillary thyroid cancer risk, whose mechanisms need to be elucidated.     

Comparison of racial differences in childhood cancer risk in case-control studies and population-based cancer registries

Introduction: Although selection bias in case-control studies has been studied extensively, little is known about selection of cases and controls among various ethnic groups. This study compares racial differences in childhood cancer rates as estimated by case-control studies with various design features. It also compares estimates of racial distribution among cases as reported by case-control studies to those observed for an ideal case series with complete ascertainment of cases for these studies or in population-based cancer registries in corresponding geographic regions and calendar periods. Methods: Peer-reviewed publications on childhood leukemia and brain tumors from North America, published between 1980 and 2007, were reviewed. Incidence data by race/ethnicity were compiled from research publications, federal cancer statistics, and cancer registries. Meta-analysis was conducted to assess racial/ethnic differences by study characteristics. Racial distributions of cases from published case-control studies were compared to those of a presumably noncensored case distribution (i.e. include both participating and non-participating cases in a case-control study) or cases recorded by cancer registries. Results: In interview-based case-control studies of childhood cancer, the proportion of Whites compared to non-Whites tended to be higher among controls than among cases; however, the opposite was true for record-based case-control studies. Additionally, the proportion of Whites tended to be higher among the participating cases in the published case-control studies compared to the proportion of Whites among the non-participating cases or in cancer registries. Conclusions: Investigators need to consider differential participation by racial group as a potential source of bias in the interpretation of case-control study results.     

Effect of Pap-smear and sociodemographic factors on cervical cancer risk in Estonia: A population-based case-control study

BackgroundLike many Eastern-European countries, Estonia struggles with ineffective cervical cancer (CC) screening. Despite a long-term organised screening programme and high overall Pap-smear coverage, CC incidence and mortality remain very high. The aim of the study was to examine the reasons for high CC incidence in Estonia by analysing the effect of Pap-smears and sociodemographic factors on CC risk.MethodsIn this population-based case-control study, women aged ≥ 25 years with an in situ/invasive CC diagnosed in Estonia in 2011–2017 were defined as cases. Using a density sampling scheme, controls were randomly selected from general population. To estimate CC risk associated with having no Pap-smears during seven years before diagnosis (cases) or index date (controls), place of residence, interruption in health insurance, and several sociodemographic factors, multivariate logistic regression was used to calculate odds ratios (OR) with 95% confidence intervals (CI). Individual-level data from three population-based registries were used.ResultsAmong 1439 cases and 4317 controls, proportion of women with no Pap-smears was 53% and 35%, respectively. Women with no Pap-smears were at higher risk for CC (OR=2.35; 95% CI: 1.85–2.98). CC risk was increased among women who were younger, living in more remote regions, lower-educated, or divorced/widowed. Interruption in health insurance was associated with a 23% risk increase. Regional differences in CC risk were observed among screened women.ConclusionTo reduce the risk of CC in Estonia, efforts are necessary to increase screening coverage among high-risk women and ensure the quality of CC screening programme. Screening approaches and communication should be tailored to the needs of different population groups. Further studies are warranted to identify the reasons for regional differences in CC risk.     

Design and validation of methods searching for risk factors in genotype case-control studies.

Accessibility of high-throughput genotyping technology allows genome-wide association studies for common complex diseases. This paper addresses two challenges commonly facing such studies: (i) searching an enormous amount of possible gene interactions and (ii) finding reproducible associations. These challenges have been traditionally addressed in statistics while here we apply computational approaches--optimization and cross-validation. A complex risk factor is modeled as a subset of single nucleotide polymorphisms (SNPs) with specified alleles and the optimization formulation asks for the one with the maximum odds ratio. To measure and compare ability of search methods to find reproducible risk factors, we propose to apply a cross-validation scheme usually used for prediction validation. We have applied and cross-validated known search methods with proposed enhancements on real case-control studies for several diseases (Crohn's disease, autoimmune disorder, tick-borne encephalitis, lung cancer, and rheumatoid arthritis). Proposed methods are compared favorably to the exhaustive search: they are faster, find more frequently statistically significant risk factors, and have significantly higher leave-half-out cross-validation rate.     

