Writing Research Articles on Benefit Design and Policy

October 2008, Vol 1, No 8 - Editorial
William E. Fassett, BSPharm, MBA, PhD
Charles R. Phillips, PharmD, PhD

Readers of American Health & Drug Benefits (AHDB) are invited to submit research articles relevant to benefit design. Successful publication of an author’s research arises from selecting a journal whose readers’ needs and interests match the topic of the manuscript. Health and drug benefit administrators and policy makers need reliable information predicting how changes in benefit design will affect healthcare outcomes and the behaviors of stakeholders. A successful manuscript clearly addresses a current issue of importance to a national audience of drug policy makers and administrators, reports results that are reliable and justified by the design and conduct of the research, and exhibits characteristics of good writing.

Aspiring authors may see achieving publication of a manuscript as a daunting task. The act of writing is hard work—Gene Fowler is reputed to have said, “Writing is easy: all you do is sit staring at a blank sheet of paper until drops of blood form on your forehead.”1 Having submitted the manuscript, the author faces possible rejection. For example, American Psychological Association journals in 2006 reported an overall rejection rate of 78%, ranging from 34% for the Journal of Comparative Psychology to 89% for JEP: Learning, Memory, and Cognition.2 Does this mean that an author must submit 2 or 3 articles to get 1 published? In fact, it does not—the publication process is not a crapshoot. Successful authors are repeatedly successful; many experienced authors find that virtually all their manuscripts reach publication. Repeatedly successful authors apply several key principles in their work.3,4 Many of these principles can help authors who submit manuscripts to AHDB.

A Solid Foundation

Every successful manuscript serves a particular need or answers a question of interest to the intended audience. Key to a successful article is welldesigned research or scholarship. Four important tasks face the researcher. First, it is critical to understand the current state of knowledge in the area selected for study, and this means to fully understand the current literature. The researcher, then, frames a research question that will add new information to the literature. Each type of question requires a particular approach, which means implementing solid research methods while designing the study. Finally, the researcher must make sure that the source of information or study population is appropriate to yield reliable results that have meaning to the intended audience.

List as authors all persons who participate significantly in the creation of the manuscript and acknowledge minor contributions. Putting one’s name on a ghostwritten article is a violation of publication ethics.5

Write for the Intended Reader

AHDB readers are policy makers and administrators responsible for implementing health and drug benefit programs. They aim to deliver the greatest value to health plan stakeholders. They seek to maximize healthcare outcomes for patients while minimizing costs to payers, which involves evaluating incentives, products, and services that are included in the benefit package. Articles submitted to the journal must provide information relevant to these concerns.

Understand, Summarize the Literature

The researcher is a member of a community of individuals interested in answering important questions about a particular area. Journals seek to publish new and unique findings of interest to their readers, and do not wish to publish material previously revealed elsewhere, whether by the same or different authors. Therefore, the aspiring author must be up to date on the current literature relevant to the proposed study.The author must summarize the literature, identify where new information is needed, and explain why his or her study fits that need.

A Clear Research Question

A clearly defined research question is essential and may propose new theory, seek to correct current thinking, ask about new methods of research, or ask how current theory applies in actual practice. The investigator must clearly define the intervention or policy change and the outcome measures that will assess the intervention’s effect.

Examples of specific research questions common to benefit design include:

• How will pharmacy spending and utilization change from implementing a plan or a public policy change (multitiered copayments, mandatory generic substitution, coinsurance, prior authorization, formularies)?

• How will medical spending and utilization change as a result of a plan or a public policy change?

• How will a proposed change affect health outcomes?6

• Which intervention or product produces the best benefit-to-cost ratio, the most cost-effective outcome, or the highest quality-adjusted life extension?

The researchers must carefully choose the interventions to study, how to measure the outcomes, and the perspective to use for the analysis. Should they base pharmacy costs on the cost to the plan, the patient, or even society? Should they define costs as per member per month, as total out of pocket, or some other measurement? The researchers need to be explicit about their measurements to be able to specifically address the research question.

Match the Method to the Question

To answer the research question, it is essential to implement well-designed and accepted research methods. Whether the research question deals with cost containment, utilization control, medical therapy management, or outcomes management, each research approach has specific practices.

Each of the accepted designs used by researchers studying pharmacy benefits follows rules of design and measurement.6 Some of these research approaches are:

• Clinical trials, retrospective reviews of pharmacy claims data during and after benefit design changes, pharmacoeconomic studies to compare competing alternatives

• Cross-section studies that depict utilization and spending under current conditions, retrospective and prospective studies assessing risk

• Longitudinal and time-series studies that track outcome changes over time.

There is no perfect study, but to obtain valid, reliable, and usable results, the appropriate research design has the fewest and least-fatal design flaws.

Use Appropriate Sample

If results are going to be applicable to a reader’s population, the sample or source of information must be generalizable to a wider group of people than just those studied. When a study’s results are not generalizable, they are merely an interesting anecdote that requires more study in other populations.

