Official Journal of the American College of Emergency Physicians



Annals of Emergency Medicine, the official journal of the American College of Emergency Physicians, is an international, peer-reviewed journal dedicated to improving the quality of care by publishing the highest quality science for emergency medicine and related medical specialties. Annals publishes original research, clinical reports, opinion, and educational information related to the practice, teaching, and environment of emergency medicine. In addition to general topics, Annals regularly publishes sections on out-of-hospital emergency medical services, pediatric emergency medicine, injury and disease prevention, health policy and ethics, disaster management, toxicology, and related topics. The journal welcomes submissions from international contributors and researchers of all specialties. Although most of our published research is clinical, we are also interested in basic science research relevant to emergency medicine.

Annals is the largest circulation peer review journal in emergency medicine (26,800, about four times its nearest competitor). It is also one of the most accessible to non-subscribing readers, since more than 2,100 medical school and hospital libraries subscribe to it in print, and another 3,300 institutions receive it electronically. Annals received over 133,000 reprint requests in 2005, and is available on the Web (with full text of all articles dating back to its inception), where it receives over 1.8 million page views a year (about 270,000 of those by subscribers or ACEP member physicians). It is also available via ScienceDirect (the world's largest electronic collection of science, technology and medicine full text and bibliographic information), which was utilized for access to the full text of Annals articles approximately 177,000 times by readers in 50 countries in 2005. Users of PubMed used direct linking to Annals full text articles approximately 72,000 times last year.

Annals' contents are also the subject of considerable interest by the lay media. Press releases are issued each month for key articles and about 950 media stories citing Annals' articles were published in the most recent year. These include stories in The New York Times, The Washington Post, The Los Angeles Times, U.S. News and World Report, and Associated Press and United Press International.

In 2009 Annals was chosen one of the 100 most influential scientific journals of the past 100 years by the Special Libraries Association (www.sla.org). The Special Libraries Association is one of the most respected and largest (11,000 members) library organizations. The entire list is at ( http://www.sla.org/content/Events/centennial/dbio100.cfm ). Some of the high profile medical journals on the list were Cell, Circulation, JAMA, The Lancet, Nature, NEJM, and Science. Annals is flattered to have received this recognition, which is testimony to the hard work, talent, and dedication of its editorial board, its staff, and all the authors who contribute to it.



These Instructions for Authors are divided into 4 different but equally important sections. Section I describes our overall philosophy and expectations regarding how original science should be conducted and reported. Section II describes the types of submissions that the journal accepts. Section III contains specific technical and formatting instructions to help authors prepare their manuscripts for submission with appropriate font, page margins, and so on. Section IV explains what you may expect from our review process.



We understand that each journal has its own requirements and that there is little uniformity among journals. Our requirements reflect the preferences of our editors and readers, but they also are tailored to reflect what is known from research about best publication practices and the clearest communication of information. Most of these instructions should be familiar to you and not unique. Those that do not fit this description were not chosen arbitrarily, but instead represent the direction toward which we believe scientific publishing is evolving. We do not expect every manuscript to comply in every regard, but the more consistent a manuscript is with these guidelines, the more likely is publication.

Style and Content

General: We seek forthright, detailed reports of scientific investigations; review and educational articles; and scientific, ethical, social, political, and economic commentary on topics of importance to emergency medicine. We value reports of original science that accurately and clearly describe what was done and why it was done. Much of the medical literature is written as if studies were perfectly conducted, but we know this is not possible. We fully expect that some part of every clinical study will deviate from the ideal. The candid disclosure of such deviations and the reasons they occurred is encouraged because it enhances the scientific process.

Writing Style: A well-written paper is more likely to be accepted for publication, and subsequently read and cited by others. We prefer a straightforward, unpretentious style whose chief purpose is to efficiently convey information. Use the active voice. Sentences should be simple and short. Never use a lengthy scientific term when a clear simpler one is available. In general, brevity conveys more genuine information than loquacity, and leads you and the reader to think more carefully about your message. The British Medical Journal is a good example of concise and effective writing that communicates a good deal of information with a modest number of words. We discourage the use of any but the most necessary of abbreviations; they may be a convenience for an author but are generally an impediment to easy comprehension for the reader. Most papers should have few or none of them. We particularly discourage the use of newly coined (and quickly forgotten) abbreviations to describe simple terms that most people say in English. Examples include BU for bedside ultrasound, UD for usual dose, CorrCrCl for corrected creatinine clearance, PEP for pediatric emergency physicians, ACE for adverse cardiac events, and VCPRCE for very confusing patients requiring a comprehensive evaluation (we invented none of these abbreviations except the last, but ones like it are occasionally also used). We appreciate the desire to save trees, but the need is not that great. For grammar, style, and punctuation, Annals uses the American Medical Association's Manual of Style for editorial style.[2]

Word count limits for each type of submission are described later. Although we do not specify limits for each section of a paper, for original research papers, we strongly suggest that the number of pages devoted to the Introduction and Discussion sections not exceed those devoted to Methods and Results sections.

Organizing Reports of Original Research

Guidelines for Different Study Designs: Original Research and Brief Research Reports should reflect your familiarity with well-established, generally accepted structural guidelines specific to your type of project. We encourage creativity and originality and recognize that not all papers can or should meet these guidelines. You will increase the likelihood of acceptance, however, by identifying and justifying deviations from guideline recommendations in the paper or in your cover letter.

Some examples of types of studies for which there are generally accepted guidelines, with references to the guideline documents, include:

• Randomized controlled trials[3,4]
- Consolidation of Standards for Reporting Trials (CONSORT)

• Studies on diagnostic tests[5-7]
- STARD

• Clinical prediction rules[8,9]

• Economic evaluations and cost-effectiveness analyses[10-13]

• Meta-analyses of trials of therapy[14]
- QUOROM

• Meta-analyses of diagnostic tests[15,16]

• Meta-analyses of observational studies[17]
- MOOSE

• Qualitative research[18-21]

• Chart reviews[22]
- Chart review methods

• Surveys[23-26]

Guidance for Specific Sections of Reports on Original Research

Abstract: Your abstract will be available online and will be read far more often than the entire paper. The abstract should be terse yet clear, accurate, and complete. Divide your 250-word abstract into the subheadings: Study hypothesis or Objective, Methods (include information on design, setting, participants, interventions, and main outcomes measured; it is not mandatory to include the subject headings), Results, and Conclusions. Include the key numerical results, but keep the amount of numerical reporting consistent with readability. Do not draw conclusions stronger or more expansive than those in the body of the paper. Take care to include all important study limitations and caveats.

