Causal models in epidemiology: past inheritance and genetic future

Rosuvastatin-induced rhabdomyolysis: case report and call for proactive multifactorial risk assessment and preventive management of statin therapy in high-risk patients

Cholesterol-lowering statins are frequently prescribed for primary and secondary prevention of ischaemic vascular events. Whereas most patients tolerate statins without problems, statin-associated myopathy is well documented, as are several risk factors. We present a case report of an 80-90-year-old man with coronary artery disease who rapidly developed severe rhabdomyolysis during treatment with rosuvastatin while in intensive care. He had several concomitant risk factors for statin-induced myopathy including high dosage, old age, renal and hepatic impairment, and a pharmacogenetic SLCO1B1*1 a/*5 variant. Single known risk factors have a low predictive value for statin-induced myopathy and may therefore be underestimated in clinical practice. However, adverse drug reactions frequently involve the joint action of a multitude of environmental and genetic component causes, and statin-induced myopathy should be regarded as a multicausal event. We therefore advocate a proactive multifactorial risk assessment to guide and individualise statin therapy in high-risk patients.

Causal Criteria in Medical and Biological Disciplines: History, Essence, and Radiation Aspects. Report 4, Part 1: The Post-Hill Criteria and Ecolgoical Criteria

Part 1 of Report 4 is focused on the development and modifications of causal criteria after A.B. Hill (1965). Criteria from B. MacMahon et al. (1970-1996), regarded as the first textbook for modern epidemiology, were considered, and it was found that the named researchers did not offer anything new despite the frequent mention of this source in relation to the theme. A similar situation emerged with the criteria of M. Susser: the three obligatory points of this author, "Association" (or "Probability" of causality), "Time order," and "Direction of effect," are trivial, and two more special criteria, which are the development of "Popperian Epidemiology," i.e., "Surviability" of the hypothesis when it is tested by different methods (included in the refinement in Hill's criterion "Consistency of association") and "Predictive performance" of the hypothesis are more theoretical and hardly applicable for the practice of epidemiology and public health. The same restrictions apply to the similar "Popperian" criteria of D.L. Weed, "Predictability" and "Testability" of the causal hypothesis. Although the universal postulates of A.S. Evans for infectious and noninfectious pathologies can be considered exhaustive, they are not used either in epidemiology or in any other discipline practice, except for the field of infectious pathologies, which is probably explained by the complication of the ten-point complex. The little-known criteria of P. Cole (1997) for medical and forensic practice are the most important. The three parts of Hill's criterion-based approaches are important in that they go from a single epidemiological study through a cycle of studies (coupled with the integration of data from other biomedical disciplines) to re-base Hill's criteria for assessing the individual causality of an effect. These constructs complement the earlier guidance from R.E. Gots (1986) on establishing probabilistic personal causation. The collection of causal criteria and the guidelines for environmental disciplines (ecology of biota, human ecoepidemiology, and human ecotoxicology) were considered. The total dominance of inductive causal criteria, both initial and in modifications and with additions, was revealed for an apparently complete base of sources (1979-2020). Adaptations of all known causal schemes based on guidelines have been found, from Henle-Koch postulates to Hill and Susser, including in the international programs and practice of the U.S. Environmental Protection Agency. The Hill Criteria are used by the WHO and other organizations on chemical safety (IPCS) to assess causality in animal experiments for subsequent extrapolation to humans. Data on the assessment of the causality of effects in ecology, ecoepidemiology, and ecotoxicology, together with the use of Hill's criteria for animal experiments, are of significant relevance not only for radiation ecology, but also for radiobiology.

Chronic mechanical irritation enhances the effect of tobacco and alcohol on the risk of oral squamous cell carcinoma: a case-control study in Argentina

OBJECTIVES

Oral squamous cell carcinoma (OSCC) is a multifactorial disease. The individual effect of each risk factor for OSCC may be conditioned by the frequency of other factors. The objective of this study was to identify the association between chronic mechanical irritation (CMI) and OSCC and to analyse the influence of CMI on other important risk factors for OSCC.

MATERIALS AND METHODS

A prospective and age/sex-matched case-control study was performed in two institutions from Argentina between 2009 and 2019, with consecutive and newly diagnosed OSCC. The frequencies of tobacco, alcohol, and CMI were analysed using conditional logistic regression. Cumulative tobacco consumption and the presence of CMI were analysed using the Mann-Whitney test.

RESULTS

CMI and OSCC were associated with an OR of 7.02 (95% CI 3.57-13.78, p < 0.001). The combination of CMI and alcohol demonstrated the highest risk of OSCC (OR 53.83, CI 95% 8.04-360, p < 0.0001), followed by the combination of CMI, tobacco, and alcohol (OR 48.06, CI 95% 8.47-272, p < 0.0001). The combination of CMI and tobacco was also significant (OR 5.61, CI 95% 1.07-29.54, p = 0.042). Patients with CMI developed OSCC with less cumulative tobacco use compared with those without CMI.

CONCLUSION

CMI is an independent risk factor for OSCC, and it could act as a risk modifier among tobacco and alcohol users having an enhancing effect.

