The Evolution of Epidemiologic Research

Moving Toward Findable, Accessible, Interoperable, Reusable Practices in Epidemiologic Research

Data sharing is essential for reproducibility of epidemiologic research, replication of findings, pooled analyses in consortia efforts, and maximizing study value to address multiple research questions. However, barriers related to confidentiality, costs, and incentives often limit the extent and speed of data sharing. Epidemiological practices that follow Findable, Accessible, Interoperable, Reusable (FAIR) principles can address these barriers by making data resources findable with the necessary metadata, accessible to authorized users, and interoperable with other data, to optimize the reuse of resources with appropriate credit to its creators. We provide an overview of these principles and describe approaches for implementation in epidemiology. Increasing degrees of FAIRness can be achieved by moving data and code from on-site locations to remote, accessible ("Cloud") data servers, using machine-readable and nonproprietary files, and developing open-source code. Adoption of these practices will improve daily work and collaborative analyses and facilitate compliance with data sharing policies from funders and scientific journals. Achieving a high degree of FAIRness will require funding, training, organizational support, recognition, and incentives for sharing research resources, both data and code. However, these costs are outweighed by the benefits of making research more reproducible, impactful, and equitable by facilitating the reuse of precious research resources by the scientific community.

Ovarian cancer epidemiology in the era of collaborative team science

PURPOSE

Over the past decade, a number of consortia have formed to further investigate genetic associations, pathogenesis, and epidemiologic risk and prognostic factors for ovarian cancer. Here, we review the benefits that ovarian cancer consortia provide as well as challenges that have arisen. Methods for managing key challenges are also discussed.

METHODS

We review the structural organization and some of the milestone epidemiologic publications of five consortia dedicated to the study of ovarian cancer, including the Ovarian Cancer Association Consortium (OCAC), the Ovarian Tumor Tissue Analysis (OTTA) Consortium, the Ovarian Cancer Cohort Consortium (OC3), the Collaborative Group on Epidemiological Studies of Ovarian Cancer (The Oxford Collaborative Group), and the Ovarian Cancer in Women of African Ancestry (OCWAA) consortium.

RESULTS

As ovarian cancer is a rare and heterogeneous disease, consortia have made important contributions in the study of risk factors by improving statistical power beyond what any single study, or even a few studies, would provide. Thus, a major accomplishment of consortial research is enhanced characterization of histotype-specific risk factor associations. In addition, consortia have facilitated impressive synergy between researchers across many institutions, spawning new collaborative research. Importantly, through these efforts, many challenges have been met, including difficulties with data harmonization and analysis, laying a road map for future collaborations.

CONCLUSIONS

While ovarian cancer consortia have made valuable contributions to the ovarian cancer epidemiological literature over the past decade, additional efforts comprising of new, well-designed case-control studies are needed to further elucidate novel, histotype-specific risk, and prognostic factors which are not consistently available in existing studies.

Epidemiology: Then and Now

Twenty-five years ago, on the 75th anniversary of the Johns Hopkins Bloomberg School of Public Health, I noted that epidemiologic research was moving away from the traditional approaches used to investigate "epidemics" and their close relationship with preventive medicine. Twenty-five years later, the role of epidemiology as an important contribution to human population research, preventive medicine, and public health is under substantial pressure because of the emphasis on "big data," phenomenology, and personalized medical therapies. Epidemiology is the study of epidemics. The primary role of epidemiology is to identify the epidemics and parameters of interest of host, agent, and environment and to generate and test hypotheses in search of causal pathways. Almost all diseases have a specific distribution in relation to time, place, and person and specific "causes" with high effect sizes. Epidemiology then uses such information to develop interventions and test (through clinical trials and natural experiments) their efficacy and effectiveness. Epidemiology is dependent on new technologies to evaluate improved measurements of host (genomics), epigenetics, identification of agents (metabolomics, proteomics), new technology to evaluate both physical and social environment, and modern methods of data collection. Epidemiology does poorly in studying anything other than epidemics and collections of numerators and denominators without specific hypotheses even with improved statistical methodologies.

Socioeconomic position in childhood and cancer in adulthood: a rapid-review

Background

The relationship of childhood socioeconomic position (SEP) to adult cancer has been inconsistent in the literature and there has been no review summarising the current evidence focused solely on cancer outcomes.