Introduction: The introduction to most papers should be less than 1.5 double-spaced manuscript pages (about 450 words); certainly no more than 2 pages. A 3-paragraph structure works well to convince the reader that your topic is new, scientifically important, and clinically relevant. In the first paragraph, under the subheading Background , succinctly describe the circumstances that set the stage for your investigation. Explain the historical context that led you to investigate the issue. Under Importance , describe why your investigation is consequential. What are its potential implications? How does it relate to issues raised in the first paragraph? Why is this specific investigation the next logical step? Conclude with a third paragraph, Goals of This Investigation , in which you state the specific research objective in a detailed manner. Include your primary outcome measure (eg, "We considered a 1-hour median decrease in length of stay important...) and the desired precision of the measurement (...and wished to enroll sufficient subjects that we could be 95% certain that our estimate was within 20 minutes of the true value.")

Methods: Readers will use your Methods section to determine the validity of your study. Provide enough detail so that a knowledgeable reader could, in principle, replicate all aspects of your study. A statement of institutional review board (IRB) approval or exemption from full review is required.

The Methods section should be organized in a logical and sequential order. Help readers by using the following subheadings to divide the Methods into meaningful sections:

Theoretical model of the problem*
Study design*
Setting*
Selection of participants*
Interventions
Methods of measurement*
Data collection and processing
Outcome measures
Primary data analysis*
Sensitivity analyses

*These subheadings should be included in almost every Original Research paper.

Authors may note that our preferences regarding analytic methods and presentation of results differ somewhat from other journals. Rest assured that we do not do this to be idiosyncratic or to create annoying roadblocks on the way to publication. Our philosophy is summarized in the editorial[27] that introduced this version of the instructions and is supported by many of the cited references. It represents our attempt to synthesize best practices regarding the conduct and presentation of clinical research. The instructions can be summarized as: show your data at the level of the unit of analysis (using graphics), report estimates of the size of effects (and your confidence in your estimates) instead of the statistical significance of effects, and account for bias when making claims about your results. Because there is no proven best way to do science, we have no absolute rules. Nevertheless, by reading and complying with what follows and having well thought-out reasons when you deviate, you will maximize your chances of getting your work published.

Begin with an explanation of the theoretical model underlying the investigation. Provide a broad overview of the study design using standard terms. Describe the setting , method for selecting participants , study protocol (including any interventions ), methods of measurement , and methods for data collection and processing . Identify your primary and secondary outcome measures . We prefer patient-centered outcomes (eg, pain, mood, mortality, days lost from work or school, quality of life) to intermediate outcomes (eg, change in FEV1, number of defibrillations), and previously validated measures to newly invented ones.

Describe the analytic plan in enough detail that a statistically sophisticated reader with access to the original data could replicate the results. Justify any data manipulations (eg, combining categories, breaking continuous responses into discrete ranges), and other adjustment techniques. Describe the rationale for the analytic strategy for each of the research questions or hypotheses instead of simply listing statistical procedures. We encourage authors to specifically and explicitly describe the assumptions and judgments made in executing their analytic strategy. We also encourage authors to recognize that, when done properly, detailed graphical presentation of the results is a complete analytic method that does not require additional statistical modeling to enhance its validity. Inform the reader of how results will be presented. Document the software used for data management and analysis. Anticipate the likely biases to your study and incorporate sensitivity analyses exploring how these biases might affect results into your design and analytic plan.[28,29]

If you find that providing this level of detail produces a Methods (or Results) section that is too long, or too complex for the typical reader, consider presenting the details in an appendix. This can be submitted with the manuscript so that the reviewers have access to all of the details. If the paper is accepted, the appendix can be included on Annals' Web site instead of in the print journal.

Results: Present the results in a logical, sequential order that parallels the organization of the Methods section. Account for all subjects, beginning with the number of subjects who could have participated in the study. Present as much data as possible at the level of the unit of analysis. Annals' preferences for reporting results, from most preferred to least are: graphical depictions of data; summaries of data (ie, means, medians, ranges); confidence intervals; point estimates; P values; and other measures of statistical significance.[30] For example, in a study with 2 groups and a continuous outcome measure, a graph showing the distribution for each group would be best; measures of central tendency and dispersion for each group next best; the sentence "the 95% confidence limits for the difference in means was ____" acceptable; and the statement, "The difference in means was significant," should be avoided. Use tables and figures to empower readers to reach their own conclusions about your work. When describing the dispersion of the data, present standard deviation, not standard error of the mean.

Emphasize the estimation of the size of effects over the determination of whether effects are statistically significant.[30-38] When possible, avoid statistical hypothesis testing. For more information on these issues see the editorial that accompanied the introduction of this version of the Instructions to Authors. At minimum, restrict estimation and testing procedures to the a priori hypotheses of interest. Statistics, whether descriptive or hypothesis testing, should not be a substitute for the presentation of data. Do not perform multiple statistical tests or adjustments in an exploratory manner to discover "significant" P values. When calculating confidence intervals, or other statistics, consider using methods that incorporate uncertainty regarding the validity of assumptions implied by classical statistical techniques.[39-43]

Do not repeat data presented in tables and figures in the text. Use the text to highlight the most important aspects of the figures and tables and to convey unique information. Round numerical results to a level of precision appropriate for the study (eg, the percent response in a study group with 80 subjects should be reported as 35%, not 35.6%). For specific guidance based on study design and analytic strategy, consider using Lang's guidelines.[44]

When using statistical models, do not restrict your analysis to the "best case" scenario. Include sensitivity analyses that explore how results change when the assumptions of the model are altered.[28,29]

You may use the following subheadings in the Results section:

Characteristics of study subjects
Main results
Sensitivity analyses

Tables and Figures: Make all tables and figures self-explanatory. Graphics should be used to convey patterns and details that cannot be succinctly conveyed in tables or text. When appropriate, include potentially important covariates in the tables and figures. We prefer graphics that show the distribution of data (eg, scatterplots, 1-way plots, box plots) to those showing summaries of data (eg, pie charts, bar graphs of means). If the data collected are paired (eg, pre and post, or 2 different measures on the same subject), then choose a graphical format that conveys the inherent pairing of the data. Avoid background gridlines and other formatting that do not convey information (eg, superfluous use of 3-dimensional formatting, background shadings).[45-49]

Arrange tables so that the primary comparisons of interest are horizontal, left-to-right (the standard reading order). Provide the N for each column or row and marginal totals where appropriate.

