- Email: [email protected]
Epidemiology in a Changing World
Epidemiology in a Changing World Jonathan M. Samet, MD, MS, Ross C. Brownson, PhD The times they are a-changin0 . –Bob Dylan
T
he context of epidemiologic research and practice is changing rapidly, driven by a rising wave of new questions based in genomics and biomarkers, new technologies for high-density data generation, a shift toward use of administrative and other databases for research and away from primary data collection by researchers, a rapidly changing communications environment, and new computing and analytic approaches based in the emerging methods of “big data.” Macro-level factors are further transforming epidemiology: generally declining research funding paylines by the NIH and its reluctance to support the development of new cohorts for epidemiologic research, the ongoing evolution of research to “big science” approaches, the rise of clinical and translational research and the increasing reliance in this area on patient databases for research, and mandates within the Affordable Care Act related to monitoring and advancement of population health. There are also rising societal expectations for impactful findings and an expectation of accountability for applying research discoveries for public/community benefit. Although there remains great uncertainty as to future directions and an ongoing need for applied “on-theground” epidemiology, a paradigm shift is truly under way. Publication trends in epidemiology over the last several decades document some of these patterns. Figures 1 and 2 show trends of publications within broad domains from 1990 to 2013. Over this time period, the number of publications identified using “epidemiology” as a search term in PubMed has grown progressively, from 26,490 in 1990 to 95,160 in 2013. Cancer has remained the leading topic, but genetic epidemiology has risen to second (Figure 1). Research in outcomes and effectiveness and on clinical/translational research themes is quickly expanding (Figure 2). From the Department of Preventive Medicine (Samet), Keck School of Medicine and Institute for Global Health, University of Southern California, Los Angeles, California; and the George Warren Brown School of Social Work (Brownson), Division of Public Health Sciences and Alvin J. Siteman Cancer Center (Brownson), School of Medicine, Washington University in St. Louis, St. Louis, Missouri Address correspondence to: Jonathan M. Samet, MD, MS, Department of Preventive Medicine, Keck School of Medicine and Institute for Global Health, University of Southern California, 2001 North Soto Street, Suite 330A, Los Angeles CA 90089-9239. E-mail: [email protected]. 0749-3797/$36.00 http://dx.doi.org/10.1016/j.amepre.2014.07.012
We are concerned that training in epidemiology has changed little, in spite of these shifts in the nature of much of epidemiologic research and practice. Epidemiologists trained today may be still active 30–40 years hence. Some predictions about their future work can be safely made: They will be using data from full-genome sequencing and a wide variety of biomarkers; much of their work will be based around data initially obtained through clinical or other databases; novel sources of big data will continue to emerge and be utilized by epidemiologists; and the conduct of research will be increasingly carried out through large, multidisciplinary teams. Given this future, we are concerned that today’s training does not align well with future directions for application of epidemiology. We have the impression that too many of the current training programs are static, reflect long-standing approaches to education, and are not forward looking. Consequently, the American College of Epidemiology has implemented a project to anticipate future directions for training of epidemiologists. As an organization concerned broadly with advancing the profession of epidemiology, the American College of Epidemiology has had a long-standing interest in training.1 Previous projects, beginning with a 2002 workshop, addressed the competencies that should be gained during master’s- and doctoral-level education.2 This effort also builds on previous efforts from the Council of State and Territorial Epidemiologists (CSTE) and CDC.3 The present project involves directed interviews with leaders in epidemiology from both the academic and practice sectors. We have asked them to forecast trends in the field and the competencies needed by current trainees to be ready for these changes. Following a search of the scientific and gray literature, semistructured interviews have been completed with 16 experienced, active, and visionary epidemiologists from both practice and research settings. On first look, the results indicate a poor match between current curricula in epidemiology and future needs. We anticipate that findings will be published in early 2015. The American College of Epidemiology will use the findings to motivate a thoughtful conversation involving the training programs, researchers, and practitioners. As educators long engaged in training epidemiologists, we have heard from our graduates that there are already gaps
& 2014 American Journal of Preventive Medicine Published by Elsevier Inc. Am J Prev Med 2014;47(5S3):S383–S385 S383 This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).
