Geographic and health system correlates of interprofessional oral health practice

Sean G. Boynes , Abigail Lauer , Amy Martin

1. Office of Interprofessional Practice, The DentaQuest Institute,Westborough, MA, USA

2. Department of Public Health Services, Medical University of South Carolina, Charleston, SC,USA

3. Division of Population Oral Health, Medical University of South Carolina, Charleston, SC,USA

Introduction

The case for interprofessional oral health care to address the epidemic of early childhood caries has been well litigated.The evidence supports the clinical and cost-effectiveness of primary care– based oral health risk assessments and fluoride varnish applications [ 1]. Many studies have demonstrated the successful adoption of interprofessional oral health practices at the clinic, system [ 2], and state [ 3, 4] levels. While the personal attributes and opinions of individual physicians and pediatricians have been explored [ 5, 6], few studies have examined the organization and system characteristics that enable or facilitate use of fluoride varnish or oral health risk assessments.Physicians have communicated they are more likely to provide fluoride varnish if they have proper training, have access to patient data to determine eligibility, have access to dentists to whom they can refer patients, and are engaged by leadership in the decision to provide the services [ 7]. The business case for oral health interprofessional practice was presented in a single state study that demonstrated its cost-efficiency albeit with diminishing returns depending on reimbursement levels [ 8]. Given the depth of knowledge of individual provider motivation, we examined correlates of interprofessional oral health practice at organizational and health system levels,including geographic characteristics. We specifically asked the following research questions: What are the geographic and organizational attributes of practices that provide oral health interprofessional care ( fluoride varnish and risk assessment)?Are they more likely to have established referral systems for dental care and if so, how are they managed?

Materials and methods

The research design centered on a cross-sectional survey of health care professionals, including clinical, administrative,and support team members from medical and dental disciplines. Respondents attended a training program on interprofessional oral health practice delivered by the Director of the DentaQuest Institute’ s Office of Interprofessional Practice at ten meetings across the United States from April through September, 2016. Training was conducted at conferencestyle venues, and not in clinical settings. Those in attendance were trained on the definition of interprofessional oral health practice, examples of how it is practiced, and effective implementation approaches. The training was voluntary, and continuing education credits were made available for most health professions. The anonymous survey was administered at the training sites at the conclusion of the program and submitted as the participants exited the venue. Participation was voluntary, and no personal or clinic-identifying information was collected. Thus this represents a convenience cohort of participants in a training program on interprofessional oral health practice and most likely captures health providers most interested and excited to engage in oral health activities.

The survey was developed by the Office of Interprofessional Practice at the DentaQuest Institute and piloted with an eightmember national content expert panel. The panel assisted with operationalizing constructs and survey items to explore the organization and system characteristics that positively encouraged the adoption of interprofessional oral health care practice. The survey asked participants whether physicians in their“ networks or systems [place of primary employment/service]routinely apply fluoride varnish and conduct oral health risk assessments” (dependent variable).

Additional survey questions were asked to capture recommended independent variables, which were organized along geographic and organizational domains. Respondents indicated their organization type such as federally qualified health centers or private practices. Included in the list of organization options was dental service organization, which is a growing trend whereby an entity contracts with dental practices to manage the business-related functions. They also indicated the states in which the practices they primarily worked were located. We operationalized the state variable as the census region of the country (Northeast, South, Midwest, and West). Respondents self-reported if the practices they worked in primarily were in rural, suburban, or urban communities. Respondents represented a wide array of professionals, which created impracticality in adjusting the data for specific provider type. Therefore two provider attributes were used: primary system of employment (medical/dental) and primary role (clinical/leadership/support). System variables were collected through questions that asked respondents’ agreement using a five-item Likert scale(strongly agree, agree, neutral, disagree, strongly disagree) that were operationalized as the following: medical-to-dental referral success; directionality of interprofessional referrals (medical to dental only, bidirectional, and none); and referral method.The referral method was operationalized as electronic transfer/information exchange, warm handoff, all other methods, and no method. Likert responses were organized as agree (strongly agree and agree) or disagree (strongly disagree, disagree, and neutral). Warm handoff has become health care lexicon for ensuring patients get to the right level of care at the right time[ 9]. The study was approved by the Institutional Review Board of the Medical University of South Carolina.