The following formats are preferred for submission of digital art: TIFF, EPS, MS Word, Powerpoint, and Excel. All illustrations should be provided as separate files and as hard copy on separate sheets. TIFF (Tagged Image File Format) is the recommended file format for bitmap, greyscale, and color images. TIFF supports several good compression schemes, ensuring file sizes are kept to a minimum to aid easy file transfer. When supplying TIFF files, please ensure that files are supplied at the correct resolution: line artwork, minimum of 1,000 dpi; halftone artwork, minimum of 300 dpi; combination artwork (line/tone), minimum of 500 dpi. For detailed technical requirements, refer to the Author Gateway.

Photographs: Please submit glossy photographs. We request a print for color verification. The following format is preferred for submission of photographs: TIFF. We accept electronic data from authors on the following media types: 3½" floppy disk, CD, Zip disk, and Jaz disk. All photographs should be provided as separate files and as hard copy on separate sheets. All photographs should be numbered according to sequence in the text. Captions should be provided on a separate sheet along with the manuscript.
The owner and/or subject of the photograph must sign the consent form.

Limitations: Explicitly discuss the limitations of your study. Describe the limitations in the context of the theoretical model of your research. You can lessen the need for a lengthy limitations section by choosing analytic strategies that account for potential biases. Consider threats to the internal and external validity of your results. Do not simply list potential limitations but examine the magnitude and direction of each bias and how it might affect the interpretation of results. Discuss the implications of any sensitivity analyses.

Discussion: Briefly summarize the results and how they relate to your area of investigation. Do not attempt a literature review. Consider only those published articles directly relevant to interpreting your results and placing them in context. Do not stress statistical significance over clinical importance. Avoid extrapolation to persons or conditions that you have not explicitly studied in your investigation. Avoid claims about cost or economic benefit unless a formal cost-effectiveness analysis was presented in the Methods and Results sections. Do not suggest "more research is needed" without stating what the specific next step is. Under the subheading, "In Retrospect," candidly discuss what you would do differently if given the opportunity to repeat the study, so others can learn from your experience. Conclude this section with a brief summary statement. Take care that the conclusion is restricted to that which can be justified by your experimental results.

You may use the following subheading in the Discussion section:

-In retrospect

Appendixes and Web Files: You may submit appendixes that describe either methods or results in more detail than will fit within our specified word limits if these are needed for clarity of understanding by either peer reviewers or readers. If appropriate, materials suitable for Web publication but not print publication (eg, audio or video files) can also be submitted. If you do so, indicate the particular reasons for the appendix and whether you are submitting it for possible Web publication or simply for peer review purposes.



Annals publishes the following types of articles:

Original Research

Original studies of basic or clinical investigations in areas relevant to emergency medicine. References and a structured abstract (see Formatting and Submitting Your Manuscript) are required. Maximum length: 4,000 words, 7 tables and/or figures, plus the abstract and references.

Brief Research Reports

Reports of preliminary data and findings or studies with small numbers demonstrating the need for further investigation. References and a structured abstract (see Formatting and Submitting Your Manuscript) are required. Maximum length: 2,000 words, plus the abstract and no more than 10 references and 3 tables and/or figures.

Concepts

Descriptions of clinical and nonclinical problems and solutions; descriptions of novel approaches to planning, management, or provision of emergency services; and practical "how-to" articles describing aspects of emergency medicine management. A narrative abstract is required. Maximum length: 4,000 words, plus the abstract (no more than 250 words) and references.

Evidence-Based Emergency Medicine

Review Articles

Extensive reviews of the literature on a narrow clinical topic. References must include, but need not be limited to, the past 3 years of the literature. A narrative abstract is required. Do not combine a case report with a literature review. Maximum length: 5,000 words, plus the narrative abstract (no more than 250 words) and references.

Case Conferences

Presentation and discussion of a case by an expert, focused on the problem-solving approach toward a particular clinical problem and discussion of differential diagnoses and subsequent management at various stages of the patient's evaluation. Maximum length: 5,000 words.

Case Reports

Brief descriptions of a previously undocumented disease process, a unique unreported manifestation or treatment of a known disease process, or unique unreported complications of treatment regimens. Entities previously reported in the emergency medicine literature will not be considered, and those reported in other specialty literature or in the foreign literature must be extremely important or pertinent to be considered. Case reports should contain an abstract, introduction, narrative, and a discussion focusing on the implications of the case reported. They should not contain a review of the literature. Maximum length: 1,500 words, no more than 15 references, and 1 table or figure.

Editorials

Authoritative comments or opinions on major current problems of emergency physicians or on controversial matters with significant implications for emergency medicine; or, qualified, thorough analysis and criticism of articles appearing in Annals. Maximum length: 1,500 words plus references.

Brief Commentaries

Brief discussion focusing on 1 or 2 key points about a single study--strengths, weaknesses, where it fits in the context of other studies, controversies, how it should or should not change our clinical practice, or how it illustrates some important principle of science or methodology. Usually written by editors or reviewers involved in the evaluation of a submitted manuscript, and published concurrently with that manuscript. Maximum length: 750 words plus references.

Correspondence

Discussion, observations, opinions, corrections, and comments on topics appearing in Annals or elsewhere; very brief reports or other items of interest. Maximum length: 500 words, plus no more than 5 references. Letters discussing an Annals' article should be received within 6 weeks of the article's publication. Authors of articles about which letters are received will be given the opportunity to reply, which will not be shared with the letter writer prior to publication. Letters of political or other topics unrelated to the science of medicine, as well as those containing personal criticisms, will not be published.



Manuscript Submission

Annals uses a Web-based peer review system, Editorial Manager^™ , to receive all submissions and no longer accepts submissions by mail. Our Web-based system provides full electronic capabilities not only for submission, but also for peer review and status updates. It also speeds manuscript turnaround and provides global access for authors, reviewers, and editors. Authors, reviewers, and editors will receive automatic e-mail messages from Editorial Manager when significant events occur. Detailed instructions and a help file are provided at the Web site.
The submission requirements of Annals of Emergency Medicine are in accordance with the "Uniform Requirements for Manuscripts Submitted to Biomedical Journals" (with the exception of our authorship requirements) and the "Declaration of Helsinki: Recommendations guiding physicians in biomedical research involving human subjects".[50] Annals uses the American Medical Association's Manual of Style for editorial style.[2]

Prior Publication: Annals publishes only original work. Manuscripts must not have been published or submitted for publication elsewhere, in whole or in part, before submission to Annals. Should there be any doubt about prior or duplicate publication of a manuscript being submitted to Annals, describe the circumstances in detail in the cover letter. If you are unsure whether some other communication on this topic might be considered duplicate publication, describe the circumstances to us in the cover letter. Failure to do so could be interpreted as deliberate duplicate publication. This restriction does not apply to abstracts or brief press reports routinely published in connection with scientific meetings, which reveal little or nothing beyond that presented in the published meeting abstract.