A sample of patients similar to those served by the readers of the journal has a greater chance of publication. A sample skewed by demographics—such as elderly or Medicaid populations, or by unique interventions unlikely useful to others7—limits the impact of their results and jeopardizes the study’s publication. Other common problems are samples that are too small for adequate statistical testing, and bias introduced by the sampling method. For example, internet surveys do not reach people who lack adequate computer access,8 and telephone surveys often miss low-income or young people.9

Follow the Author Guidelines Carefully

Each journal has identified a potential audience and has a plan that defines the types of articles sought and the style and format of the journal. The guidelines for authors set forth the criteria for acceptance derived from this plan. Research articles in AHDB are judged at the outset according to these guidelines and must follow the required format. The AHDB guidelines are relatively general, so it is helpful here to describe a typical outline for a research article. At least the following sections should be present:

Abstract: The abstract provides the reader with a concise summary of the article. The abstract must include the objective, methods, results, and conclusions. Readers must be able to learn from the abstract the key study results and decide whether the article is applicable to their environment and, therefore, worthy of further review.

Introduction: The introduction must concisely summarize the literature and the importance of the research question to the reader. It should identify key gaps in current knowledge and suggest how the proposed study addresses those gaps. An introduction that is beyond a page or so is generally too long.

Methods: The methods section should inform thelocation of the study, and the data collection time frame. The design of the study is described, and relevant statistical tests must be disclosed. The author must explain how the survey instrument was developed and validated. When human subject research is involved, the author must document the nature of Institutional Review Board approval. Include appropriate references that justify the particular methods used in the study. The methods section should be sufficient to allow another researcher to replicate the study.

Results: This section presents the study findings. It should show results that address the research questions and objectives, as well as any results supporting the validity of the data. In general, reserve interpretations of the findings or comments on weaknesses or strengths of the data for the discussion section.

Discussion: The discussion explains how the author interprets the results in the context of current knowledge. It should also mention potential limitations of the study.

Conclusion: The author’s conclusions should comment on additional implications of the study, and if the study met its intended purpose. The author should explain the applicability of the results to health and benefit design and implementation. In general, the conclusions section should be a brief summary that encapsulates the study findings and their meaning.

Revise, Revise, Revise

Rejected manuscripts fail for 2 major reasons (and often for both): (1) the article is conceptually unsound in the design and conduct of the study; and (2) the article is poorly organized and written. Sometimes an otherwise acceptably designed study is rejected because the author fails to describe it clearly and fully. We recommend that an author revise a manuscript at least 3 times before submitting it for publication.

Revise first for clarity and organization. This should include asking another colleague to review the article and comment on any major missing pieces, overall organization, and if the colleague understands the author’s descriptions and conclusions.

Revise next for conciseness. Journals have limited space and readers have limited time. Most journals specify upper limits for word counts; AHDB’s target for peer-reviewed articles is 3000 words. AHDB generally limits abstracts to 250 words and references to 30.

Finally, revise to conform to the journal’s format and style guidelines. Ensure correct citation of every reference, and verify that all data and figures are correct and accurate. References must be numbered consecutively and listed at the end of the article. Figures and tables should be placed after the references.

Successful authors usually must revise the manuscript again after peer review. When the editors ask the author to make these changes, congratulations are in order. The peer reviewers and editors have found the article generally appropriate and suitable for publication, subject to the revisions. Perform this revision promptly, and comply with all the suggestions. Look forward to seeing the manuscript in print.


1. Brainy Quote: Gene Fowler quotations. http://www.brainy quotes/authors/g/gene_fowler.html. Accessed August 10, 2008.

2. American Psychological Association. Summary Report of Journal Operations, 2006. 2007. operations. pdf. Accessed August 10, 2008.

3. Nahata MC. Tips for writing and publishing an article. Ann Pharmacother. 2008;42:273-277.

4. Emerald Group Publishing Limited. How to…survive peer review and revise your paper. authors/guides/ review.htm?part=1. Accessed August 10, 2008.

5. DeAngelis CD, Fontanarosa PB. Impugning the integrity of medical science: the adverse effects of industry influence. JAMA. 2008;299:1833-1835.

6. Goldman DP, Joyce GF, Zheng Y. Prescription drug cost sharing: associations with medication and medical utilization and spending and health. JAMA. 2007;298:61-69.

7. Joyce GF, Escarce JJ, Solomon MD, et al. Employer drug benefit plans and spending on prescription drugs. JAMA. 2002;288:1733-1739.

8. Hartford K, Carey R, Mendonca J. Sampling bias in an international internet survey of diversion programs in the criminal justice system. Eval Health Prof. 2007;30:35-46.

9. Blumberg SJ, Luke JV. Coverage bias in traditional telephone surveys of low-income and young adults. Public Opin Quart. 2007;71:734-749.

Related Items
Post-Pandemic Physician Leadership
David B. Nash, MD, MBA
September 2021 Vol 14, No 3 published on September 14, 2021 in Editorial
Thinking About 2030
David B. Nash, MD, MBA
June 2021 Vol 14, No 2 published on June 17, 2021 in Editorial
What Is Your IQ(VIA)?
David B. Nash, MD, MBA
March 2021 Vol 14, No 1 published on March 30, 2021 in Editorial
The Dirty Dozen Redux
David B. Nash, MD, MBA
Web Exclusives published on December 28, 2020 in Editorial
Pandemic Posting
David B. Nash, MD, MBA
October/November 2020 Vol 13, No 5 published on November 30, 2020 in Editorial
Last modified: August 30, 2021
Copyright © Engage Healthcare Communications, LLC. All rights reserved.