Survey data were examined through bivariate analyses using chi-square tests comparing the dependent variable and each independent variable. Multivariate logistic regression models were used that included all variables that were in the bivariate analysis, and we adjusted the data for a variety of characteristics to determine the effect of specific variables,holding all other variables constant.

Adjusted odds ratios were estimated to identify characteristics associated with a higher likelihood of agreeing with the statement that “ networks or systems routinely apply fluoride varnish and conduct oral health risk assessments” were provided by medical professionals in the respondents’ sites or networks. SAS version 9.4 was used for all statistical analyses [ 10].

Results

A total of 673 training attendees, which included clinical,administrative, and support health care team members, volunteered to complete the survey. One hundred and thirteen were eliminated because of incomplete responses or undecipherable writing. Respondents who indicated they practiced clinically in the military were also excluded. Our original intention was to keep them in the analysis as a subpopulation but we received too few responses from military personnel to include them in the analysis ( n ＜ 5). Anyone not responding to the survey question that captured data for the dependent variable was excluded. Our convenience cohort resulted in 560 completed surveys. Because attendance was not taken, no denominator is available to calculate a response rate, although we speculate most of the attendees completed surveys. In descriptive analysis, most (68.7%) agreed with the following statement:“ Medical providers at our site, or part of our network, are administering fluoride varnish and identifying oral health risk factors in the majority of patients seen.” A description of our cohort is presented in Table 1 along with results from our descriptive analysis that delineates our cohort’ s attributes along independent variables.

Geographically, our study cohort represented 44 states.Table 1 delineates characteristics of respondents and the organizations where they are employed or practice. We had a reasonable representation from each of the four census regions(South [31.7%], Northeast [31.3%], West [20.6%], Midwest[16.4%]) and three levels of rurality (rural [42.9%], suburban[32.1%], urban [25.0%]).

Participants were asked to select the organizational type that best described their current work situation; most of our cohort represented federally qualified health centers and private practices. All organizational types had majority agreement with the statement that medical providers at their site or in their network are administering fluoride varnish and identifying oral health risk factors in the majority of patients seen (68.7%, n = 385).Respondents’ referral system attributes are delineated in Table 2. Of note, a minority of respondents used electronic means for referral facilitation, while the majority disagreed that their electronic health record (EHR) made referral management easy.

In bivariate analysis, organization type ( P = 0.0067), having a successful medical-to-dental referral system ( P ＜ 0.0001),referral method ( P ＜ 0.0001), and EHR ease of use for referrals ( P ＜ 0.0001) all had significant relationships with the dependent variable. Table 3 shows a summary of all bivariate comparisons.

No geographic indicators (census region or rurality) were significant in the multivariate analysis. All referral system indicators were significant. Dependable medical-to-dental referrals ( P ＜ 0.0001), ease of EHR use in making dental referrals ( P = 0.0054), and how referrals were made ( P = 0.0009)were all predictors of the dependent variable. The odds of those who agreed with having a dependable medical-to-dental referral system and also agreed with the dependent variable were 4.5 times greater than for those who disagreed about having a dependable medical-to-dental referral system (odds ratio 4.54, confidence interval 2.79– 7.39). Respondents who agreed/strongly agreed that they had a successful network for medical-to-dental referrals had 4.54 times greater odds of agreeing that most medical providers deliver fluoride varnish and complete risk-based oral health assessments than respondentswho disagreed/strongly disagreed with that statement (odds ratio 2.4, confidence interval 1.29– 4.37). Table 4 summarizes logistical regression and adjusted odds ratio findings.

Table 1. Results from descriptive analyses of the survey respondents attending interprofessional oral health training

Discussion

In a previous analysis completed by the American Dental Association’ s Health Policy Institute, it was reported that a key component of physician dissatisfaction was a lack of adequate sources for dental care referrals [ 11]. Our results corroborate these findings and those of Bernstein et al. [ 12] that physicians are more likely to facilitate interprofessional oral health practice through fluoride varnish applications when an established medical-to-dental referral system is in place. While the importance of treating dental referrals the same as any other specialty referral is explained by the US Department of Health and Human Services in its guidance around interprofessionaloral health practice [ 13], difficulties with achieving this level of referral between medical and dental care teams are frequently reported [ 14, 15].

