To develop a guideline for reports of web-based interventions we broadly followed the standard methodology developed by the CONSORT group, reported in detail elsewhere [14 (link)]. We started the work on CONSORT-EHEALTH in October 2010 with writing a grant proposal requesting funding for a consensus workshop from the Canadian Institutes of Health Research (CIHR). Unfortunately, this funding request was turned down (with some rather odd explanations, such as “[it is unclear] why journal editors [private sector] need funding to complete this project.”) Without funding, our initial plan to use a 3-phase process of premeeting item generation, a meeting with invited stakeholders, and postmeeting consolidation, had to be modified, with only a very short face-to-face workshop in the context of a scientific meeting, and the bulk of the work being conducted through online consultations.
The core international group of CONSORT-EHEALTH contributors included researchers, funders, consumers, journal editors, and industry, listed under acknowledgments. This is (and remains) an open and dynamic group.
In the premeeting item-generation process, we used the current CONSORT guideline items as a framework, and generated additional items and subitems through literature searches, extracting reported items from published RCTs as well as relevant guidelines. We had access to the referee reports of JMIR, which helped us to evaluate which items are frequently pointed out by reviewers as “missing” in the original submissions of the authors. Additional input came from a face-to-face session hosted by the International Society for Research on Internet Interventions (ISRII), in Sydney, Australia on April 6-8, 2011.
A preliminary version of the CONSORT-EHEALTH checklist (V1.5) was published in April 2010. In a web-based Delphi process we gathered some data on the importance of the items [14 (link)]. The initial questionnaire with the list of items is shown inMultimedia Appendix 1 . Participants were asked to suggest missing items (under each CONSORT subheading), and to rate each proposed subitem on a scale of 1-5 (where 1 was “subitem not at all important” and 5 was “essential”).
We kept items as “essential” in the CONSORT-EHEALTH when at least 50% of respondents rated an item as “5-essential”. We downgraded items as “highly recommended” when at least 50% of respondents rated an item as 4 or 5 (but less than 50% said it is “essential”). We eliminated items when less than 50% of respondents answered 4 or 5.
The core international group of CONSORT-EHEALTH contributors included researchers, funders, consumers, journal editors, and industry, listed under acknowledgments. This is (and remains) an open and dynamic group.
In the premeeting item-generation process, we used the current CONSORT guideline items as a framework, and generated additional items and subitems through literature searches, extracting reported items from published RCTs as well as relevant guidelines. We had access to the referee reports of JMIR, which helped us to evaluate which items are frequently pointed out by reviewers as “missing” in the original submissions of the authors. Additional input came from a face-to-face session hosted by the International Society for Research on Internet Interventions (ISRII), in Sydney, Australia on April 6-8, 2011.
A preliminary version of the CONSORT-EHEALTH checklist (V1.5) was published in April 2010. In a web-based Delphi process we gathered some data on the importance of the items [14 (link)]. The initial questionnaire with the list of items is shown in
We kept items as “essential” in the CONSORT-EHEALTH when at least 50% of respondents rated an item as “5-essential”. We downgraded items as “highly recommended” when at least 50% of respondents rated an item as 4 or 5 (but less than 50% said it is “essential”). We eliminated items when less than 50% of respondents answered 4 or 5.
Full text: Click here