Methods and results

We performed a rapid review of the literature, which identified 22 publications from 13 studies, primarily in the UK and northern European countries that specifically analysed individual measures of SEP in childhood and cancer outcomes in adulthood. Most of these studies adjusted for adult SEP as a critical mediator of the relationship of interest.

Conclusions

Results confirm that childhood socioeconomic circumstances have a strong influence on stomach cancer and are likely to contribute, along with adult circumstances, to lung cancer through cumulative exposure to smoking. There was also some evidence of increased risk of colorectal, liver, cervical and pancreatic cancers with lower childhood SEP in large studies, but small numbers of cancer deaths made these estimates imprecise. Gaps in knowledge and potential policy implications are presented.

The next generation of large-scale epidemiologic research: implications for training cancer epidemiologists

There is expanding consensus on the need to modernize the training of cancer epidemiologists to accommodate rapidly emerging technological advancements and the digital age, which are transforming the practice of cancer epidemiology. There is also a growing imperative to extend cancer epidemiology research that is etiological to that which is applied and has the potential to affect individual and public health. Medical schools and schools of public health are recognizing the need to develop such integrated programs; however, we lack the data to estimate how many current training programs are effectively equipping epidemiology students with the knowledge and tools to design, conduct, and analyze these increasingly complex studies. There is also a need to develop new mentoring approaches to account for the transdisciplinary team-science environment that now prevails. With increased dialogue among schools of public health, medical schools, and cancer centers, revised competencies and training programs at predoctoral, doctoral, and postdoctoral levels must be developed. Continuous collection of data on the impact and outcomes of such programs is also recommended.

From administrative infrastructure to biomedical resource: Danish population registries, the "Scandinavian laboratory," and the "epidemiologist's dream"

Since the 1970s, Danish population registries were increasingly used for research purposes, in particular in the health sciences. Linked with a large number of disease registries, these data infrastructures became laboratories for the development of both information technology and epidemiological studies. Denmark's system of population registries had been centralized in 1924 and was further automated in the 1960s, with individual identification numbers (CPR-numbers) introduced in 1968. The ubiquitous presence of CPR-numbers in administrative routines and everyday lives created a continually growing data archive of the entire population. The resulting national-level database made possible unprecedented record linkage, a feature epidemiologists and biomedical scientists used as a resource for population health research. The specific assemblages that emerged with their practices of data mining were constitutive of registry-based epidemiology as a style of thought and of a distinct relationship between science, citizens, and the state that emerged as “Scandinavian.”

Collaborative cancer epidemiology in the 21st century: the model of cancer consortia

During the last two decades, epidemiology has undergone a rapid evolution toward collaborative research. The proliferation of multi-institutional, interdisciplinary consortia has acquired particular prominence in cancer research. Herein, we describe the characteristics of a network of 49 established cancer epidemiology consortia (CEC) currently supported by the Epidemiology and Genomics Research Program (EGRP) at the National Cancer Institute (NCI). This collection represents the largest disease-based research network for collaborative cancer research established in population sciences. We describe the funding trends, geographic distribution, and areas of research focus. The CEC have been partially supported by 201 grants and yielded 3,876 publications between 1995 and 2011. We describe this output in terms of interdisciplinary collaboration and translational evolution. We discuss challenges and future opportunities in the establishment and conduct of large-scale team science within the framework of CEC, review future prospects for this approach to large-scale, interdisciplinary cancer research, and describe a model for the evolution of an integrated Network of Cancer Consortia optimally suited to address and support 21st-century epidemiology.

Bigger, better, sooner--scaling up for success

Over the last twenty years, the field of epidemiology has seen a rapidly increasing interest in, and need for, addressing low-level risks, interactions as well as main effects, and simultaneous assessment of vast numbers of biomarkers. Multiple examples over this time have shown the necessity for very large, high-quality individual studies (e.g., biobanks) or consortia of studies for these efforts to be successful. The need for this will continue to increase in the foreseeable future. It will also be important to analyze and publish aggregated data much earlier in the discovery process than typical for past efforts.

Cohorts and consortia conference: a summary report (Banff, Canada, June 17-19, 2009)

Epidemiologic studies have adapted to the genomics era by forming large international consortia to overcome issues of large data volume and small sample size. Whereas both cohort and well-conducted case-control studies can inform disease risk from genetic susceptibility, cohort studies offer the additional advantages of assessing lifestyle and environmental exposure-disease time sequences often over a life course. Consortium involvement poses several logistical and ethical issues to investigators, some of which are unique to cohort studies, including the challenge to harmonize prospectively collected lifestyle and environmental exposures validly across individual studies. An open forum to discuss the opportunities and challenges of large-scale cohorts and their consortia was held in June 2009 in Banff, Canada, and is summarized in this report.

