To identify the essential components within each PCDH characteristic, we used a modified Delphi process to obtain structured feedback and gain consensus among our project’s national advisory committee (NAC). The 51-member NAC was composed of experts and organizational representatives representing dental and medical care providers including small and large private practices, integrated delivery systems, academic health centers, and community health centers; private and public payers; professional dental, medical, and public health associations; health services research, public health and health policy experts; accrediting organizations; and representatives from federal agencies. Recruitment of members to the NAC expert panel was completed prior to the first phase of this project using a purposive, snowball sampling process to ensure appropriate content expertise and stakeholder representation [7]. The modified Delphi method used for Phase 2 was similar to that used in Phase 1 and included an initial email invitation to complete an online survey that asked NAC members to rank the importance of each potential component. The need for successive rounds was based on a priori criteria regarding the median ratings and level of agreement.

A draft list of components was derived from the following sources: NAC member  feedback from the Phase 1 development of the PCDH characteristics, existing PCMH accreditation/recognition tools (e.g., National Committee for Quality Assurance, Accreditation Association for Ambulatory Health Care, Joint Commission),
resources from governmental and non-profit agencies concerned with quality measurement (e.g., National Academy of Medicine, Agency for Healthcare Research and Quality, National Quality Forum), and published literature related to the dental home model [13–15]. Through this process the research team identified 34 components.

A web-based survey was sent to each NAC member by email in April 2017. Reminder emails were sent to nonrespondents at 1 and 2 weeks. The survey presented 34 proposed components to be rated, along with descriptions of each component and/or example measure concepts that could fit within the component. Participants were asked the following question for each proposed component “Thinking about what makes a PCDH [characteristic], to what degree is [identified component] an essential component?” For example: “Thinking about what makes a PCDH accessible, to what degree is timeliness an essential component?” Response options ranged from 1 (Not essential) to 9 (Definitely essential). Rating guidance included that ratings of 7–9 meant essential (include as a component), 4–6 uncertain (needs more discussion), and 1–3 not essential (exclude as a component) with instructions to rate each component on its own merit rather than relative to other components. The rating criteria are described in Table 1
and are the same as those used to identify the PCDH characteristics in Phase 1.

Respondents were encouraged to include their rationale for component ratings through open-ended feedback. Respondents were also asked to suggest additional components.  The research team discussed concerns and suggested changes offered by NAC members and modified the PCDH components to improve clarity and address identified gaps. As component modifications mainly consisted of rewording rather than major conceptual changes, it was determined that an additional Delphi survey round was
not indicated.

Following the RAND Appropriateness Method, the criteria for consensus on component inclusion/exclusion combined a median rating and measure of disagreement
[16]. Components with a median rating of 7–9 without disagreement, were to be included in the model, median rating of 4–6 without disagreement or any other rating with disagreement required additional assessment, and median rating of 1–3 without disagreement were to be excluded from the model. The criterion for disagreement was met if the 30%–70% interpercentile range (IPR) was greater than the interpercentile range adjusted for symmetry (IPRAS). Additional detail on the statistical approach has been published previously [7]. This study was approved by the University of Iowa Institutional Review Board.

Patient Centered Dental Home: Building a Framework for Dental Quality Measurement and Improvement. J of Public Health Dentistry (2021). Available at: https://doi.org/10.1111/jphd.12482