S384
Samet and Brownson / Am J Prev Med 2014;47(5S3):S383–S385 25000
Applied epidemiology Genetic epidemiology Cancer epidemiology
20000
Cardiovascular epidemiology Environmental epidemiology
15000
10000
5000
0
Figure 1. Number of annual publications in various epidemiology fields indexed in PubMed, 1990–2013. Relative change from 1990 to 2013 by field: applied epidemiology, 548%; genetic epidemiology, 1,376%; cancer epidemiology, 227%; cardiovascular epidemiology, 504%; environmental epidemiology, 271%. Note: Search terms: (1) applied[All Fields] AND (“epidemiology”[Subheading] OR “epidemiology”[All Fields] OR “epidemiology”[MeSH Terms]); (2) molecular epidemiology[MeSH Terms] OR (“molecular”[All Fields] AND “epidemiology”[All Fields]) OR “molecular epidemiology” [All Fields] OR (“genetic”[All Fields] AND “epidemiology”[All Fields]) OR “genetic epidemiology”[All Fields].
between what is taught and the full array of skills needed after completion of graduate training. Former trainees often single out the need for skills in leadership, administration, written and verbal communication, and financial management.
We anticipate further discrepancies between needs and training, as the trends identified by the project’s interviewees accelerate. Above all, the project’s initial findings point to a need for lifelong learning by epidemiologists, a topic that has received too little attention to date and
4500
4000
Outcomes and effectiveness
3500
Clinical translational research 3000 2500
2000 1500 1000 500 0
Figure 2. Number of annual publications in outcomes of effectiveness and clinical translational research indexed in PubMed, 1990–2013. Relative change from 1990 to 2013 for number of publications in outcomes of effectiveness, 4,957%; clinical translational research, 11,332%. Note: Search terms: (1) outcomes[All Fields] AND effectiveness[All Fields]; (2) clinical[All Fields] AND (“translational medical research”[MeSH Terms] OR (“translational”[All Fields] AND “medical”[All Fields] AND “research”[All Fields]) OR “translational medical research”[All Fields] OR (“translational”[All Fields] AND “research”[All Fields]) OR “translational research”[All Fields]).
www.ajpmonline.org
Samet and Brownson / Am J Prev Med 2014;47(5S3):S383–S385
should be addressed by professional organizations and academic institutions, partnered with other key stakeholders (e.g., CDC, CSTE, the Association of Schools and Programs in Public Health). Is the field of epidemiology ready to change and to do so as quickly as is needed? Teaching of introductory epidemiology has roots in descriptive epidemiology, etiologic research, and causal inference.4 More advanced methods are taught quite variably and there are few textbooks beyond the introductory level. Much of the instruction on emerging methods is left to courses in biostatistics, bioinformatics, and computer sciences. The emerging topics identified in our survey have probably yet to enter course curricula and may be the focus of special seminars or lectures. They should also be learned during internship and practical experiences for master’s- and doctoral-level students. Change is likely to be driven by the needs of learners and the job market. Publication of this article was supported by the U.S. Centers for Disease Control and Prevention (CDC), an Agency of the Department of Health and Human Services, under the
November 2014
S385
Cooperative Agreement with the Public Health Foundation and University of Michigan Center of Excellence in Public Health Workforce Studies (CDC RFA-OT13-1302). The ideas expressed in the articles are those of the authors and do not necessarily reflect the official position of CDC. No other financial disclosures were reported by the authors of this paper.
References 1. American College of Epidemiology. Strategic plan. Raleigh NC: American College of Epidemiology, 2010. www.acepidemiology.org/ sites/default/files/ACE%20Strategic%20Plan%20%28draft%20051710% 29.pdf. 2. American College of Epidemiology. Education. acepidemiology.org/ content/education. 3. Birkhead GS, Koo D. Professional competencies for applied epidemiologists: a roadmap to a more effective epidemiologic workforce. J Public Health Manag Pract 2006;12(6):501–4. 4. Keyes KM, Galea S. Epidemiology matters: a new introduction to methodological foundations. Oxford: Oxford University Press, 2014.