Inference on haplotype effects in case-control studies using unphased genotype data

A variety of statistical methods exist for detecting haplotype-disease association through use of genetic data from a case-control study. Since such data often consist of unphased genotypes (resulting in haplotype ambiguity), such statistical methods typically apply the expectation-maximization (EM) algorithm for inference. However, the majority of these methods fail to perform inference on the effect of particular haplotypes or haplotype features on disease risk. Since such inference is valuable, we develop a retrospective likelihood for estimating and testing the effects of specific features of single-nucleotide polymorphism (SNP)-based haplotypes on disease risk using unphased genotype data from a case-control study. Our proposed method has a flexible structure that allows, among other choices, modeling of multiplicative, dominant, and recessive effects of specific haplotype features on disease risk. In addition, our method relaxes the requirement of Hardy-Weinberg equilibrium of haplotype frequencies in case subjects, which is typically required of EM-based haplotype methods. Also, our method easily accommodates missing SNP information. Finally, our method allows for asymptotic, permutation-based, or bootstrap inference. We apply our method to case-control SNP genotype data from the Finland-United States Investigation of Non-Insulin-Dependent Diabetes Mellitus (FUSION) Genetics study and identify two haplotypes that appear to be significantly associated with type 2 diabetes. Using the FUSION data, we assess the accuracy of asymptotic P values by comparing them with P values obtained from a permutation procedure. We also assess the accuracy of asymptotic confidence intervals for relative-risk parameters for haplotype effects, by a simulation study based on the FUSION data.     

Risk factors for breast cancer in women biopsied for benign breast disease: A nested case-control study

Aim: Women with a history of benign breast disease are at increased risk of subsequent breast cancer. However, few studies have examined whether established breast cancer risk factors other than histology are associated with an altered risk of breast cancer in women with benign breast disease. We used a nested case-control design within a large, multi-center cohort of women biopsied for benign breast disease (BBD) to estimate odds ratios for breast cancer in association with exposure to a range of personal and lifestyle factors. Methods: Cases were women biopsied for BBD who subsequently developed breast cancer; controls were individually matched to cases on center and age at diagnosis and were women biopsied for BBD who did not develop breast cancer in the same follow-up interval as that for the cases. After excluding women with prevalent breast cancer, 1357 records (661 case records and 696 records) were available for analysis. We used conditional logistic regression to obtain crude and multivariable-adjusted estimates of the association between specific factors and risk of breast cancer. Results: In multivariable analyses age at first live birth, number of pregnancies, and postmenopausal status were inversely associated with risk of breast cancer. The odds ratio for women with age at first birth <25 years and ≥3 pregnancies, relative to nulliparous women, was 0.49, 95% confidence interval 0.13–0.79, and that for postmenopausal women relative to premenopausal women was 0.60, 95% CI 0.37–0.99. Conclusions: Further study of personal factors influencing the risk of breast cancer in women with BBD may help to identify subgroups of the population at increased risk of invasive disease.     

Lysophosphatidic acid and apolipoprotein A1 predict increased risk of developing World Trade Center-lung injury: a nested case-control study

Abstract

Rationale: Metabolic syndrome, inflammatory and vascular injury markers measured in serum after World Trade Center (WTC) exposures predict abnormal FEV₁. We hypothesized that elevated LPA levels predict FEV₁?<?LLN.

Methods: Nested case-control study of WTC-exposed firefighters. Cases had FEV₁?<?LLN. Controls derived from the baseline cohort. Demographics, pulmonary function, serum lipids, LPA and ApoA1 were measured.

Results: LPA and ApoA1 levels were higher in cases than controls and predictive of case status. LPA increased the odds by 13% while ApoA1 increased the odds by 29% of an FEV₁?<?LLN in a multivariable model.

Conclusions: Elevated LPA and ApoA1 are predictive of a significantly increased risk of developing an FEV₁?<?LLN.     

Inference of haplotype effects in case-control studies using unphased genotype and environmental data

A retrospective likelihood-based approach was proposed to test and estimate the effect of haplotype on disease risk using unphased genotype data with adjustment for environmental covariates. The proposed method was also extended to handle the data in which the haplotype and environmental covariates are not independent. Likelihood ratio tests were constructed to test the effects of haplotype and gene-environment interaction. The model parameters such as haplotype effect size was estimated using an Expectation Conditional-Maximization (ECM) algorithm developed by Meng and Rubin (1993). Model-based variance estimates were derived using the observed information matrix. Simulation studies were conducted for three different genetic effect models, including dominant effect, recessive effect, and additive effect. The results showed that the proposed method generated unbiased parameter estimates, proper type I error, and true beta coverage probabilities. The model performed well with small or large sample sizes, as well as short or long haplotypes.     