Author Responsibilities: The authors are responsible for creating all components of the manuscript. If writers are provided by a sponsoring or funding institution or corporation to draft or revise the article, the name of the writer and their sponsoring organization must be provided. Their names and contributions will be provided with the acknowledgments. We do not recognize "honorary" authorship (when authorship is granted as a favor to someone powerful or prestigious who would not have qualified for it otherwise). All authors must take responsibility in writing for the accuracy of the manuscript, and one author must be the guarantor and take responsibility for the work as a whole.Annals' requirements for authorship are described in more detail in a previously published article.[51] Changes in authorship after the initial submission require written agreement by all authors.

Statistical Consultant: All manuscripts that contain statistics must have been prepared in consultation with an individual who has expertise in the field. One of the authors or an outside consultant may perform this analysis. The individual responsible for statistical analysis must be specifically identified in the cover letter, and may be listed as an author or in the acknowledgments.

Conflict of Interest: Potential conflicts of interests by authors (and others) are increasingly a matter of public concern and are best dealt with by full disclosure of relationships that might constitute a conflict of interest. Potential conflicts of interest must be identified in the cover letter, on the Manuscript Submission Agreement, and on the title page. The title page should list all sources of outside funding or support of any kind for the work, or any financial interest of the authors in the product studied or the company that produces it.

In the Manuscript Submission Agreement, cite all affiliations or involvement that might be perceived as conflicts of interest. These include but are not limited to honoraria; education grants; speakers' bureaus; membership, employment, consultancies, stock ownership, or other equity interest; and expert testimony or patent-licensing arrangements with any organization or entity with any financial interest in the subject matter or materials used in the research. If there are no such potential conflicts, the authors must declare this. If the manuscript is accepted,Annals will disclose the relevant information with the article. Annals has a detailed policy[52]on this and other ethical issues.

Other Conflicts of Interest: Potential conflicts of interest that might arise on the part of journal reviewers or editors are discussed separately.[52]

Permission: Any submitted material (including figures and tables) that has been published elsewhere must be accompanied by the written consent of the author and publisher for reproduction in Annals. All photographs, especially photographs of subjects showing any recognizable features, must be accompanied by their signed release authorizing publication, as must materials that contain personal information with respect to any individual. The approved consent form is available here. Provide a statement of clearance to publish, if this is an institutional requirement. Obtain written permission from any persons acknowledged by name. Articles appear in both the print and online versions of the journal, and the wording of the permission letter should specify permission in all forms and media. Failure to obtain electronic permission rights may result in the images not appearing in the online version.

Access to Data: During the peer review process, Annals may require access to the authors' original data for the sole purpose of better reviewing the submission. Annals reserves the right to review the complete primary data on which manuscripts are based, either before publication or, should questions arise, afterward. Annals will keep the data confidential and not otherwise directly benefit from the advance access to the data or the in-depth examination of them that stems from the peer review process.

Required Submission Documents

When submitting your manuscript to Annals via our Web-based peer review system, Editorial Manager, each type of submission has its unique items for submission. The following documents are required for most types of manuscripts submitted to Annals and should be saved as separate electronic files for uploading to the Web system (Note: only the Manuscript Submission Agreement can be submitted offline):

• Cover Letter
• Manuscript Submission Agreement
• Author Contributions Statement (for Original Research and Brief Research Reports only)
• Title Page
• Abstract, Article, References

We strongly suggest you keep copies of all submission documents in the event of any problem.

Cover Letter: The cover letter should identify and briefly describe the manuscript. In addition, it should:

• list the title of the article
• identify the journal category for which your manuscript is intended
• identify the corresponding author
• indicate whether it is a randomized controlled trial or other standardized study type described below
• provide full information about any form of prior publication (see "Prior Publication" above)
• describe any situation that might be perceived as a conflict of interest
• list any copyright constraints

Save the cover letter as a separate electronic file for uploading to Editorial Manager.

Manuscript Submission Agreement: A Manuscript Submission Agreement is printed in every issue of the journal and is available here in PDF form. The completed and signed document should be faxed and mailed to the editorial office when the manuscript is submitted. All authors' signatures are required before the editor's final decision is sent to the corresponding author. If the Manuscript Submission Agreement is not on file in the Annals' office when a revision is received, the manuscript will not be processed until it arrives.

The sections on IRB/Informed Consent, Conflict of Interest, and Statistical Consultant should be especially noted. Any subsequent changes to the authorship status of individuals listed on this document will require written consent from those authors themselves.

Author Contributions Statement: In all Original Research and Brief Research Reports, the corresponding author must provide information on the contributions each author has made to the article. The purpose of this listing is to give credit where it is due. Additionally, this will serve to clearly identify who is responsible for the quality, accuracy, and ethics of the work, and to whom we may turn for details of the research not included in the manuscript. Listings should be brief and to the point. The details of our reasons for this requirement, and a discussion of the various types of authorship (along with samples) is elaborated elsewhere.[51]

An example of a typical description of a multicenter clinical trial might be:

MBK, BD, and NT conceived the study, designed the trial, and obtained research funding. MBK, BD, ML, and NT supervised the conduct of the trial and data collection. EW, SF, and MG undertook recruitment of participating centers and patients and managed the data, including quality control. NT and BD provided statistical advice on study design and analyzed the data; ML chaired the data oversight committee. BD drafted the manuscript, and all authors contributed substantially to its revision. MBK takes responsibility for the paper as a whole. Save the author contributions information as a separate electronic file for uploading to Editorial Manager.

