Standards for Reporting Implementation Studies (StaRI) StatementBMJ 2017; 356 doi: https://doi.org/10.1136/bmj.i6795 (Published 06 March 2017) Cite this as: BMJ 2017;356:i6795
All rapid responses
We thank the authors of the SQUIRE reporting standards for their interest in StaRI. It is, perhaps, not so surprising that parallel groups working in related fields reach conclusions that resonate – indeed the similarities may be perceived as reinforcing the conclusions.
We were, of course, aware of the SQUIRE 1.0 (2008) reporting standards1 when we wrote our protocol (registered with EQUATOR in September 20132). Our decision to proceed with StaRI stemmed from a concern that SQUIRE 1.0’s stated focus was on quality improvement interventions focussing ‘primarily on making care better at unique local sites, rather than on generating new, generalisable scientific knowledge’.3 StaRI’s focus was broader: reporting studies that built on and extended a more formal evaluative approach (for example the MRC’s cyclical complex intervention framework,4,5 or the Dissemination and Implementation (D&I) linear framework6) and that explicitly aimed to generate scientific knowledge applicable beyond the individual system under study.7 A National Institutes of Health working group on D&I research in 2013 systematically assessed existing reporting standards (including SQUIRE) and similarly concluded that ‘further exploration of D&I-specific reporting guidelines with EQUATOR methodology was warranted’.8
The defining concepts and over-arching components that characterise StaRI emerged early in the process. For example, the ‘dual aims of reporting on the process of implementation and effectiveness of the intervention’, and the need to ‘monitor fidelity to an intervention whilst encouraging adaptation to suit diverse local contexts’ were predicted by the free text comments in the e-Delphi exercise that we undertook early in 2014 and are discussed in the subsequent paper.9 These were picked up in the consensus meeting in April 2015 and shaped the conclusions of the meeting which were subsequently refined for publication.10,11 In the interim the updated SQUIRE 2.0 was published,12 which broadened its remit to ‘the many approaches used for systematically improving the quality, safety and value of healthcare’, though the checklist still reflects a focus on ‘the local problem’ and the ‘evolving’ process associated with improvement interventions.
Davies et al in their response helpfully highlight the similarities between StaRI and SQUIRE 2.0. Both reporting standards now embrace multiple study designs and refer to other standards for methodological details. Of note, the World Health Organization guidelines (to which StaRI contributed) which focus on improving reporting of their fieldwork adopted the alternative approach of incorporating a range of methodological standards in one document.13 Both StaRI and SQUIRE emphasise the concept of an underpinning rationale/logic pathway, the importance of context, and the need for detailed descriptions of the intervention and implementation strategy (as in SQUIRE,11 the StaRI Explanation and Elaboration document 12 refers authors to TIDIER14 as a useful framework). The important tension between fidelity and adaptation is implicit in SQUIRE but explicitly highlighted in StaRI.
Davies et al, however, do not identify the fundamental difference between StaRI and SQUIRE; the key distinction between the implementation strategy and the evidence-based clinical, healthcare, global health or public health intervention being implemented. This was presaged in the StaRI e-Delphi,9 and discussed at length in the consensus meeting, emerging as a defining feature of the StaRI reporting standards.10,11 This imposes a conceptual rigour not only on reporting (illustrated as two strands in the checklist), but potentially also on the design of implementation studies. Explicitly distinguishing the implementation strategy from the intervention was a turning point in our discussions and we hope the concept will prove helpful not only to researchers, but also to healthcare professionals and managers developing strategies to improve the delivery of healthcare.
Implementation science is an emerging field, and both StaRI and SQUIRE 2.0 (and other initiatives) add to the debate about frameworks, standards, definitions and terminology.15 As Davies et al observe ‘we share the same spirit’ and it behoves us now to discuss and harmonise the insights from our different perspectives. We hope that this exchange of letters will lead to on-going discussion and collaboration.