CLINICAL RELEVANCE

Elimination of CMI could decrease the risk of OSCC.

Big Data to Knowledge: Application of Machine Learning to Predictive Modeling of Therapeutic Response in Cancer

Background

In recent years, the availability of high throughput technologies, establishment of large molecular patient data repositories, and advancement in computing power and storage have allowed elucidation of complex mechanisms implicated in therapeutic response in cancer patients. The breadth and depth of such data, alongside experimental noise and missing values, requires a sophisticated human-machine interaction that would allow effective learning from complex data and accurate forecasting of future outcomes, ideally embedded in the core of machine learning design.

Objective

In this review, we will discuss machine learning techniques utilized for modeling of treatment response in cancer, including Random Forests, support vector machines, neural networks, and linear and logistic regression. We will overview their mathematical foundations and discuss their limitations and alternative approaches in light of their application to therapeutic response modeling in cancer.

Conclusion

We hypothesize that the increase in the number of patient profiles and potential temporal monitoring of patient data will define even more complex techniques, such as deep learning and causal analysis, as central players in therapeutic response modeling.

Exploring the Biological Activity and Mechanism of Xenoestrogens and Phytoestrogens in Cancers: Emerging Methods and Concepts

Xenoestrogens and phytoestrogens are referred to as "foreign estrogens" that are produced outside of the human body and have been shown to exert estrogen-like activity. Xenoestrogens are synthetic industrial chemicals, whereas phytoestrogens are chemicals present in the plant. Considering that these environmental estrogen mimics potentially promote hormone-related cancers, an understanding of how they interact with estrogenic pathways in human cells is crucial to resolve their possible impacts in cancer. Here, we conducted an extensive literature evaluation on the origins of these chemicals, emerging research techniques, updated molecular mechanisms, and ongoing clinical studies of estrogen mimics in human cancers. In this review, we describe new applications of patient-derived xenograft (PDX) models and single-cell RNA sequencing (scRNA-seq) techniques in shaping the current knowledge. At the molecular and cellular levels, we provide comprehensive and up-to-date insights into the mechanism of xenoestrogens and phytoestrogens in modulating the hallmarks of cancer. At the systemic level, we bring the emerging concept of window of susceptibility (WOS) into focus. WOS is the critical timing during the female lifespan that includes the prenatal, pubertal, pregnancy, and menopausal transition periods, during which the mammary glands are more sensitive to environmental exposures. Lastly, we reviewed 18 clinical trials on the application of phytoestrogens in the prevention or treatment of different cancers, conducted from 2002 to the present, and provide evidence-based perspectives on the clinical applications of phytoestrogens in cancers. Further research with carefully thought-through concepts and advanced methods on environmental estrogens will help to improve understanding for the identification of environmental influences, as well as provide novel mechanisms to guide the development of prevention and therapeutic approaches for human cancers.

'Dark matter', second waves and epidemiological modelling

Recent reports using conventional Susceptible, Exposed, Infected and Removed models suggest that the next wave of the COVID-19 pandemic in the UK could overwhelm health services, with fatalities exceeding the first wave. We used Bayesian model comparison to revisit these conclusions, allowing for heterogeneity of exposure, susceptibility and transmission. We used dynamic causal modelling to estimate the evidence for alternative models of daily cases and deaths from the USA, the UK, Brazil, Italy, France, Spain, Mexico, Belgium, Germany and Canada over the period 25 January 2020 to 15 June 2020. These data were used to estimate the proportions of people (i) not exposed to the virus, (ii) not susceptible to infection when exposed and (iii) not infectious when susceptible to infection. Bayesian model comparison furnished overwhelming evidence for heterogeneity of exposure, susceptibility and transmission. Furthermore, both lockdown and the build-up of population immunity contributed to viral transmission in all but one country. Small variations in heterogeneity were sufficient to explain large differences in mortality rates. The best model of UK data predicts a second surge of fatalities will be much less than the first peak. The size of the second wave depends sensitively on the loss of immunity and the efficacy of Find-Test-Trace-Isolate-Support programmes. In summary, accounting for heterogeneity of exposure, susceptibility and transmission suggests that the next wave of the SARS-CoV-2 pandemic will be much smaller than conventional models predict, with less economic and health disruption. This heterogeneity means that seroprevalence underestimates effective herd immunity and, crucially, the potential of public health programmes.

Effective immunity and second waves: a dynamic causal modelling study

This technical report addresses a pressing issue in the trajectory of the coronavirus outbreak; namely, the rate at which effective immunity is lost following the first wave of the pandemic. This is a crucial epidemiological parameter that speaks to both the consequences of relaxing lockdown and the propensity for a second wave of infections. Using a dynamic causal model of reported cases and deaths from multiple countries, we evaluated the evidence models of progressively longer periods of immunity. The results speak to an effective population immunity of about three months that, under the model, defers any second wave for approximately six months in most countries. This may have implications for the window of opportunity for tracking and tracing, as well as for developing vaccination programmes, and other therapeutic interventions.