Title Page: On the title page, include the title; the authors' full names, academic degrees (provide no more than 2 per author; do not include honorary affiliations, such as fellow status in an organization), and affiliations (including department, division, institution, city, state, and country) at the time of the study; the name of the meeting, city, state, and date (month and year) if the paper has been presented; acknowledgment of grants (including grant number) or other financial support, including compensation for consulting; the phrase "word count" followed by a numeric word count of the text (excluding abstract and references), and the phrase, "Address for reprints..." followed by the full name, address, telephone number, fax number, and e-mail address of the appropriate author. (If you do not wishreprints, simply write the phrase "Reprints not available from the authors" in this space). The same should be given for the Corresponding Author if it is different. Save the title page as a separate file for uploading to Editorial Manager.

Abstract, Article, and References: The abstract, main text of your manuscript, and the references should be combined into 1 electronic file for uploading to Editorial Manager. Number the pages beginning with the abstract. It is optional whether any tables or figures appear after thereferences or are uploaded as separate items in Editorial Manager.

Additional Required Documents

For certain types of manuscripts, additional items are required. If you are reprinting previously published tables or figures and have requested and/or obtained permission to use these items with your Annals' manuscript, you must forward the letters granting permission to the Annals' office. These items are required at the time the manuscript is uploaded to Editorial Manager and should be faxed to the editorial office, along with the Manuscript Submission Agreement.

If a revision of your manuscript is requested, you may be asked to send an original hard copy of any figures you included. If the manuscript is accepted for publication, the original hard copy of any figures may be required if the quality of the digital (electronic) image is insufficient to reproduce in the journal.

Randomized Controlled Trials (RCTs): All RCTs must be identified as such. The CONSORT checklist will help you make sure all key information is provided in the manuscript and we recommend you use it for this purpose; faxing it to the editorial office with the manuscript is optional. Annals encourages authors to register RCTs with the Cochrane Collaboration or another international trial registry and supply the trial number. The Trial Registration Form[53] for unreported controlled trials has been published in Annals previously.

Permissions: Any submitted material (including figures and tables) that has been published elsewhere must be accompanied by the written consent of the author and publisher for reproduction in Annals. Photographs of subjects showing any recognizable features must be accompanied by their signed release authorizing publication, as must case reports that provide enough unique identification of a person (other than name) to make recognition possible. Provide a statement of clearance to publish, if this is an institutional requirement. Obtain written permission from any persons acknowledged by name. Articles appear in both the print and online versions of the journal, and the wording of the permission letter should specify permission in all forms and media. Failure to obtain electronic permission rights may result in the images not appearing in the online version.

Institutional Review Board (IRB): Formal written IRB approval or exemption from full review must be obtained for any study involving human subjects or their records or animal research. In any research on human beings, the subject's freely given informed consent should be obtained in writing. Note this in the Methods section of the article. If your institution exempted your study from formal review, state this in the Methods section. If reviewers request it, you may be asked to provide documentation from the IRB.

Manuscript Preparation

Format: All manuscripts should be double-spaced with 1.5-inch margins. Number pages consecutively, beginning with the abstract. Be sure that your manuscript does not contain identifying information that would unblind the peer review process (see below). For additional guidelines, see the Instructions for Web-Based Submission .

Blinded Peer Review: Our policy is to blind reviewers to the authors' names and institutions. Although this process has not been shown to affect the quality of reviews, we believe it increases the likelihood of fairness. Do not identify the author, institution, or city anywhere in the abstract or text (including running heads), including referring to your own referenced work in the first person, of the article. Manuscripts that do not meet this requirement will not be processed. If you need to cite a previous paper of your own that describes methods or preceding studies, you can provide the citation but do not refer to it in the text in a way that reveals to the reader that you authored it.

Title Page: Follow the guidelines list in Section III. Formatting and Submitting Your Manuscript. Required Submission Documents

Abstract: For Original Research and Brief Research Reports, follow the instructions for original research listed above. For Concepts, Review Articles, and Case Reports, include a narrative abstract of no more than 250 words summarizing the paper.

Text: For Original Research and Brief Research Reports, divide the text into the sections: Introduction, Methods, Results, Limitations, and Discussion (including a final paragraph that summarizes the conclusion); and subheadings

Units of Measure: Provide units of measure in common reference values, followed by Systeme International (SI) units in parentheses.[54]

Drugs: Use generic names and, if necessary in the Methods section, list brand names (including the manufacturer's name, city, and state) in parentheses. Please include the International Nonproprietary Name (INN) as well.[55]

References: Do not use the endnote or footnote function of word processing software to generate a list of references. Number references (including references to unpublished information) consecutively in the order of their appearance in the manuscript. Type a list of references in their order of mention in the text, not alphabetically, at the end of the manuscript.[56] Abbreviate journal names according to Index Medicus. Indicate abstracts by "abstract" in parentheses. Annals' style is to list the first 3 authors, followed by "et al" if there are more than 3. Accuracy of citations is the author's responsibility. Examples of correct referencing forms are as follows:

Journal Article: Raftery KA, Smith-Coggins R, Chen AHM. Gender-associated differences in emergency department pain management. Ann Emerg Med. 1995;26:414-421.

Book: Huddy J. Emergency Department Design: A Practical Guide to Planning for the Future. Dallas, TX: American College of Emergency Physicians; 2002.

Book Chapter: Mengert TJ, Eisenberg MS. Prehospital and emergency medicine thrombolytic therapy. In: Tintinalli JE, Ruiz E, Krome RL, eds. Emergency Medicine: A Comprehensive Study Guide. 4th ed. New York NY: McGraw-Hill; 1996:337-343.

Courses, lectures (unpublished): Sokolove PE. Needlesticks and high-risk exposure. Course lecture presented at: American College of Emergency Physicians, Scientific Assembly, October 12, 1998; San Diego, CA.

Internet: Gore L. ACEP hails House passage of the HEALTH Act [press release]. American College of Emergency Physicians Web site. Available at: http://www.acep.org/1,32181,0.html . Accessed March 14, 2003.

Personal Communication: Avoid reference to personal communications, but when necessary, include the person's name, his or her title, month, and year. A letter granting permission to publish from the person providing the information must be included at the time of submission.

Tables: The purpose of a table is to summarize data in an easy-to-read format. Tables should be self-explanatory and able to stand alone; the data presented in a table should not be presented in the text. Number tables consecutively. Refer to each table consecutively in the text. Each table must be on a separate page after the references. Omit internal horizontal and vertical rules.

Figures: Figures and legends should be self-explanatory and able to stand alone; the data presented in a figure should not be duplicated in the text. Refer to each figure consecutively in the text. Good graphical design is of critical importance, and an effort should be made to maximize the amount of useful information portrayed.[45-49] Do not use 3-dimensional graphs or charts or solitary pie charts. If you are reporting a randomized controlled trial, your first figure should be the one described under Participant Flow in the CONSORT criteria.