Semiparametric methods for evaluating risk prediction markers in case-control studies

The performance of a well-calibrated risk model for a binary disease outcome can be characterized by the population distribution of risk and displayed with the predictiveness curve. Better performance is characterized by a wider distribution of risk, since this corresponds to better risk stratification in the sense that more subjects are identified at low and high risk for the disease outcome. Although methods have been developed to estimate predictiveness curves from cohort studies, most studies to evaluate novel risk prediction markers employ case-control designs. Here we develop semiparametric methods that accommodate case-control data. The semiparametric methods are flexible, and naturally generalize methods previously developed for cohort data. Applications to prostate cancer risk prediction markers illustrate the methods.     

Tests of homogeneity for the relative risk in multiply-matched case-control studies

We examine several tests of homogeneity of the odds ratio in the analysis of 2 x 2 tables arising from epidemiologic 1:R matched case-control studies. The T4 and T5 statistics proposed by Liang and Self (1985, Biometrika 72, 353-358) are unable to detect obvious inhomogeneity in two numerical examples and in simulation studies. The null hypothesis is rejected by the chi-square statistic of Ejigou and McHugh (1984, Biometrika 71, 408-411) and by a new proposed method whose significance level must be simulated.     

Capitation and enhanced fee-for-service models for primary care reform: a population-based evaluation

Background

Primary care reform in Ontario, Canada, included the initiation of a blended capitation model in 2001–2002 and an enhanced fee-for-service model in 2003. Both models involve patient rostering, incentives for preventive care and requirements for after-hours care. We evaluated practice characteristics and patterns of care under both models.

Methods

Using administrative data, we identified physicians belonging to either the capitation or the enhanced fee-for-service group throughout the period from Sept. 1, 2005, to Aug. 31, 2006, and their enrolled patients. Practices were stratified by location (urban v. rural). We compared the groups in terms of practice characteristics and patterns of care, including comprehensiveness of care, continuity of care, after-hours care, visits to the emergency department and uptake of new patients.

Results

Patients in the capitation and enhanced fee-for-service practices had similar demographic characteristics. Patients in capitation practices had lower morbidity and comorbidity indices. Comprehensiveness and continuity of care were similar between the 2 groups. Compared with patients in enhanced fee-for-service practices, those in capitation practices had less after-hours care (adjusted rate ratio [RR] 0.68, 95% confidence interval [CI] 0.61–0.75) and more visits to emergency departments (adjusted RR 1.20, 95% CI 1.15–1.25). Overall, physicians in the capitation group enrolled fewer new patients than did physicians in the enhanced fee-for-service group (37.0 v. 52.0 per physician); the same was true of new graduates (60.3 v. 72.1 per physician).

Interpretation

Physicians enrolled in the capitation model had different practice characteristics than those in the enhanced fee-for-service model. These characteristics appeared to be pre-existing and not due to enrolment in a new model. Although the capitation model provides an alternative to fee-for-service practice, its characteristics should be the focus of future policy development and research.Primary health care is facing a number of serious challenges internationally, with questions being raised about whether it will even survive in some settings.¹ Fundamental issues include shortages in human resources and maldistribution of physicians; dissatisfaction on the part of providers and patients; gaps between guideline-recommended care and provided care; and a preference of trainees to choose specialty careers. Close to 4 million Canadians do not have a family physician, and more than 2 million report difficulties in accessing routine or ongoing care at any time of day as well as immediate care for minor health problems at any time of day.² Canadians in rural areas face geographic barriers to care, fewer available health care professionals than in urban areas and higher rates of disease.³In response to these challenges, policy-makers in Canada and elsewhere are considering or are implementing interdisciplinary teams, new organizational structures, new governance and reimbursement models, requirements for after-hours care, provision of after-hours advice by telephone, electronic health records and other information technology, and pay-for-performance initiatives. Many of these directions are incorporated in the Medical Home concept in the United States⁴ and in the Quality and Outcomes Framework in the United Kingdom.⁵ Although there is evidence for the effectiveness of some of these initiatives, most have not been rigorously evaluated. Reimbursement models, perhaps the best-studied aspect of primary care reform, seem to influence some aspects of physician behaviour. However, there is a lack of evidence about their ultimate impact on patient outcomes.⁶In Ontario, Canada, a blended capitation model called the Family Health Network was introduced in 2001–2002. An enhanced fee-for-service blended model called the Family Health Group was introduced in 2003. These models rapidly attracted physicians. By 2006, they were the most common models of care in Ontario, exceeding the straight fee-for-service plan.Physicians are free to select one of the models or remain in the straight fee-for-service plan. Many make decisions based on a free revenue analysis that uses their previous billings to project their income under the capitation model. Our evaluation, involving more than 500 physicians and close to half a million patients under the capitation model, is therefore an examination of one of the world’s largest short-term voluntary shifts from fee-for-service to capitation. Our objective was to evaluate practice characteristics and patterns of care under the capitation model, including comprehensiveness, continuity, after-hours care, visits to the emergency department and uptake of unattached patients. We used practices in the enhanced fee-for-service model as a contemporaneous comparison group.     