Effective immunity and second waves: a dynamic causal modelling study

This technical report addresses a pressing issue in the trajectory of the coronavirus outbreak; namely, the rate at which effective immunity is lost following the first wave of the pandemic. This is a crucial epidemiological parameter that speaks to both the consequences of relaxing lockdown and the propensity for a second wave of infections. Using a dynamic causal model of reported cases and deaths from multiple countries, we evaluated the evidence models of progressively longer periods of immunity. The results speak to an effective population immunity of about three months that, under the model, defers any second wave for approximately six months in most countries. This may have implications for the window of opportunity for tracking and tracing, as well as for developing vaccination programmes, and other therapeutic interventions.

Computational Nosology and Precision Psychiatry

This article provides an illustrative treatment of psychiatric morbidity that offers an alternative to the standard nosological model in psychiatry. It considers what would happen if we treated diagnostic categories not as causes of signs and symptoms, but as diagnostic consequences of psychopathology and pathophysiology. This reformulation (of the standard nosological model) opens the door to a more natural description of how patients present-and of their likely responses to therapeutic interventions. In brief, we describe a model that generates symptoms, signs, and diagnostic outcomes from latent psychopathological states. In turn, psychopathology is caused by pathophysiological processes that are perturbed by (etiological) causes such as predisposing factors, life events, and therapeutic interventions. The key advantages of this nosological formulation include (i) the formal integration of diagnostic (e.g., DSM) categories and latent psychopathological constructs (e.g., the dimensions of the Research Domain Criteria); (ii) the provision of a hypothesis or model space that accommodates formal, evidence-based hypothesis testing (using Bayesian model comparison); and (iii) the ability to predict therapeutic responses (using a posterior predictive density), as in precision medicine. These and other advantages are largely promissory at present: The purpose of this article is to show what might be possible, through the use of idealized simulations.

Transdisciplinary approaches enhance the production of translational knowledge

The primary goal of translational research is to generate and apply knowledge that can improve human health. Although research conducted within the confines of a single discipline has helped us to achieve this goal in many settings, this unidisciplinary approach may not be optimal when disease causation is complex and health decisions are pressing. To address these issues, we suggest that transdisciplinary approaches can facilitate the progress of translational research, and we review publications that demonstrate what these approaches can look like. These examples serve to (1) demonstrate why transdisciplinary research is useful, and (2) stimulate a conversation about how it can be further promoted. While we note that open-minded communication is a prerequisite for germinating any transdisciplinary work and that epidemiologists can play a key role in promoting it, we do not propose a rigid protocol for conducting transdisciplinary research, as one really does not exist. These achievements were developed in settings where typical disciplinary and institutional barriers were surmountable, but they were not accomplished with a single predetermined plan. The benefits of cross-disciplinary communication are hard to predict a priori and a detailed research protocol or process may impede the realization of novel and important insights. Overall, these examples demonstrate that enhanced cross-disciplinary information exchange can serve as a starting point that helps researchers frame better questions, integrate more relevant evidence, and advance translational knowledge more effectively. Specifically, we discuss examples where transdisciplinary approaches are helping us to better explore, assess, and intervene to improve human health.

Considerations for Using Genetic and Epigenetic Information in Occupational Health Risk Assessment and Standard Setting

Risk assessment forms the basis for both occupational health decision-making and the development of occupational exposure limits (OELs). Although genetic and epigenetic data have not been widely used in risk assessment and ultimately, standard setting, it is possible to envision such uses. A growing body of literature demonstrates that genetic and epigenetic factors condition biological responses to occupational and environmental hazards or serve as targets of them. This presentation addresses the considerations for using genetic and epigenetic information in risk assessments, provides guidance on using this information within the classic risk assessment paradigm, and describes a framework to organize thinking about such uses. The framework is a 4 × 4 matrix involving the risk assessment functions (hazard identification, dose-response modeling, exposure assessment, and risk characterization) on one axis and inherited and acquired genetic and epigenetic data on the other axis. The cells in the matrix identify how genetic and epigenetic data can be used for each risk assessment function. Generally, genetic and epigenetic data might be used as endpoints in hazard identification, as indicators of exposure, as effect modifiers in exposure assessment and dose-response modeling, as descriptors of mode of action, and to characterize toxicity pathways. Vast amounts of genetic and epigenetic data may be generated by high-throughput technologies. These data can be useful for assessing variability and reducing uncertainty in extrapolations, and they may serve as the foundation upon which identification of biological perturbations would lead to a new paradigm of toxicity pathway-based risk assessments.

Considerations for Using Genetic and Epigenetic Information in Occupational Health Risk Assessment and Standard Setting

Risk assessment forms the basis for both occupational health decision-making and the development of occupational exposure limits (OELs). Although genetic and epigenetic data have not been widely used in risk assessment and ultimately, standard setting, it is possible to envision such uses. A growing body of literature demonstrates that genetic and epigenetic factors condition biological responses to occupational and environmental hazards or serve as targets of them. This presentation addresses the considerations for using genetic and epigenetic information in risk assessments, provides guidance on using this information within the classic risk assessment paradigm, and describes a framework to organize thinking about such uses. The framework is a 4 × 4 matrix involving the risk assessment functions (hazard identification, dose-response modeling, exposure assessment, and risk characterization) on one axis and inherited and acquired genetic and epigenetic data on the other axis. The cells in the matrix identify how genetic and epigenetic data can be used for each risk assessment function. Generally, genetic and epigenetic data might be used as endpoints in hazard identification, as indicators of exposure, as effect modifiers in exposure assessment and dose-response modeling, as descriptors of mode of action, and to characterize toxicity pathways. Vast amounts of genetic and epigenetic data may be generated by high-throughput technologies. These data can be useful for assessing variability and reducing uncertainty in extrapolations, and they may serve as the foundation upon which identification of biological perturbations would lead to a new paradigm of toxicity pathway-based risk assessments.