T
he context of epidemiologic research and practice is changing rapidly, driven by a rising wave of new questions based in genomics and biomarkers, new technologies for high-density data generation, a shift toward use of administrative and other databases for research and away from primary data collection by researchers, a rapidly changing communications environment, and new computing and analytic approaches based in the emerging methods of “big data.” Macro-level factors are further transforming epidemiology: generally declining research funding paylines by the NIH and its reluctance to support the development of new cohorts for epidemiologic research, the ongoing evolution of research to “big science” approaches, the rise of clinical and translational research and the increasing reliance in this area on patient databases for research, and mandates within the Affordable Care Act related to monitoring and advancement of population health. There are also rising societal expectations for impactful findings and an expectation of accountability for applying research discoveries for public/community benefit. Although there remains great uncertainty as to future directions and an ongoing need for applied “on-theground” epidemiology, a paradigm shift is truly under way. Publication trends in epidemiology over the last several decades document some of these patterns. Figures 1 and 2 show trends of publications within broad domains from 1990 to 2013. Over this time period, the number of publications identified using “epidemiology” as a search term in PubMed has grown progressively, from 26,490 in 1990 to 95,160 in 2013. Cancer has remained the leading topic, but genetic epidemiology has risen to second (Figure 1). Research in outcomes and effectiveness and on clinical/translational research themes is quickly expanding (Figure 2). From the Department of Preventive Medicine (Samet), Keck School of Medicine and Institute for Global Health, University of Southern California, Los Angeles, California; and the George Warren Brown School of Social Work (Brownson), Division of Public Health Sciences and Alvin J. Siteman Cancer Center (Brownson), School of Medicine, Washington University in St. Louis, St. Louis, Missouri Address correspondence to: Jonathan M. Samet, MD, MS, Department of Preventive Medicine, Keck School of Medicine and Institute for Global Health, University of Southern California, 2001 North Soto Street, Suite 330A, Los Angeles CA 90089-9239. E-mail: [email protected]. 0749-3797/$36.00 http://dx.doi.org/10.1016/j.amepre.2014.07.012
We are concerned that training in epidemiology has changed little, in spite of these shifts in the nature of much of epidemiologic research and practice. Epidemiologists trained today may be still active 30–40 years hence. Some predictions about their future work can be safely made: They will be using data from full-genome sequencing and a wide variety of biomarkers; much of their work will be based around data initially obtained through clinical or other databases; novel sources of big data will continue to emerge and be utilized by epidemiologists; and the conduct of research will be increasingly carried out through large, multidisciplinary teams. Given this future, we are concerned that today’s training does not align well with future directions for application of epidemiology. We have the impression that too many of the current training programs are static, reflect long-standing approaches to education, and are not forward looking. Consequently, the American College of Epidemiology has implemented a project to anticipate future directions for training of epidemiologists. As an organization concerned broadly with advancing the profession of epidemiology, the American College of Epidemiology has had a long-standing interest in training.1 Previous projects, beginning with a 2002 workshop, addressed the competencies that should be gained during master’s- and doctoral-level education.2 This effort also builds on previous efforts from the Council of State and Territorial Epidemiologists (CSTE) and CDC.3 The present project involves directed interviews with leaders in epidemiology from both the academic and practice sectors. We have asked them to forecast trends in the field and the competencies needed by current trainees to be ready for these changes. Following a search of the scientific and gray literature, semistructured interviews have been completed with 16 experienced, active, and visionary epidemiologists from both practice and research settings. On first look, the results indicate a poor match between current curricula in epidemiology and future needs. We anticipate that findings will be published in early 2015. The American College of Epidemiology will use the findings to motivate a thoughtful conversation involving the training programs, researchers, and practitioners. As educators long engaged in training epidemiologists, we have heard from our graduates that there are already gaps
& 2014 American Journal of Preventive Medicine Published by Elsevier Inc. Am J Prev Med 2014;47(5S3):S383–S385 S383 This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).