Neonatal risk factors for cerebral palsy in very preterm babies: case-control study.

OBJECTIVE: To identify neonatal risk factors for cerebral palsy among very preterm babies and in particular the associations independent of the coexistence of antenatal and intrapartum factors. DESIGN: Case-control study. SETTING: Oxford health region. SUBJECTS: Singleton babies born between 1984 and 1990 at less than 32 weeks'' gestation who survived to discharge from hospital: 59 with cerebral palsy and 234 randomly selected controls without cerebral palsy. MAIN OUTCOME MEASURES: Adverse neonatal factors expressed as odds ratios and 95% confidence intervals. RESULTS: Factors associated with an increased risk of cerebral palsy after adjustment for gestational age and the presence of previously identified antenatal and intrapartum risk factors were patent ductus arteriosus (odds ratio 2.3; 95% confidence interval 1.2 to 4.5), hypotension (2.3; 1.3 to 4.7), blood transfusion (4.8; 2.5 to 9.3), prolonged ventilation (4.8; 2.5 to 9.0), pneumothorax (3.5; 1.6 to 7.6), sepsis (3.6; 1.8 to 7.4), hyponatraemia (7.9; 2.1 to 29.6) and total parenteral nutrition (5.5; 2.8 to 10.5). Seizures were associated with an increased risk of cerebral palsy (10.0; 4.1 to 24.7), as were parenchymal damage (32; 12.4 to 84.4) and appreciable ventricular dilatation (5.4; 3.0 to 9.8) detected by cerebral ultrasound. CONCLUSION: A reduction in the rate of cerebral palsy in very preterm babies requires an integrated approach to management throughout the antenatal, intrapartum, and neonatal periods.     

Serum 25-hydroxyvitamin D and risk of lung cancer in male smokers: a nested case-control study

Background

A role for vitamin D in cancer risk reduction has been hypothesized, but few data exist for lung cancer. We investigated the relationship between vitamin D status, using circulating 25-hydroxyvitamin D [25(OH)D], and lung cancer risk in a nested case-control study within the Alpha-Tocopherol, Beta-Carotene Cancer Prevention Study of Finnish male smokers.

Methods

Lung cancer cases (n = 500) were randomly selected based on month of blood collection, and 500 controls were matched to them based on age and blood collection date. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated using multivariate-adjusted conditional logistic regression. To account for seasonal variation in 25(OH)D concentrations, season-specific and season-standardized quintiles of 25(OH)D were examined, and models were also stratified on season of blood collection (darker season = November–April and sunnier season = May–October). Pre-determined, clinically-defined cutpoints for 25(OH)D and 25(OH)D as a continuous measure were also examined.

Results

Overall, 25(OH)D was not associated with lung cancer. Risks were 1.08 (95% CI 0.67–1.75) and 0.83 (95% CI 0.53–1.31) in the highest vs. lowest season-specific and season-standardized quintiles of 25(OH)D, respectively, and 0.91 (95% CI 0.48–1.72) for the ≥75 vs. <25 nmol/L clinical categories. Inverse associations were, however, suggested for subjects with blood collections from November–April, with ORs of 0.77 (95% CI 0.41–1.45, p-trend = 0.05) and 0.65 (95% CI 0.37–1.14, p-trend = 0.07) in the highest vs. lowest season-specific and season-standardized quintiles of 25(OH)D, respectively, and 0.61 (95% CI 0.24–1.52, p-trend = 0.01) for ≥75 vs. <25 nmol/L. We also found 11% lower risk for a 10 nmol/L increase in 25(OH)D in the darker season based on the continuous measure (OR = 0.89, 95% CI 0.81–0.98, p = 0.02).

Conclusion

In this prospective study of male smokers, circulating 25(OH)D was not associated with lung cancer risk overall, although inverse associations were suggested among those whose blood was drawn during darker months.