Making valid causal inferences from observational data

The ability to make strong causal inferences, based on data derived from outside of the laboratory, is largely restricted to data arising from well-designed randomized control trials. Nonetheless, a number of methods have been developed to improve our ability to make valid causal inferences from data arising from observational studies. In this paper, I review concepts of causation as a background to counterfactual causal ideas; the latter ideas are central to much of current causal theory. Confounding greatly constrains causal inferences in all observational studies. Confounding is a biased measure of effect that results when one or more variables, that are both antecedent to the exposure and associated with the outcome, are differentially distributed between the exposed and non-exposed groups. Historically, the most common approach to control confounding has been multivariable modeling; however, the limitations of this approach are discussed. My suggestions for improving causal inferences include asking better questions (relates to counterfactual ideas and "thought" trials); improving study design through the use of forward projection; and using propensity scores to identify potential confounders and enhance exchangeability, prior to seeing the outcome data. If time-dependent confounders are present (as they are in many longitudinal studies), more-advanced methods such as marginal structural models need to be implemented. Tutorials and examples are cited where possible.

Epigenetic influences in the aetiology of cancers arising from breast and prostate: a hypothesised transgenerational evolution in chromatin accessibility

Epidemiological studies have consistently supported the notion that environmental and/or dietary factors play a central role in the aetiology of cancers of the breast and prostate. However, for more than five decades investigators have failed to identify a single cause-and-effect factor, which could be implicated; identification of a causative entity would allow the implementation of an intervention strategy in at-risk populations. This suggests a more complex pathoaetiology for these cancer sites, compared to others. When one examines the increases or decreases in incidence of specific cancers amongst migrant populations, it is notable that disease arising in colon or stomach requires one or at most two generations to exhibit a change in incidence to match that of high-incidence regions, whereas for breast or prostate cancer, at least three generations are required. This generational threshold could suggest a requirement for nonmutation-driven epigenetic alterations in the F0/F1 generations (parental/offspring adopting a more westernized lifestyle), which then predisposes the inherited genome of subsequent generations to mutagenic/genotoxic alterations leading to the development of sporadic cancer in these target sites. As such, individual susceptibility to carcinogen insult would not be based per se on polymorphisms in activating/detoxifying/repair enzymes, but on elevated accessibility of crucial target genes (e.g., oncogenes, tumour suppressor genes) or hotspots therein to mutation events. This could be termed a genomic susceptibility organizational structure (SOS). Several exposures including alcohol and heavy metals are epigens (i.e., modifiers of the epigenome), whereas others are mutagenic/genotoxic, for example, heterocyclic aromatic amines; humans are continuously and variously exposed to mixtures of these agents. Within such a transgenerational multistage model of cancer development, determining the interaction between epigenetic modification to generate a genomic SOS and genotoxic insult will facilitate a new level of understanding in the aetiology of cancer.

Interaction of occupational and personal risk factors in workforce health and safety

Most diseases, injuries, and other health conditions experienced by working people are multifactorial, especially as the workforce ages. Evidence supporting the role of work and personal risk factors in the health of working people is frequently underused in developing interventions. Achieving a longer, healthy working life requires a comprehensive preventive approach. To help develop such an approach, we evaluated the influence of both occupational and personal risk factors on workforce health. We present 32 examples illustrating 4 combinatorial models of occupational hazards and personal risk factors (genetics, age, gender, chronic disease, obesity, smoking, alcohol use, prescription drug use). Models that address occupational and personal risk factors and their interactions can improve our understanding of health hazards and guide research and interventions.

Expanding the scope of risk assessment: methods of studying differential vulnerability and susceptibility

Several methodological issues have been identified in analysis of epidemiological data to better assess the distributional effects of exposures and hypotheses about effect modification. We discuss the hierarchical mixed model and some more complex methods. Methods of capturing inequality are a second dimension of risk assessment, and simulation studies are important because plausible choices for air pollution effects and effect modifiers could result in extremely high risks in a small subset of the population. Future epidemiological studies should explore contextual and individual-level factors that might modify these relationships. The Environmental Protection Agency should make this a standard part of their risk assessments whenever the necessary information is available.

Causal models for investigating complex disease: I. A primer

BACKGROUND/AIMS

To illustrate the utility of causal models for research in genetic epidemiology and statistical genetics. Causal models are increasingly applied in risk factor epidemiology, economics, and health policy, but seldom used in statistical genetics or genetic epidemiology. Unlike the statistical models usually used in genetic epidemiology, causal models are explicitly formulated in terms of cause and effect relationships occurring at the individual level.