1. Davidoff F, Batalden PB, Stevens DP, Ogrinc GS, Mooney SE, for the Squire development group. Publication guidelines for quality improvement in health care: evolution of the SQUIRE project. Qual Saf Health Care 2008;17(Suppl I):i3–i9
2. Pinnock H, Taylor S, Epiphaniou E, et al. Developing Standards for Reporting Phase IV Implementation studies. Available from http://www.equator-network.org/wp-content/uploads/2013/09/Proposal-for-r... (Accessed May 2017)
3. Ogrinc G, Mooney SE, Estrada C, et al. The SQUIRE (Standards for QUality Improvement Reporting Excellence) guidelines for quality improvement reporting: explanation and elaboration. Qual Saf Health Care 2008;17(Suppl I):i13–i32
4. Craig P, Dieppe P, Macintyre S, Michie S, Nazareth I, Petticrew M. Developing and evaluating complex interventions: the new Medical Research Council guidance. London: MRC; 2008. Available from www.mrc.ac.uk/complexinterventionsguidance (Accessed June 2016)
5. Pinnock H, Epiphaniou E, Taylor SJC. Phase IV implementation studies: the forgotten finale to the complex intervention methodology framework. Annals ATS 2014;11: S118-S122
6. Lansdverk J, Brown CH, Chamberlain P et al. Chapter 12. Design and analysis in dissemination and implementation research, in Brownson RC, Colditz GA, Proctor EK (eds) Dissemination and implementation research in health: translating science into practice. New York: Oxford University Press 2012
7. Bauer MS, Damschroder L, Hagedorn H, Smith J, Kilbourne AM. An introduction to implementation science for the non-specialist. BMC Psychology 2015; 3:32
8. Neta G, Glasgow RE, Carpenter CR, et al. A Framework for Enhancing the Value of Research for Dissemination and Implementation Am J Public Health 2015;2105:49-57
9. Pinnock H, Epiphaniou E, Sheikh A, et al. Developing Standards for Reporting Implementation studies (StaRI): an e-Delphi. Implement Sci 2015;10:42
10. Pinnock H, Barwick M, Carpenter C, et al., for the StaRI group. Standards for Reporting Implementation Studies (StaRI) statement. BMJ 2017;347:f6753
11. Pinnock H, Barwick M, Carpenter C, et al., for the StaRI group. Standards for Reporting Implementation Studies (StaRI). Explanation and Elaboration document. BMJ Open 2017;7:e013318
12. Ogrinc G, Davies L, Goodman D, Batalden P, Davidoff F, Stevens D. SQUIRE 2.0 (Standards for QUality Improvement Reporting Excellence): revised publication guidelines from a detailed consensus process. BMJ Qual Saf 2016;25:986–992
13. Hales S, Lesher-Trevino A, Ford N, et al. Reporting guidelines for implementation and operational research. Bull World Health Org 2016;94:58-64
14. Hoffmann TC, Glasziou PP, Boutron I, et al. Better reporting of interventions: template for intervention description and replication (TIDieR) checklist and guide. BMJ 2014;348:g1687
15. Lilford R, Chilton P. And Today We Have the Naming of Parts. NIHR CLAHRC West Midlands News Blog. June 2015. Available from https://clahrcwmblog.wordpress.com/2017/05/05/naming-of-parts/#DCB201705... (accessed May 2017)
Competing interests: No competing interests
To the Editor,
We were surprised to see the recently published StaRI guidelines for implementation science [Pinnock et al]. StaRI is strikingly similar to SQUIRE 2.0 – the publication guidelines to promote excellence in healthcare improvement reporting1. We are concerned that StaRI may create unintended confusion in the field.
The SQUIRE guidelines represent the input of hundreds of people over a 10 year period of transparent development, evaluation, and revision2-5. We recognize that the StaRI team may have been unaware of the philosophy and development that SQUIRE represents, just as we were unaware of the work of StaRI until its release. We believe the divisions between the science of improvement and implementation science are a false dichotomy. Thus, we will use this forum to briefly outline where the StaRI and SQUIRE converge. We hope this will facilitate a dialogue that can move the field forward as a whole.
The purpose of StaRI and SQUIRE share the same spirit. SQUIRE “provide[s] a framework for the reporting of new knowledge about how to improve healthcare”. StaRI’s purpose is “improve reporting of implementation studies … with the aim of enhancing adoption and sustainability of effective interventions”.
The scope of StaRI and SQUIRE are the same: both cover ‘the range of designs’ used in the field. Additionally, both guidelines indicate that not all items will apply to every study, and should not be required for every manuscript.
A goal for the most recent version of SQUIRE was to create clarity around terminology – favoring plain language over field-specific jargon. It has three key concepts. Authors who use SQUIRE should provide a rationale for their intervention, should study their intervention(s) - not simply report outcomes, and should describe the context of the work. SQUIRE’s three key concepts are recapitulated in StaRI ‘s two defining concepts and 3 overarching components.
In the first defining concept of StaRI, authors are urged to describe the strategy of implementation. Though the authors do not reference TIDIER, these guidelines were released in 2014 and provide a framework for clear reporting of interventions6. SQUIRE directs authors to this tool for the appropriate guidance.
In the second defining concept, StaRI asks authors to report the “impact of the intervention on the health of the target population … [and] even when the evidence is strong, [consider] the possibility that the intervention may be attenuated…”. The items for this include assessments of fidelity, resource use, costs, etc. This is nearly the same as the SQUIRE key concept that invites authors to ‘study their intervention’. The items for StaRI and SQUIRE in this area are similar.