Electronic Illustration Submission: On initial submission, figures must be submitted in electronic format. Images can be provided in Microsoft Word, WordPerfect, Microsoft Excel, RTF, TXT, LaTeX2e, AMSTex, TIF, GIF, JPEG, EPS, Postscript, PICT, or PDF format. Images created in presentation software, such as PowerPoint, CorelDraw, or Harvard Graphics, cannot be uploaded with your submission. If a revision of your manuscript is requested, only images in TIF or EPS format (Macintosh or PC generated) will be accepted. Images should be no more than 5 inches wide. Graphics software, such as Photoshop and Illustrator, should be used in the creation of the art. Gray scale images should be at least 300 DPI. Combinations of gray scale and line art should be at least 1200 DPI. Line art (black and white or color) should be at least 1200 DPI. If the manuscript is accepted for publication, the original hard copy of figures may be requested if the quality of the digital image is insufficient to reproduce in the journal.

Submission Checklist

Complete all applicable sections of this page, and submit with the manuscript.

1. __ Prepared manuscript following Instructions for Authors

2. __ Completed Manuscript Submission Agreement (fax [972-580-0051] and mail to the editorial office)

__ Conflict of interest/funding section completed

__ IRB/Informed Consent and Statistical Consultant sections completed

__ Signed by all authors

3. __ Cover letter including (saved as separate file for uploading)

__ Title

__ Brief description (include submission category)

__ Corresponding author

__ Potential conflicts of interest

__ Any potential forms of prior publication or concurrent submission

__ Copyright constraints

4. __ Statement of author contributions (saved as separate file for uploading)

5. __ Title page stating (saved as separate file for uploading):

__ Title

__ Authors' full names

__ Academic degrees (no more than 2; no honorary degrees)

__ Affiliations (at the time of the study, no honorary affiliations)

__ Presentation information

__ Funding or outside support information or financial interest in the product studied or the company that produces it

__ Word count of text (exclude abstract and references)

__ Address for reprints and correspondence; include e-mail address

6. __ Abstract (structured, for Original Research and Brief Research Reports), main text of the manuscript, and references (the abstract, text, and references can be saved as one file for uploading)

__ Numbered pages starting with abstract

__ Margins 1.5 inches

__ Entire text, double-spaced

__ No mention of author, institution, or city anywhere for blinded peer review

__ IRB/Informed consent statement in Methods section

7. __ Permissions to reproduce published material or personal communications in all forms and media (mail to the editorial office)

Failure to comply with these requirements will delay consideration of your manuscript.



Peer Review

The review process begins when Annals' staff reviews each manuscript uploaded into Editorial Manager for completeness and adherence to submission requirements. The manuscript is then forwarded electronically to one or more Annals' editors, based on subject matter. These editors read the manuscript and then either send it to expert peer reviewers or determine that it should be returned to the author without additional peer review. Some papers are returned at this stage because of quality issues, but also because of redundancy with material that has been published or is in press, or lack of suitability for this journal's readership. In such cases, a decision will be made by the editor to reduce the burden on our expert peer reviewers, and to provide the author the quickest response possible. The journal publishes an annual report in July listing characteristics of the peer review process (eg, acceptance rates of manuscripts, time from submission to decision).

Reviewers who are sent manuscripts are blinded to the identity of the authors or originating institution of the manuscript, and authors do not know the identity of the reviewers. Extensive ethical requirements[52] govern the peer review process. The assessments of the expert peer reviewers are read by the assigned editor, and often discussed with associate editors or the deputy editors of the journal before a decision is made. Before any manuscript can be accepted for publication, it is additionally reviewed and approved by the editor in chief.

When a decision is made, that decision is communicated to the corresponding author by e-mail. Reviewers' comments are included in the letter and often contain an editor's summary description of some of the main reasons for the decision. In the case of revision, authors should write a detailed cover letter responding to each of the points raised in the review. Even if the authors believe no change is necessary for a particular critique, a brief response is required that summarizes the reasons for their conclusion.

Peer Review and Conflict of Interest

Unless otherwise stated, all original articles published in Annals of Emergency Medicine are peer reviewed by members of the editorial board and expert reviewers. This includes original research, case conferences and reports, concepts and state-of-the-art reviews, special contributions, Clinical Topic Reviews in Evidence-Based Emergency Medicine, and Skills for Evidence-Based Emergency Care. Departments in the journal (Correspondence; Editorials; Brief Commentaries; Health Policy Report; Residents' Perspective; CDC Update; NHTSA Notes; Evidence-Based Emergency Medicine "Feedback," "Updates," and "Abstracts"; Change of Shift; and Media and Book Reviews) are peer reviewed by an editorial board member. In all Annals' peer review, the identity of authors is concealed from reviewers and vice versa. ACEP Policy Statements and Clinical Policies are not peer reviewed by the journal.

Annals makes every effort to minimize the opportunity for potential conflicts of interest by authors, reviewers, and editors. Authors are required, on submission of their manuscript, to state in writing whether or not outside funding or other form of support was received and its source, as well as whether or not they have any financial interest in the subject of their study. This information is published with the article. Reviewers are required to declare any potential conflict of interest in their review of a manuscript, and the editor then decides whether it is of a nature that requires they be replaced with another reviewer. Editors who have a conflict of interest regarding a manuscript are required to recuse themselves from the decisionmaking process.

The full details of these and other publication ethics policies of this journal are available on the Web at www.mosby.com/AnnEmergMed. Select Browse All Issues, select January 2003, and view "Journal policy on ethics in scientific publication" on page 82.

Withdrawing Manuscripts After Submission

Corresponding authors who wish to withdraw a manuscript after it has been submitted to Annals must provide a letter that is signed by the corresponding author and indicate the corresponding author represents the wishes of all authors in withdrawing the manuscript. Manuscripts will not be withdrawn from consideration once they have been submitted until the journal office receives the request in writing. Manuscripts cannot be withdrawn after final acceptance by the journal except for reasons of scientific error or misconduct.

Editing

All accepted manuscripts are edited for clarity and conformity to the American Medical Association's Manual of Style for editorial style. Authors will receive a page proof for approval. No changes will be accepted after the page proof is approved.

Responsibility for Published Information

Annals and the publisher are not responsible for statements made by any contributor. Authors are responsible for all statements made in the text, including editorial changes approved by the author. Authors should retain copies of all submitted material.