METHODS

We describe two causal models, the sufficient component cause model and the potential outcomes model, and show how key concepts in genetic epidemiology, including penetrance, phenocopies, genetic heterogeneity, etiologic heterogeneity, gene-gene interaction, and gene-environment interaction, can be framed in terms of these causal models. We also illustrate how potential outcomes models can provide insight into the potential for confounding and bias in the measurement of causal effects in genetic studies.

RESULTS

Our analysis illustrates how causal models can elucidate the relationships among underlying causal mechanisms and measures obtained from statistical analysis of observed data.

CONCLUSION

Causal models can enhance research aimed at identifying causal genes.

Challenges in the use of literature-based meta-analysis to examine gene-environment interactions

Statistical interactions between genes and environmental exposures with respect to disease outcomes may help to identify biologic mechanisms and pathways and inform behavioral interventions. The number of persons required for a single study to have sufficient statistical power to detect such interactions may be considered prohibitively large, making a meta-analysis of published literature an apparently attractive alternative. However, meta-analysis of gene-environment interactions using published literature is challenging, with the conclusions being likely to suffer from bias and lack of generalizability. The authors highlight these challenges and biases using an illustrative example: meta-analysis of interactions between the Pro12Ala variant of the peroxisome proliferator-activated receptor gamma (PPARgamma) gene and various diet and lifestyle factors in the risk of diabetes. The authors conclude that literature-based meta-analysis conducted to examine gene-environment interactions is unlikely to provide a meaningful quantitative conclusion. Alternative strategies are required, including analyses in scientific consortia established to assess main genetic effects, where individual participant data can be shared, allowing both greater power and consistency of analysis methods. However, these consortia are likely to be limited by lack of standardization of the measures of environmental factors. This issue may ultimately only be resolvable by the de novo establishment of large single or multicenter cohorts using comparable methods.

Approaches to advancing quantitative human health risk assessment of environmental chemicals in the post-genomic era

The contribution of genomics and associated technologies to human health risk assessment for environmental chemicals has focused largely on elucidating mechanisms of toxicity, as discussed in other articles in this issue. However, there is interest in moving beyond hazard characterization to making more direct impacts on quantitative risk assessment (QRA)--i.e., the determination of toxicity values for setting exposure standards and cleanup values. We propose that the evolution of QRA of environmental chemicals in the post-genomic era will involve three, somewhat overlapping phases in which different types of approaches begin to mature. The initial focus (in Phase I) has been and continues to be on "augmentation" of weight of evidence--using genomic and related technologies qualitatively to increase the confidence in and scientific basis of the results of QRA. Efforts aimed towards "integration" of these data with traditional animal-based approaches, in particular quantitative predictors, or surrogates, for the in vivo toxicity data to which they have been anchored are just beginning to be explored now (in Phase II). In parallel, there is a recognized need for "expansion" of the use of established biomarkers of susceptibility or risk of human diseases and disorders for QRA, particularly for addressing the issues of cumulative assessment and population risk. Ultimately (in Phase III), substantial further advances could be realized by the development of novel molecular and pathway-based biomarkers and statistical and in silico models that build on anticipated progress in understanding the pathways of human diseases and disorders. Such efforts would facilitate a gradual "reorientation" of QRA towards approaches that more directly link environmental exposures to human outcomes.

The impact of new research technologies on our understanding of environmental causes of disease: the concept of clinical vulnerability

In spite of decades of epidemiological research, the etiology and causal patterns for many common diseases, such as breast and colon cancer or neurodegenerative diseases, are still largely unknown. Such chronic diseases are likely to have an environmental origin. However, "environmental" risks have been often elusive in epidemiological studies. This is a conundrum for current epidemiological research. On the other side, the relative contribution of genes to chronic diseases, as emerging from GWAS, seems to be modest (15-50% increase in disease risk). What is yet to be explored extensively is a model of disease based on long-term effects of low doses of environmental exposures, incorporating both genetic and acquired susceptibility ("clinical vulnerability"), and the cumulative effects of different exposures. Such a disease model would be compatible with the weak associations found by GWAS and the still elusive role of many (low-level) environmental exposures. We also propose that the introduction of "-omic" high-throughput technologies, such as transcriptomics, proteomics and metabolomics, may provide, in the next years, powerful tools to investigate early effects of environmental exposures and understand the etiology of common diseases better, according to the "clinical vulnerability model". The development of "-omics", in spite of current limitations and lack of sound validation, could greatly contribute to the elucidation of the disease model we propose.

Invited commentary: genes, environment, and hybrid vigor

In the 1950s, case-control studies of smoking and lung cancer established a paradigm for epidemiologic studies of risk factors for chronic diseases. Since then, thousands of case-control studies have examined possible associations of countless risk factors with numerous diseases, rarely finding associations as strong or consistent as that of smoking with lung cancer. Recently, researchers have applied advances in molecular genetics to conduct candidate gene and genome-wide association studies of lung cancer. Skeptics among both epidemiologists and geneticists have argued that genomic research adds little value when most cases of disease can be attributed to a preventable exposure; however, well-conducted studies of gene-environment interactions that draw on data from more than 50 years of research in toxicology, pathophysiology, and behavioral science offer important models for the development of more comprehensive approaches to understanding the etiology of chronic diseases.