Table 2. Results from the descriptive analyses of the study cohort’ s referral system or network

Respondents who reported EHR ease of use were 2.4 times more likely to administer fluoride varnish and conduct risk assessments. A recent report from a multistate interprofessional practice collaborative found that when a risk assessment template was embedded within a site’ s EHR, risk-based oral evaluations by medical providers increased [ 14]. A second analysis found that adding oral health provider prompts within an EHR system assisted in standardization and decreasing variation between multiple providers during well child care [ 2]. In an additional analysis of medical care providers’ perspectives on dental information needs in EHRs, the authors reported that respondents recommended “ user-friendly access” to dental information communication was crucial to the delivery of holistic care [ 15].

While the type of referral system used by our respondents varied, use of electronic health information and referral transmission processes for interprofessional practice-related communication was positively related to the administration of fluoride varnish and oral health risk assessment evaluations. Respondents with interprofessional electronic transfer capabilities were nearly two times more likely to administer fluoride varnish and risk assessments. Although their impact on work flow processes, return on investment, and ability for interoperability need further examination, the benefits of electronic referral tools with EHR systems have been reported to result in more timely and improved communications as well as to provide better triage opportunities for provider teams [ 16,17]. A lack of electronic information and referral exchange capability has been associated with oral health referral dissatisfaction by physicians [ 11, 14].

While our study identified organizational- and system-level correlates for the adoption of certain interprofessional oral health practice, it does have limitations. The quality of referrals and patient outcomes were not explored. Respondents provided perceptions and descriptions of interprofessional oral health practices. Our dependent variable was defined with two elements: fluoride varnish and risk assessment. It is possible some respondents operationalized their response to only one of the two-part construct. Participants were not given the opportunity to report the age of their patient populations, and survey questions were not designed to isolate information about the application of fluoride varnish in adult or pediatric care. Given that fluoride varnish application is most often associated with the well-child visit, a missed opportunity to gain knowledge of providers administering oral health care to nonpediatric populations may have occurred. In addition, it did not confirm the impact these practices had on the patients they serve. Data were collected anonymously from respondents who voluntarilyparticipated in the study. As such, we were unable to identify the number of individual practices represented; therefore it is possible that multiple people from a single practice participated in the survey. Given that we collected data across 44 states, we anticipate potential clustering to be minimized.

Table 3. Results from bivariate analysis of dependent and independent variables

Table 4. Results from adjusted odds ratio estimates*

Conclusion

This analysis demonstrated, in this motivated study population, the importance of a dependable medical-to-dental referral systems and processes for fluoride varnish administration and identification of oral health risk factors during primary health care. In addition, the ease of EHR use and the availability of electronic referral mechanisms were found to impact primary care oral health prevention and intervention. Given the findings, further research is warranted to understand the limitations and importance with participation in interprofessional care by dental care providers. More detailed analysis on how interprofessional processes may reduce oral health risk and new cavitation, as well as the actual and perceived impact on patient populations, may also be warranted.

Acknowledgments

The authors thank Dr. Patricia A. Braun from the Department of Pediatric and Family Medicine, University of Colorado School of Medicine, for her thoughtful scholarly contributions to our study.

Author contributions

Sean G. Boynes conceptualized and designed the study and survey instrument, collected primary data, and drafted the initial manuscript. Abigail Lauer designed the statistical analysis,performed the statistical analysis, and reviewed and revised the manuscript. Amy Martin developed the study framework,supervised the statistical analysis, and reviewed and revised the manuscript. All authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.

Financial disclosure

The authors declare that have no financial relationships relevant to this article to disclose.

Conflict of interest

The authors declare that they have no conflict of interest.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-pro fit sectors.