The first of StaRI’s three overarching components is that there must be a hypothesis. This approximates the first key concept of SQUIRE, that there should be a theory or rationale underpinning the proposed interventions7.
The second component of StaRI is the requirement to describe the balance between fidelity to and adaptation of the intervention. It is highly similar to the TIDIER guidelines and also contains elements that SQUIRE places under the heading of ‘the study of the intervention’. This concept is hard to understand and communicate (whether in SQUIRE or StaRI), but at least one publication addresses it8. When we teach about SQUIRE we use the following plain language to describe this concept: ‘did things get better for the reasons you think they did? Were there unintended consequences? What was the impact of the intervention on the people, processes and systems involved?’
The third component of StaRI is the requirement that authors describe context. SQUIRE incorporates items for the description of context in the methods, results and discussion section.
StaRI describes SQUIRE as a publication guideline for quality improvement reports, but this conflates SQUIRE with the classic eight point guideline for quality improvement reports by Moss and Thompson9. SQUIRE is distinguished from Moss and Thompson’s work by having a different purpose, which is to support the reporting of “formal planned empirical studies on the development and testing of improvement interventions”2 and the “reporting of new knowledge about how to improve health care”1. Moss and Thompson’s quality improvement report guideline offers an alternative to the IMRaD format of biomedical reporting, while SQUIRE stays within the IMRaD format.
During the development of SQUIRE 2.0, we noted that the words ‘quality improvement’ had become confusing, because the phrase is associated with specific methodologies that are used in this area of work. SQUIRE is not intended to be restricted to these, but rather applies to any systematic method to improve the quality, safety, and value of healthcare. To resolve the confusion, we retained the SQUIRE acronym to hold on to the guideline’s origins and history, but explicitly moved toward clearer language – SQUIRE’s tag line is now: “promoting excellence in healthcare improvement reporting”. This accurately reflects the intent of the guidelines by leaving space for the many ways and methods that one might use to improve healthcare.
To be sure, some differences between StaRI and SQUIRE remain, especially in the assumptions about timing of iteration and adaptation of interventions. StaRI emphasizes alterations to a specific planned intervention during execution (“implementation cycle”). In SQUIRE the alterations are assumed to most typically occur during both development and execution. SQUIRE does not constrain alterations to a specific time frame.
As we reflect, we are struck by the fact that one field has named itself for the process of making healthcare better (implementation science) while the other has named itself for the hoped for outcome (the science of healthcare improvement). There are more similarities between our work than differences, and we hope this is the beginning of a discussion that can harmonize our science and decrease confusion.
Louise Davies - For the SQUIRE Leadership Team
David P. Stevens
1. Ogrinc G, Davies L, Goodman D, Batalden P, Davidoff F, Stevens D. SQUIRE 2.0 (Standards for QUality Improvement Reporting Excellence): revised publication guidelines from a detailed consensus process. BMJ quality & safety. Sep 14 2015.
2. Davidoff F, Batalden PB, Stevens DP, Ogrinc GS, Mooney SE, Group SD. Development of the SQUIRE Publication Guidelines: evolution of the SQUIRE project. BMJ quality & safety. Nov 2008;34(11):681-687.
3. Davies L, Batalden P, Davidoff F, Stevens D, Ogrinc G. The SQUIRE Guidelines: an evaluation from the field, 5 years post release. BMJ quality & safety. Dec 2015;24(12):769-775.
4. Davidoff F, Batalden P. Toward stronger evidence on quality improvement. Draft publication guidelines: the beginning of a consensus project. Qual Saf Health Care. Oct 2005;14(5):319-325.
5. Davies L, Donnelly KZ, Goodman DJ, Ogrinc G. Findings from a novel approach to publication guideline revision: user road testing of a draft version of SQUIRE 2.0. BMJ quality & safety. Apr 2016;25(4):265-272.
6. Hoffmann TC, Glasziou PP, Boutron I, et al. Better reporting of interventions: template for intervention description and replication (TIDieR) checklist and guide. BMJ. 2014;348:g1687.
7. Davidoff F, Dixon-Woods M, Leviton L, Michie S. Demystifying theory and its use in improvement. BMJ quality & safety. Jan 23 2015.
8. Portela MC, Pronovost PJ, Woodcock T, Carter P, Dixon-Woods M. How to study improvement interventions: a brief overview of possible study types. BMJ quality & safety. May 2015;24(5):325-336.
9. Moss F, Thompson R. A new structure for quality improvement reports. Quality in health care : QHC. Jun 1999;8(2):76.
Competing interests: No competing interests