S384
Samet and Brownson / Am J Prev Med 2014;47(5S3):S383–S385 25000
Applied epidemiology Genetic epidemiology Cancer epidemiology
20000
Cardiovascular epidemiology Environmental epidemiology
15000
10000
5000
0
Figure 1. Number of annual publications in various epidemiology fields indexed in PubMed, 1990–2013. Relative change from 1990 to 2013 by field: applied epidemiology, 548%; genetic epidemiology, 1,376%; cancer epidemiology, 227%; cardiovascular epidemiology, 504%; environmental epidemiology, 271%. Note: Search terms: (1) applied[All Fields] AND (“epidemiology”[Subheading] OR “epidemiology”[All Fields] OR “epidemiology”[MeSH Terms]); (2) molecular epidemiology[MeSH Terms] OR (“molecular”[All Fields] AND “epidemiology”[All Fields]) OR “molecular epidemiology” [All Fields] OR (“genetic”[All Fields] AND “epidemiology”[All Fields]) OR “genetic epidemiology”[All Fields].
between what is taught and the full array of skills needed after completion of graduate training. Former trainees often single out the need for skills in leadership, administration, written and verbal communication, and financial management.
We anticipate further discrepancies between needs and training, as the trends identified by the project’s interviewees accelerate. Above all, the project’s initial findings point to a need for lifelong learning by epidemiologists, a topic that has received too little attention to date and
4500
4000
Outcomes and effectiveness
3500
Clinical translational research 3000 2500
2000 1500 1000 500 0
Figure 2. Number of annual publications in outcomes of effectiveness and clinical translational research indexed in PubMed, 1990–2013. Relative change from 1990 to 2013 for number of publications in outcomes of effectiveness, 4,957%; clinical translational research, 11,332%. Note: Search terms: (1) outcomes[All Fields] AND effectiveness[All Fields]; (2) clinical[All Fields] AND (“translational medical research”[MeSH Terms] OR (“translational”[All Fields] AND “medical”[All Fields] AND “research”[All Fields]) OR “translational medical research”[All Fields] OR (“translational”[All Fields] AND “research”[All Fields]) OR “translational research”[All Fields]).
www.ajpmonline.org
Samet and Brownson / Am J Prev Med 2014;47(5S3):S383–S385
should be addressed by professional organizations and academic institutions, partnered with other key stakeholders (e.g., CDC, CSTE, the Association of Schools and Programs in Public Health). Is the field of epidemiology ready to change and to do so as quickly as is needed? Teaching of introductory epidemiology has roots in descriptive epidemiology, etiologic research, and causal inference.4 More advanced methods are taught quite variably and there are few textbooks beyond the introductory level. Much of the instruction on emerging methods is left to courses in biostatistics, bioinformatics, and computer sciences. The emerging topics identified in our survey have probably yet to enter course curricula and may be the focus of special seminars or lectures. They should also be learned during internship and practical experiences for master’s- and doctoral-level students. Change is likely to be driven by the needs of learners and the job market. Publication of this article was supported by the U.S. Centers for Disease Control and Prevention (CDC), an Agency of the Department of Health and Human Services, under the
November 2014
S385
Cooperative Agreement with the Public Health Foundation and University of Michigan Center of Excellence in Public Health Workforce Studies (CDC RFA-OT13-1302). The ideas expressed in the articles are those of the authors and do not necessarily reflect the official position of CDC. No other financial disclosures were reported by the authors of this paper.
References 1. American College of Epidemiology. Strategic plan. Raleigh NC: American College of Epidemiology, 2010. www.acepidemiology.org/ sites/default/files/ACE%20Strategic%20Plan%20%28draft%20051710% 29.pdf. 2. American College of Epidemiology. Education. acepidemiology.org/ content/education. 3. Birkhead GS, Koo D. Professional competencies for applied epidemiologists: a roadmap to a more effective epidemiologic workforce. J Public Health Manag Pract 2006;12(6):501–4. 4. Keyes KM, Galea S. Epidemiology matters: a new introduction to methodological foundations. Oxford: Oxford University Press, 2014.