Since its inception in 1990, Epidemiology has published a substantial number of articles that are primarily statistical (or, more broadly, methodologic), rather than substantive or specific to a subject area. These methodologic articles, which often appeal to the broad readership in a way that substantive papers do not, have helped to establish Epidemiology as an interesting and informative journal.
Allen Wilcox and the Editors of Epidemiologyhave asked me to join them specifically to handle statistical and methodologic submissions to the journal. It may be useful to prospective authors to know what types of methodologic papers are likely to fare well. I am looking for pieces that lead to higher-quality science (broadly defined) in epidemiology. More specifically, I would like to see methods that prove useful in subsequent studies. One crude measure of success might be how often methods pieces published in Epidemiologyinfluence epidemiology papers published thereafter.
Several areas call out for new ideas and methods:
- How do we use new designs and other techniques to reduce cost or increase information yield? Some examples might be two-phase designs or pooling of biospecimen samples.
- Are there better and more practical ways to assess and improve fieldwork or quality control, and perhaps even to incorporate results from these into measures of uncertainty?
- How do we use the existing knowledge base to determine the most promising new substantive questions to study or to decide whether to abandon a research area that has not yielded much insight to date?
Furthermore, there are venerable questions that need to be reconsidered in light of social changes, molecular methods or deepening understanding of key concepts or new statistical tools. For example:
- How should methods of disease ascertainment change in response to advances in molecular diagnostics or to increased population mobility?
- How can we increase the usefulness of vital statistics databases?
- How do we choose controls in an era of caller ID and answering machines and fears that genotyping for research might lead to loss of health insurance?
- Can we address bias from unmeasured confounders, selection factors or determinants of loss to follow-up?
- How do we interpret the results from one study, or summarize the results from many?
Moreover, I believe that a methods paper appropriate for publication in Epidemiology should be accessible to its readers, few of whom are statisticians. Successful communication with working scientists, including well-documented and user-friendly software, also serves the interest of authors who want to see their ideas used. I encourage articles that transform seemingly arcane statistical ideas (whether well established or cutting edge) into practical epidemiologic tools, using language that the readers of Epidemiology will understand. These will not all be original biostatistical research papers, though those are welcome. In that spirit, I would welcome papers aimed primarily at students and readers who might feel out of date on important new ideas.
To obtain an outside view of the utility or accessibility of a paper, I plan to send statistical manuscripts not only to referees who are themselves methodologists, but also to researchers who are the target audience for ideas developed in the paper. Such referees can help the Editors evaluate whether the pragmatic mission of the journal is served by publication of the paper. These are the kinds of comments I will find useful when I ask you for a review:
- “I do not have a clue what the author is saying.”
- “It might be relevant and important, but I cannot really tell.”
- “The effort required to implement the proposed method is far greater than the impact it is likely to have on the results.”
- “Excellent idea, but expressed in overly esoteric language.”
- “This paper makes clear how the technique works. It no longer seems like a black box.”
- “This technique solves a problem that has been bothering me. I now know how to learn more from the data in my study.”
Allen Wilcox once wrote that an editor’s job is not only to reflect the field, but to shape it. 1 The readers of this journal know that applying improved methods can strengthen design, fieldwork, analysis and interpretation of epidemiologic studies. Epidemiology will continue to provide a forum for presenting important work on epidemiologic methods. With your help—as contributors, reviewers and readers—we hope to point the way to further improvements in epidemiologic research.
1. Wilcox AJ. Epidemiology: Chapter two [editorial]. Epidemiology 2001; 12: 285–286.
A Call for Nominations: The 2003 Rothman Epidemiology Prize
Epidemiology presents an annual award for the best paper published by the journal during the previous year.
This prize of $3,000 and a plaque goes to the author whose paper is selected by the Editors and the Editorial Board for its originality, importance, clarity of thought, and excellence in writing.
With this issue, we close our 2002 volume. We invite our readers to nominate papers published during the past year.
Please submit your nominations to: Epidemiology The Snow Building 331 West Main Street, Suite 606 Durham, North Carolina 27701 Nominations must be received no later than 1 December 2002.
The winner will be announced in the May 2003 issue. This award is made possible by an endowment from Hoffman-LaRoche Ltd., managed by the International Society for Pharmacoepidemiology (ISPE).