Copyright

Accepted manuscripts become the property of Annals and may not be published elsewhere without the written permission of Annals and the Publisher, Mosby, a division of Elsevier. Authors retain the right to reuse their published material in lectures, lecture notes, and exhibits.

Access to Data

During the peer review process, Annals may require access to the authors' original data for the sole purpose of better reviewing the submission. Annals reserves the right to review the complete primary data on which manuscripts are based, either before publication or, should questions arise, afterward. Annals will keep the data confidential and not otherwise directly benefit from the advance access to the data or the in-depth examination of them that stems from the peer review process.

Reprints

A reprint order form will be sent to the corresponding author at the time page proofs are sent. Readers should contact authors to obtain single copies of reprints. For multiple copies of an article, readers may also contact Mosby's reprint department at 314-453-4314.

Contacting the Editorial Office

If additional information is needed, contact the Annals' editorial office at 800-803-1403 or 972-550-0911; fax 972-580-0051; or e-mail to AnnEmergMed@acep.org.

For written editorial communication, contact:
Annals of Emergency Medicine
PO Box 619911
Dallas, TX 75261-9911
or
1125 Executive Circle
Irving, TX 75038-2522



References

1. ISI Web of Knowledge. Available at: http://www.isiwebofknowledge.com/. Accessed March 14, 2003.

2. American Medical Association. Manual of Style: A Guide for Authors and Editors. 9th ed. Chicago, IL: American Medical Association; 1998.

3. Begg C, Cho M, Eastwood S, et al. Improving the quality of reporting of randomized controlled trials. The CONSORT statement. JAMA. 1996;276:637-639.

4. Elbourne DR, Campbell MK. Extending the CONSORT statement to cluster randomized trials: for discussion. Stat Med. 2001;20:489-496.

5. Jaeschke R, Guyatt GH, Sackett DL. Users? guides to the medical literature: III. How to use an article about a diagnostic test. A. Are the results of the study valid? JAMA. 1994;271:389-391.

6. Jaeschke R, Guyatt GH, Sackett DL. Users? guides to the medical literature: III. How to use an article about a diagnostic test. B. What are the results and will they help me in caring for my patients? JAMA. 1994;271:703-707.

7. Bossuyt PM, Reitsma JB, Bruns DE, et al. Towards complete and accurate reporting of studies of diagnostic accuracy: the STARD initiative BMJ. 2003;326:41-44.

8. Wasson JH, Sox HC, Neff RK, et al. Clinical prediction rules. Applications and methodological standards. N Engl J Med. 1985;313:793-799.

9. Laupacis A, Sekar N, Stiell IG. Clinical prediction rules. A review and suggested modifications of methodological standards. JAMA. 1997;277:488-494.

10. Gold MR, Siegel JE, Russell LB, et al, eds. Cost-Effectiveness in Health and Medicine. Oxford, United Kingdom: Oxford University Press; 1996.

11. Russell LB, Gold MR, Siegel JE, et al. The role of cost-effectiveness analysis in health and medicine. JAMA. 1996;276:1172-1177.

12. Siegel JE, Weinstein MC, Russell LB, et al. Recommendations for reporting cost-effectiveness analyses. JAMA. 1996;276:1339-1341.

13. Weinstein MC, Siegel JE, Gold MR, et al. Recommendations of the panel on cost-effectiveness in health and medicine. JAMA. 1996;276:1253-1258.

14. Moher D, Cook DJ, Eastwood S, et al. Improving the quality of reports of meta-analyses of randomised controlled trials: the QUOROM statement. Quality of Reporting of Meta-analyses. Lancet. 1999;354:1896-1900.

15. Irwig L, Tosteson ANA, Gatsonis C, et al. Guidelines for meta-analyses evaluating diagnostic tests. Ann Intern Med. 1994;120:667-676.

16. Moses LE, Shapiro D, Littenberg B. Combining independent studies of a diagnostic test into a summary ROC curve: data-analytic approaches and some additional considerations. Stat Med. 1993;12:1293-1316.

17. Stroup DF, Berlin JA, Morton SC, et al. Meta-analysis of observational studies in epidemiology: a proposal for reporting. Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group. JAMA. 2000;283:2008-2012.

18. Mays N, Pope C. Qualitative research in health care. Assessing quality in qualitative research. BMJ. 2000;320:50-52.

19. Giacomini MK, Cook DJ. Users? guides to the medical literature: XXIII. Qualitative research in health care. A. Are the results valid? JAMA. 2000;284:357-362.

20. Giacomini MK, Cook DJ. Users? guides to the medical literature: XXIII. Qualitative research in health care. B. What are the results and how do they help me care for my patients? Evidence-Based Medicine Working Group. JAMA. 2000;284:478-482.

21. Pope C, Mays N, eds. Qualitative Research in Health Care. London, United Kingdom: BMJ Books; 2000.

22. Gilbert EH, Lowenstein SR, Koziol-McLain J, et al. Chart reviews in emergency medicine: where are the methods? Ann Emerg Med. 1996;27:305-308.

23. Fink A, Kosecoff J, eds. How to Conduct Surveys: A Step by Step Guide. Beverly Hills, CA: Sage; 1998.

24. Cox BG, Cohen SB. Methodological Issues for Health Care Surveys. New York, NY: Marcel Dekker, Inc.; 1985.

25. Couper MP, Traugott MW Lamias MJ. Web survey design. Public Opin Q. 2001;65:231-253.

26. Fink A, ed. The Survey Kit. Newbury Park, CA: Sage; 2002.

27. Schriger DL, Wears RL, Cooper RJ, et al. Upgrading our Instructions for Authors. Ann Emerg Med. 2003;41:565-567.

28. Greenland S. Basic methods for sensitivity analysis of biases. Int J Epidemiol. 1996;25:1107-1116.

29. Greenland S. Basic methods for sensitivity analysis30. Rozeboom W. The fallacy of the null hypothesis significance test. Psychol Bull. 1960;57:416-428.

31. Cohen J. The earth is round (p < 0.05). Am Psychol. 1994;47:997-1003.

32. Krueger J. Null hypothesis significance testing. On the survival of a flawed method. Am Psychol. 2001;56:16-26.

33. Gallagher J. P > 0.05: threshold for decerebrate genuflection. Acad Emerg Med. 1999;6:1084-1087.

34. Sterne J, Smith G. Sifting the evidence what's wrong with significance tests? BMJ. 2001;322:226-230.

35. Berger J, Berry D. Statistical analysis and the illusion of objectivity. Am Scientist. 1987;76:159-165.

36. Braitman LE. Confidence intervals extract clinically useful information from data. Ann Intern Med. 1988;108:296-298.

37. Gardner MJ, Altman DG. Confidence intervals rather than P values: estimation rather than hypothesis testing. BMJ. 1986;292:746-750.