Bioinformatics: Tools to accelerate population science and disease control research

Population science and disease control researchers can benefit from a more proactive approach to applying bioinformatics tools for clinical and public health research. Bioinformatics utilizes principles of information sciences and technologies to transform vast, diverse, and complex life sciences data into a more coherent format for wider application. Bioinformatics provides the means to collect and process data, enhance data standardization and harmonization for scientific discovery, and merge disparate data sources. Achieving interoperability (i.e. the development of an informatics system that provides access to and use of data from different systems) will facilitate scientific explorations and careers and opportunities for interventions in population health. The National Cancer Institute's (NCI's) interoperable Cancer Biomedical Informatics Grid (caBIG) is one of a number of illustrative tools in this report that are being mined by population scientists. Tools are not all that is needed for progress. Challenges persist, including a lack of common data standards, proprietary barriers to data access, and difficulties pooling data from studies. Population scientists and informaticists are developing promising and innovative solutions to these barriers. The purpose of this paper is to describe how the application of bioinformatics systems can accelerate population health research across the continuum from prevention to detection, diagnosis, treatment, and outcome.

Effectiveness of strategies to increase the validity of findings from association studies: size vs. replication

Background

The capacity of multiple comparisons to produce false positive findings in genetic association studies is abundantly clear. To address this issue, the concept of false positive report probability (FPRP) measures "the probability of no true association between a genetic variant and disease given a statistically significant finding". This concept involves the notion of prior probability of an association between a genetic variant and a disease, making it difficult to achieve acceptable levels for the FPRP when the prior probability is low. Increasing the sample size is of limited efficiency to improve the situation.

Methods

To further clarify this problem, the concept of true report probability (TRP) is introduced by analogy to the positive predictive value (PPV) of diagnostic testing. The approach is extended to consider the effects of replication studies. The formula for the TRP after k replication studies is mathematically derived and shown to be only dependent on prior probability, alpha, power, and number of replication studies.

Results

Case-control association studies are used to illustrate the TRP concept for replication strategies. Based on power considerations, a relationship is derived between TRP after k replication studies and sample size of each individual study. That relationship enables study designers optimization of study plans. Further, it is demonstrated that replication is efficient in increasing the TRP even in the case of low prior probability of an association and without requiring very large sample sizes for each individual study.

Conclusions

True report probability is a comprehensive and straightforward concept for assessing the validity of positive statistical testing results in association studies. By its extension to replication strategies it can be demonstrated in a transparent manner that replication is highly effective in distinguishing spurious from true associations. Based on the generalized TRP method for replication designs, optimal research strategy and sample size planning become possible.

The continuum of translation research in genomic medicine: how can we accelerate the appropriate integration of human genome discoveries into health care and disease prevention?

Advances in genomics have led to mounting expectations in regard to their impact on health care and disease prevention. In light of this fact, a comprehensive research agenda is needed to move human genome discoveries into health practice in a way that maximizes health benefits and minimizes harm to individuals and populations. We present a framework for the continuum of multidisciplinary translation research that builds on previous characterization efforts in genomics and other areas in health care and prevention. The continuum includes four phases of translation research that revolve around the development of evidence-based guidelines. Phase 1 translation (T1) research seeks to move a basic genome-based discovery into a candidate health application (e.g., genetic test/intervention). Phase 2 translation (T2) research assesses the value of a genomic application for health practice leading to the development of evidence-based guidelines. Phase 3 translation (T3) research attempts to move evidence-based guidelines into health practice, through delivery, dissemination, and diffusion research. Phase 4 translation (T4) research seeks to evaluate the "real world" health outcomes of a genomic application in practice. Because the development of evidence-based guidelines is a moving target, the types of translation research can overlap and provide feedback loops to allow integration of new knowledge. Although it is difficult to quantify how much of genomics research is T1, we estimate that no more than 3% of published research focuses on T2 and beyond. Indeed, evidence-based guidelines and T3 and T4 research currently are rare. With continued advances in genomic applications, however, the full continuum of translation research needs adequate support to realize the promise of genomics for human health.