Should burden of disease estimates include cannabis use as a risk factor for psychosis?

Louise Degenhardt and colleagues discuss the evidence and the debate about whether Global Burden of Disease estimates should include cannabis use as a risk factor for psychosis.

Interpreting epidemiologic studies of developmental neurotoxicity: conceptual and analytic issues

This paper discusses issues pertaining to the validity, precision, and interpretation of epidemiologic studies of neurotoxicity. With regard to validity, the critical issues pertain to the appropriate strategy for confounder adjustment, particularly when confounders are complex, multi-faceted constructs, and to the need for greater clarity and transparency in articulating the causal relationships implicit in the analytic approach applied. With regard to precision, the critical issue is a need to identify the contributors to the substantial variability observed in the effect estimates that describe dose-response and dose-effect relationships. In addition to methodological sources, such as imprecision in estimating dose at the critical organ site(s), true inter-individual differences in susceptibility to a neurotoxicant could also contribute to the variability. Variability might be reduced by taking full account of factors such as co-exposures or health co-morbidities, genetic polymorphisms, and the social ecology of exposure. With regard to interpretation, we need to do a better job as a field conveying to risk assessors and others the ecological significance of the types of performance deficits observed following neurotoxicant exposure, emphasizing the distinction between individual and population risk. A final issue discussed is the need to define standards for the conduct, analysis, and reporting of epidemiologic studies of neurotoxicity, similar to those developed for other fields.

A lifestage-specific approach to hazard and dose-response characterization for children's health risk assessment

In 2006, the U.S. EPA published a report entitled A Framework for Assessing Health Risks of Environmental Exposures to Children (hereafter referred to as the "Framework") describing a lifestage approach to risk assessment that includes the evaluation of existing data from a temporal perspective (i.e., the timing of both the exposure and the outcome). This article summarizes the lifestage-specific issues discussed in the Framework related to the qualitative and the quantitative hazard and dose-response characterization. Lifestage-specific hazard characterization includes an evaluation of relevant human and experimental animal studies, focusing on the identification of critical windows of development (i.e., exposure intervals of maximum susceptibility) for observed outcomes, evaluation of differential exposure at individual lifestages, the relevance and impact of lifestage-specific toxicokinetic and toxicodynamic data, mode of action information, variability and latency of effects from early lifestage exposure, and describing uncertainties. The interpretation of the hazard data to determine the strength of association between early life exposures and the timing and type of outcomes depends upon the overall weight of evidence. Lifestage-specific dose-response characterization relies on the identification of susceptible lifestages in order to quantify health risk, information on the point of departure, key default assumptions, and descriptions of uncertainty, sensitivity, and variability. Discussion of the strength and limitations of the hazard and dose-response data provides a basis for confidence in risk determinations. Applying a lifestage approach to hazard and dose-response characterization is likely to improve children's health risk assessment by identifying data gaps and providing a better understanding of sources of uncertainty.

Personalizing nutrigenomics research through community based participatory research and omics technologies

Personal and public health information are often obtained from studies of large population groups. Risk factors for nutrients, toxins, genetic variation, and more recently, nutrient-gene interactions are statistical estimates of the percentage reduction in disease in the population if the risk were to be avoided or the gene variant were not present. Because individuals differ in genetic makeup, lifestyle, and dietary patterns than those individuals in the study population, these risk factors are valuable guidelines, but may not apply to individuals. Intervention studies are likewise limited by small sample sizes, short time frames to assess physiological changes, and variable experimental designs that often preclude comparative or consensus analyses. A fundamental challenge for nutrigenomics will be to develop a means to sort individuals into metabolic groups, and eventually, develop risk factors for individuals. To reach the goal of personalizing medicine and nutrition, new experimental strategies are needed for human study designs. A promising approach for more complete analyses of the interaction of genetic makeups and environment relies on community-based participatory research (CBPR) methodologies. CBPR's central focus is developing a partnership among researchers and individuals in a community that allows for more in depth lifestyle analyses but also translational research that simultaneously helps improve the health of individuals and communities. The USDA-ARS Delta Nutrition Intervention Research program exemplifies CBPR providing a foundation for expanded personalized nutrition and medicine research for communities and individuals.

Response: if society is our patient, how shall we take its pulse?

A society's health status is not simply the sum of the health of individuals, but an emergent property. While epidemiology has evolved from the study of simple epidemics to the study of the determinants of population health, it is still appropriate to consider the population as the epidemiologist's patient. At the same time, the population is also our study subject-we study populations to identify ways to prevent disease in populations. This article illustrates some of the implications for the prevention of disease that result from treating society as our patient.

Consanguinity and birth defects in the jerusalem perinatal study cohort

BACKGROUND

While parental consanguinity is known to increase the risk of birth defects in offspring, it is hard to quantify this risk in populations where consanguinity is prevalent.

