The CRISP Working Group completed a series of studies to assess current practices, needs, best practices, and potential for improving PC research reporting (Table 2).14-18,*

CRISP Studies on the Reporting of Primary Care Research That Nominated Potential Reporting Items

We conducted a scoping review to map the published literature on PC research reporting quality, strengths, weaknesses, recommendations, and efforts to improve reporting.14 Our search of 7 major databases for articles published in English during 2000-2020, supplemented by a secondary search of references and expert panel suggestions, yielded 2,847 unique titles, of which 126 underwent full-text review and 25 met predetermined inclusion criteria. All publications identified the need to improve reporting and recommended items to include in reports. Most cited the need for more detailed reporting on the context of study interventions, clinical settings, and health care systems. Most publications endorsed reporting guidelines and recognized the unique needs of PC research reporting.

We conducted a needs assessment of the international, interprofessional community of producers and users of PC research.15 Our online survey, conducted during 2018-2019, yielded 255 respondents across 24 nations, including physicians, scientists, educators, public health professionals, patients, study participants, and community members. Respondents indicated difficulty interpreting, synthesizing, and applying PC research reports “50% or more of the time.” Overall, 37% reported problems using current PC research reports. Regarding specific types of research, 49% reported difficulty for qualitative research, 46% for mixed methods research, and 38% for observational research. The most common problems were synthesizing findings/results (58%) and assessing generalizability (42%). The majority of users wanted richer reporting of theoretical foundations (54%); teams, roles, and organization of care (53%); and patient involvement in the research process (53%). Some described challenges with the reporting of the context of the health care setting; practical details of interventions; patient-clinician and team relationships; and generalizability, applicability, and impact in various PC settings. Respondents nominated a list of potential reporting items for PC research. We concluded that opportunities exist to improve the reporting of PC research to make it more useful for its many users, suggesting a role for new research reporting guidelines specific to PC.

We conducted a second international, interprofessional, online survey in 2019, focused on PC clinicians who provided clinical care to patients for more than one-half of their working week.16 The survey yielded 252 respondents across 29 nations, including 88% physicians, 5% nurses, and 3% physician assistants. Of these practicing clinicians, 33% accessed original reports of PC research in academic journals weekly or daily, but only 36% found reports met their needs “frequently” or “always.” We concluded that PC practitioners read original research reports, but current reports meet their information needs less than one-half the time. Practitioners desired improved reporting of study context, interventions, relationships, generalizability, and implementation. Respondents nominated potential reporting items, adding to our list.

After identifying the needs voiced by survey respondents, we scanned the 14 EQUATOR Network guidelines most relevant to PC research.14 We found that currently published guidelines do not adequately address many of the concerns voiced by our respondents.

In addition, we surveyed editors of 12 major journals publishing PC research in 2020, to collect their recommendations for improving the reporting of PC research.*

Our CRISP Working Group also gathered peer comments on our developing work through presentations, open meetings, and workshops at national and international conferences in Australia, Canada, Europe, and the United States during 2018-2022.†

At each stage of this research, participants nominated potential reporting items. Following a prespecified, iterative analytic plan, our CRISP Working Group reviewed new items to add to the growing list. Some potential items were combined, split into separate items, or reworded for clarity and interpretation across nations, health care systems, and PC settings. The revised aggregate list of items was then presented in the following survey. Fewer new items were suggested at each successive stage, suggesting we had reached saturation across the study groups.

To move from this aggregate list to a consensus set of reporting items, we conducted a Delphi study.17 Using a prespecified, published protocol,18 we recruited an international, interdisciplinary, interprofessional Delphi panel of PC researchers and research users for an online survey in 2021. We presented the list of potential reporting items and asked participants to vote whether each item should be included, required, or recommended in a guideline. An item advanced to the next Delphi round if more than 50% of panelists voted to include it. Eighty-nine respondents completed both Round 1 (84% response rate) and Round 2 (91%), with representation of a wide variety of demographic characteristics, health professions, scientific disciplines, research roles, levels of experience, and world regions. Round 1 presented 29 potential items, of which 25 moved on to Round 2. After the 2 rounds, 11 items were endorsed for inclusion by at least 90% of panelists, and an additional 12 items were endorsed by more than 50%. The Delphi study thereby identified a consensus set of items with broad endorsement from the worldwide community of producers and users of PC research.

On the basis of the Delphi results and earlier studies, the CRISP Working Group made final edits for clarity and broad understanding to produce the draft CRISP Checklist with 24 reporting items.