38. Schriger DL. Problems with current methods in data analysis and reporting, and suggestions for moving beyond incorrect ritual. Eur J Emerg Med. 2002;9:203-207.

39. Royall RM. Statistical Evidence: A Likelihood Paradigm. Monographs on Statistics and Applied Probability Vol 71. Boca Raton, FL: Chapman & Hall/CRC; 1997:191.

40. Poole C. Beyond the confidence interval. Am J Public Health. 1987;77:195-198.

41. Goodman SN. Toward evidence-based medical statistics. 2: the Bayes factor. Ann Intern Med. 1999;130:1005-1013.

42. Matthews RAJ. Why should clinicians care about Bayesian methods? Available at: http://ourworld.compuserve.com/homepages/rajm/jspib.htm . Accessed February 1, 2003.

43. Good IJ. The Bayesian influence, or how to sweep subjectivism under the carpet. In: Harper WL, Hooker CA, eds. Foundations of Probability Theory, Statistical Inference and Statistical Theories of Science. Volume 2. Dordrecht, Holland; Reidel 1976:125-174.

44. Lang TA, Secic M. How to Report Statistics in Medicine: Annotated Guidelines for Authors, Editors, and Reviewers. Philadelphia, PA: American College of Physicians; 1997.

45. Tufte ER. The Visual Display of Quantitative Information. Cheshire, CT: Graphics Press; 1983.

46. Tufte ER. Envisioning Information. Cheshire, CT: Graphics Press; 1990.

47. Tufte ER. Visual Explanations. Cheshire, CT: Graphics Press; 1997.

48. Cleveland WS. Visualizing Data. Summit, NJ: Hobart Press; 1993.

49. Schriger DL, Cooper RJ. Achieving graphical excellence: suggestions and methods for creating high-quality visual displays of experimental data. Ann Emerg Med. 2001;37:75-87.

50. http://www.wma.net/e/policy/17-c_e.html . Accessed February 1, 2003.

51. Callaham ML, Waeckerle JF. Deciphering the authorship code. Ann Emerg Med. 2001;37:386-387.

52. Callaham ML. Journal policy on ethics in scientific publication. Ann Emerg Med. 2003;41:82-89.

53. Waeckerle JF, Callaham ML. Medical editors trial amnesty. Ann Emerg Med. 1998;31:400-401.

54. American Medical Association. Manual of Style: A Guide for Authors and Editors. 9th ed. Chicago, IL: American Medical Association; 1998:328, 485.

55. American Medical Association. Manual of Style: A Guide for Authors and Editors. 9th ed. Chicago, IL: American Medical Association; 1998:355.

56. American Medical Association. Manual of Style: A Guide for Authors and Editors. 9th ed. Chicago, IL: American Medical Association; 1998:34.

57. Pearl J. Causality: Models, Reasoning, and Inference. Cambridge, United Kingdom: Cambridge University Press; 2001.

58. Greenland S, Pearl J, Robins JM. Causal diagrams for epidemiologic research. Epidemiology. 1999; 10:37-48.

59. Cooper RJ, Schriger DL. Evidence-based emergency department discharge planning: how do we get there? Ann Emerg Med. 1999;34:667-669.

60. Altman DG, Dore CJ. Randomisation and baseline comparisons in clinical trials. Lancet. 1990;335:149-153.

61. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet. 1986;1:307-310.

62. Bland JM, Altman DG. Comparing methods of measurement: why plotting difference against standard method is misleading. Lancet. 1995;346:1085-1087.

63. Altman DG, Matthews JNS. Statistics notes: interaction 1: heterogeneity of effects. BMJ. 1996;313:486.

64. Matthews JNS, Altman DG. Statistics notes: interaction 2: compare effect sizes not P values. BMJ. 1996;313:808.

65. Matthews JNS, Altman DG. Statistics notes: interaction 3: how to examine heterogeneity. BMJ. 1996;313:862.

66. Altman DG, Bland JM. Statistics notes: units of analysis. BMJ. 1997;314:1874.

67. Bland JM, Kerry SM. Statistics notes. Trials randomised in clusters. BMJ. 1997;315:600.

68. Kerry SM, Bland JM. Sample size in cluster randomisation. BMJ. 1998;316:549.

69. Kerry SM, Bland JM. Analysis of a trial randomised in clusters. BMJ. 1998;316:54.

70. Wears RL. Advanced statistics: statistical methods for analyzing cluster and cluster-randomized data. Acad Emerg Med. 2002;9:330-341.

71. Bland JM. Cluster designs: a personal view. Available at: http://www.sghms.ac.uk/depts/phs/staff/jmb/clustalk.htm . Accessed February 2, 2003.

72. Greenland S. Rothman KJ. Fundamentals of epidemiologic data analysis. In: Modern Epidemiology. 2nd ed. Philadelphia, PA: Lippincott-Raven; 1998:225-229.

73. Bland JM, Altman DG. Multiple significance tests: the Bonferroni method. BMJ. 1995;310:170.

74. Perneger TV. What?s wrong with Bonferroni adjustments. BMJ. 1998;316:1236-1238.

75. Bender R, Lange S. Multiple test procedures other than Bonferroni's deserve wider use. BMJ. 1999;318:600-601.

76. Greenland S. Introduction to regression models. In: Rothman KJ, Greenland S, eds. Modern Epidemiology. 2nd ed. Philadelphia, PA: Lippincott-Raven; 1998:359-399.

77. Greenland S. Introduction to regression modeling. In: Rothman KJ, Greenland S, eds. Modern Epidemiology. 2nd ed. Philadelphia, PA: Lippincott-Raven; 1998:401-432.

78. Harrell FE. Regression Modeling Strategies: With Applications to Linear Models, Logistic Regression and Survival Analysis. New York, NY: Springer-Verlag; 2001.

79. Kleinbaum DG, Kupper LL, Morgenstern H. Epidemiologic Research: Principles and Quantitative Methods. New York, NY: Van Nostrand Reinhold Company; 1982.

80. Schwartz RJ, Panacek EA. Basics of research (part 7): archival data research. Air Med J. 1996;15:119-124.