METHODS

To support ongoing studies of cancer and of psychiatric disease, we studied relationships of consanguinity to 1,053 major birth defects in 29,815 offspring, born in 1964-1976. To adjust for confounding variables (geographic origin, social class and hospital), we constructed logistic regression models, using GEE to take into account correlations between sibs. Odds ratios (ORs) and 95% confidence limits were estimated in comparison to a reference group of offspring with grandfathers born in different countries.

RESULTS

With 10.1% of offspring having consanguineous parents, the adjusted OR for major birth defect was 1.41 (1.12-1.74). Offspring of marriages between uncles-nieces, first cousins and more distant relatives showed adjusted ORs of 2.36 (0.98-5.68), 1.59 (1.22-2.07) and 1.20 (0.89-1.59) respectively. For descendents of grandfathers born in the same country, but not known to be related, the OR was 1.05 (0.91-1.21); these showed increased risk associated with ancestries in Western Asia (1.27, 1.04-1.55, p < 0.02) or Europe (1.13, 0.79-1.80).

CONCLUSIONS

A strong association of consanguinity with poverty and low education points to the need to avoid exposure to environmental hazards in these families.

Epidemiology of basal-like breast cancer

Risk factors for the newly identified "intrinsic" breast cancer subtypes (luminal A, luminal B, basal-like and human epidermal growth factor receptor 2-positive/estrogen receptor-negative) were determined in the Carolina Breast Cancer Study, a population-based, case-control study of African-American and white women. Immunohistochemical markers were used to subtype 1,424 cases of invasive and in situ breast cancer, and case subtypes were compared to 2,022 controls. Luminal A, the most common subtype, exhibited risk factors typically reported for breast cancer in previous studies, including inverse associations for increased parity and younger age at first full-term pregnancy. Basal-like cases exhibited several associations that were opposite to those observed for luminal A, including increased risk for parity and younger age at first term full-term pregnancy. Longer duration breastfeeding, increasing number of children breastfed, and increasing number of months breastfeeding per child were each associated with reduced risk of basal-like breast cancer, but not luminal A. Women with multiple live births who did not breastfeed and women who used medications to suppress lactation were at increased risk of basal-like, but not luminal A, breast cancer. Elevated waist-hip ratio was associated with increased risk of luminal A in postmenopausal women, and increased risk of basal-like breast cancer in pre- and postmenopausal women. The prevalence of basal-like breast cancer was highest among premenopausal African-American women, who also showed the highest prevalence of basal-like risk factors. Among younger African-American women, we estimate that up to 68% of basal-like breast cancer could be prevented by promoting breastfeeding and reducing abdominal adiposity.

Linking causal concepts, study design, analysis and inference in support of one epidemiology for population health

In this paper, which is dedicated to Dr. Calvin Schwabe, I review the concepts of causation and how they impact on study design, analysis and interpretation of results. Notwithstanding the fact that no observational study can prove causation, there are a number of issues that if addressed sufficiently, can improve the validity and usefulness of our studies. These are elaborated with specific recommendations for the conduct of future observational research. Approaches that have been useful in my teaching and research experience are also described.

Methodological Issues in Genetic Association Studies of Inherited Thrombophilia: Original Report of Recent Practice

The aims of this article are to evaluate the methodological quality of genetic association studies on the inherited thrombophilia published during 2003 to 2005, to identify the most common mistakes made by authors of those studies, and to examine if overall quality of the article correlates with the quality of the journal. Articles were evaluated by 2 independent reviewers using the checklist of 16 items. A total of 58 eligible studies were identified. Average total score was 7.59 ± 1.96. Total article score did not correlate with the journal impact factor (r = 0.3971; 95% confidence interval [CI], 0.1547-0.5944, P = .002). Total score did not differ across years ( P = .624). Finally, it is concluded that methodological quality of genetic association studies is not optimal, and it does not depend on the quality of the journal. Journals should adopt methodological criteria for reporting the genetic association studies, and editors should encourage authors to strictly adhere to those criteria.

Late neurodevelopmental effects of early exposures to chemical contaminants: reducing uncertainty in epidemiological studies

Early exposures to environmental chemicals are reliably associated with late neurotoxicities in children. However, substantial scatter of observations exists around the estimated dose-effect relationships. This variability has many potential sources, one of which is interindividual differences in susceptibility. Such differences imply that the long-term impacts of exposure will not the same for all individuals, but will vary depending on a variety of factors that might either aggravate or mitigate contaminant effects. These include co-exposures, genetic polymorphisms and characteristics of the social environment. The context dependence of contaminant effects has implications both for study designs and analytical approaches. In addition, a systems approach to understanding the associations among contaminant exposures, covariates and health outcomes is necessary.

