Article Text


New guidelines for reporting improvement research
Why new guidelines for reporting improvement research? And why now?
  1. D P Stevens
  1. Correspondence to:
 Dr D P Stevens
 Association of American Medical Colleges, 2450 N Street, NW Washington, DC 20037, USA;

Statistics from

An opportunity for readers, authors, and reviewers to voice their opinions on the draft new guidelines

It is reasonable to ask why QSHC would embark on additional guidelines for reporting improvement research—and why now?1 Useful and valuable guidelines for quality improvement reports (QIRs) were initially published in our predecessor journal, Quality in Health Care, in 1999,2 and reprinted again in QSHC in 2004. The BMJ also adopted this format in 2001.3 QIRs were initially slow to appear, but they have been published recently with increasing regularity in QSHC, particularly since the QIR guidelines were reprinted.

In their paper in this issue of QSHC Davidoff and Batalden emphasize that the proposed new guidelines are not intended to supersede QIRs.1 Indeed, they are best seen as part of the continuum of improvement scholarship that ranges from individual case reports and QIRs to studies that address complex systems and safety issues.4 In their accompanying commentaries, Berwick5 and Thomson6 serve to expand on this perspective.

Granted, this initiative is in one sense about scholarship and the scholar. But at its core it is about patient care and the patient. The purpose of critical, scholarly investigation in healthcare improvement is to provide the important new knowledge that leads to better, safer care. Moss and Thomson, in framing guidelines for QIRs, put their focus on the patient.2 So it should be with this initiative. To that end, the assessment of the impact of these draft guidelines should include their contribution to improvement in patient care.

The authors of the CONSORT guidelines for reporting randomized controlled trials have provided a road map for the refinement of publication guidelines.7 They have emphasized that it is an ongoing process, and that important components of the process include transparency, consensus, regular updating to reflect their usefulness, and constant testing to assure that the guidelines that are retained are indeed of demonstrable utility.7

By publishing these proposed guidelines,1 the editors of QSHC commit to several strategies that we believe will contribute value to this consensus project. First, the guidelines are published in this issue as a draft. Accordingly, we invite responses to a questionnaire regarding this draft on our home page ( Secondly, we will commission formal commentaries and editorials in subsequent issues of QSHC that will address key issues surrounding these guidelines. Thirdly, QSHC will road test the guidelines by asking authors, reviewers, and editors to use them in their work with this journal. We will encourage their responses to a brief questionnaire that will be used specifically to assess its utility for authors and reviewers. Fourthly, we plan to hold a consensus meeting in 2006 to foster further discussion and consensus among the broader academic and editorial communities regarding the framework proposed by these authors. Fifthly, a revised version of the guidelines—the product of these extended tests and discussions—will be published in subsequent issues of QSHC, currently planned next for October 2006.

One of Tom Nolan’s basic tenets of improvement is that one must answer the question: “How do we know that a change is an improvement?” His answer is by measurement.8 It is clear that the outcomes of this process will challenge our best assessment skills. Here are five suggested outcome measures.

  • If crafted well, guidelines will offer better and more explicit guidance for authors and editors.

  • Their availability should accelerate the appearance of publications of consistently higher value, thereby contributing to what Grove has called “research efficiency”.9 (On the other side of the ledger, if they add unnecessarily doctrinaire or superfluous tasks, they could conceivably slow innovation.10)

  • Their utility should be reflected in more frequent publication of improvement work in higher impact journals.

  • By extension, guidelines should contribute to better design of studies.

  • They could conceivably influence granting agencies in their criteria for supporting such research.8,10

There is more at stake here than comprehensive rules for scientific reporting. Hence the need to get it right.

Why new guidelines? And why now? The answer in the largest context is that patient care is not nearly at the level of quality that is possible. The explanations for this gap between knowledge and practice are many and complex,11,12 and go well beyond medical journals and scientific publications. Nevertheless, these guidelines invite an expanded discussion regarding the important research that can contribute to healthcare quality and patient safety. We at QSHC commit to serve in every way possible to facilitate that discussion.

An opportunity for readers, authors, and reviewers to voice their opinions on the draft new guidelines

View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.