Mode of action frameworks: a critical analysis

Mode of action (MOA) information is increasingly being applied in human health risk assessment. The MOA can inform issues such as the relevance of observed effects in laboratory animals to humans, and the variability of response within the human population. Several collaborative groups have developed frameworks for analyzing and utilizing MOA information in human health risk assessment of environmental carcinogens and toxins, including the International Programme on Chemical Safety, International Life Sciences Institute, and U.S. Environmental Protection Agency. With the goal of identifying gaps and opportunities for progress, we critically evaluate several of these MOA frameworks. Despite continued improvement in incorporating biological data in human health risk assessment, several notable challenges remain. These include articulation of the significant role of scientific judgment in establishing an MOA and its relevance to humans. In addition, binary (yes/no) decisions can inappropriately exclude consideration of data that may nonetheless be informative to the overall assessment of risk. Indeed, the frameworks lack a broad consideration of known causes of human disease and the potential for chemical effects to act additively with these as well as endogenous background processes. No integrated analysis of the impact of multiple MOAs over the same dose range, or of varying MOAs at different life stages, is included. Separate consideration of each MOA and outcome limits understanding of how multiple metabolites, modes, and toxicity pathways contribute to the toxicological profile of the chemical. An extension of the analyses across outcomes with common modes is also needed.

Epidemiological evidence for Mycobacterium avium subspecies paratuberculosis as a cause of Crohn's disease

Mycobacterium avium subspecies paratuberculosis is the causative agent of Johne's disease, a chronic enteritis in ruminants including cattle, sheep, goats, and farmed deer. Recently, this bacterium has received an increasingly wide interest because of a rapidly growing body of scientific evidence which suggests that human infection with this microorganism may be causing some, and possibly all, cases of Crohn's disease. Recent studies have shown that a high percentage of people with Crohn's disease are infected with M. avium subsp. paratuberculosis; whether the association of this bacterium and Crohn's disease is causal or coincidental is not known. Crohn's disease is a gastrointestinal disease in humans with similar histopathological findings to those observed in the paucibacillary form of Johne's disease in cattle. The search for risk factors in Crohn's disease has been frustrating. However, epidemiologists have gathered enough information that points to an association between M. avium subsp. paratuberculosis and Crohn's disease. This paper reviews epidemiological models of disease causation, the major philosophical doctrines about causation, the established epidemiological criteria for causation, and the currently known epidemiological evidence of M. avium subsp. paratuberculosis as a possible cause of Crohn's disease.

Insulin and IGF-1 mediated inhibition of apoptosis in CHO cells grown in suspension in a protein-free medium

When Chinese hamster ovary (CHO) cells were grown in suspension and deprived of serum, 40% of them became apoptotic after 72 hours, as determined by flow cytometry analysis of TUNEL-labelled cells. Cell viability, assessed by erythrocin B staining, decreased correspondingly. An increase in the total fraction of cells expressing interleukin converting enzyme (ICE; caspase 1), B-cell lymphoma 2 protein (Bcl-2,) and Bcl-2 associated x protein (Bax) was shown by antibody probing and subsequent flow cytometry. The p53 tumour suppressor gene product level remained low within the cell population. Insulin-like growth factor-1 (IGF-1) inhibited cell death in a concentration-dependent manner, and at 20 ng/ml, cell viability was maintained close to 100% and no apoptotic cells were detected. Also, insulin was shown to inhibit cell death - at 1.0 microg/ml, cell viability was 95%, whereas 10% of the cells stained for apoptosis. At the highest concentrations of IGF-1 and insulin, the expression of ICE, Bcl-2 and Bax was fully suppressed, whereas the p53 product level increased, despite still being detectable in a minority of cells. Under these conditions, IGF-1 may increase p53 expression to restrain abnormal cell proliferation. It is concluded that special attention should be paid to exposure and culture conditions that induce acquired susceptibility to a toxic insult, during the development and validation of cell-based assays.

Exposure and dose modelling in occupational epidemiology

Complex and dynamic physiologic processes underlie the exposure-response relations that occupational and environmental epidemiologists study. Simple summary measures of exposure such as the average, cumulative exposure, or duration of exposure, can be applied suitably in exposure-response analyses in many instances. However, there are situations where these metrics may not be directly proportional to risk, in which case their use will result in misclassification and biased estimates of exposure-response associations. We outline methods for developing exposure or dose metrics which may reduce misclassification, as illustrated with some recent examples. Selecting better exposure or dose metrics can be thought of as a problem of choosing appropriate weights on the exposure history of each cohort member. Dosimetric modeling involves choosing exposure weights based on formal hypotheses about underlying physiologic or pathogenetic processes. Dosimetric modeling is still not widely used in epidemiology, and so the forms of mathematical models and the criteria for choosing one model over another are not yet standardized. We hope to stimulate further applications through this presentation.

Bayesian homeopathy: talking normal again

Homeopathy has a communication problem: important homeopathic concepts are not understood by conventional colleagues. Homeopathic terminology seems to be comprehensible only after practical experience of homeopathy. The main problem lies in different handling of diagnosis. In conventional medicine diagnosis is the starting point for randomised controlled trials to determine the effect of treatment. In homeopathy diagnosis is combined with other symptoms and personal traits of the patient to guide treatment and predict response. Broadening our scope to include diagnostic as well as treatment research opens the possibility of multi factorial reasoning. Adopting Bayesian methodology opens the possibility of investigating homeopathy in everyday practice and of describing some aspects of homeopathy in conventional terms.