1. Clark MB, Slayton RL. Fluoride use in caries prevention in the primary care setting. Pediatrics 2014;134(3):626– 33.

2. Dooley D, Moultrie NM, Heckman B, Gansky SA, Potter MB,Walsh MM. Oral health prevention and toddler well-child care: routine integration in a safety net system. Pediatrics 2016;137(1):e20143532.

3. Silk H, Sachs Leichner E, Alvarado V, Cote E, Cote S. A multistate initiative to implement pediatric oral health in primary care practice and clinical education. J Public Health Dent 2017.doi:10.1111/jphd.12225. [Epub ahead of print].

4. Veschusio CN, Probst JC, Martin AB, Hardin JW, L Hale N.Impact of South Carolina’ s Medicaid fluoride varnish reimbursement policy on children’ s receipt of fluoride varnish in medical and dental settings. J Public Health Dent 2016;76(4):356– 61.

5. Weatherspoon DJ, Horowitz AM, Kleinman DV. Maryland physicians’ knowledge, opinions, and practices related to dental caries etiology and prevention in children. Pediatr Dent 2016;38(1):61– 7.

6. Quinonez RB, Kranz AM, Lewis CW, Barone L, Bouleter S,O’ Connor KG, et al. Oral health opinions and practices of pediatricians: updated results from a national survey. AcadPediatr 2014;14(6):616– 23.

7. Lewis C, Lynch H, Richardson L. Fluoride varnish use in primary care: what do providers think? Pediatrics 2005;115(1):e69– 76.

8. Stearns SC, Rozier RG, Kranz AM, Pahael BT, Quinonez RB.Cost-effectiveness of preventive oral health care in medical of fices for young Medicaid enrollees. Arch Pediatr Adolesc Med 2012;166(10):945– 51.

9. Hastings & Prince Edward Children and Youth Services Network(HPECYSN). What do “ no wrong door” and “ warm hand-off” really mean? Ontario, Canada: HPECYSN; 2013 [accessed 2017 Nov 6].Available from: www.hpechildrenandyouth.ca/2013/08/what-doesno-wrong-door-and-warm-hand-off-really-mean/.

10. SAS Institute Inc. 2013. SAS/ACCCESS ® 9.4 interface to ADABAS. Cary, SAS Institute Inc.

11. Miloro MB, Vujicic M. Physicians dissatis fied with current referral process to dentists: American Dental Association Health Policy Institute research brief. Chicago, IL: American Dental Association; 2016 [accessed 2017 Jun 9]. Available from: www.ada.org/～/media/ADA/Science%20and%20Research/HPI/Files/HPIBrief_0316_5.pdf.

12. Bernstein J, Gebel C, Vargas C, Geltman P, Walter A, Garcia R,et al. Listening to paediatric primary care nurses: a qualitative study of the potential for interprofessional oral health practice in six federally qualified health centres in Massachusetts and Maryland. BMJ Open 2017;7(3):e014124.

13. United States Department of Health and Human Services. Considerations for oral health integration in primary care practice for children. Rockville: United States Department of Health and Human Services, Health Resources and Service Administration; 2012.

14. Boynes SG, Davis L, Adams G, Mills M, Deutchman M. MORE care: narrowing the rural interprofessional oral health care gap.Boston, MA: DentaQuest Institute; 2017 [accessed 2017 Jun 9].Available from: www.dentaquestinstitute.org/sites/default/ files/Initiating-Rural-IPOHNs_Closing-the-Gap_42717_v3.pdf.

15. Acharya A, Shimpi N, Mahnke A, Mathias R, Ye Z. Medical care providers’ perspectives on dental information needs in electronic health records. J Am Dent Assoc 2017;148(5):328– 37.

16. Gandhi TK, Keating NL, Ditmore M, Kiernan D, Johnson R,Burdick E, et al. Improving referral communication using a referral tool within an electronic medical record. Advances in patient safety: new directions and alternative approaches (Vol. 3:performance and tools). Rockville, MD: Agency for Healthcare Research and Quality (US); 2008.

17. Kim Y, Chen AH, Keith E, Yee HR, Kushel MB. Not perfect, but better: primary care providers’ experiences with electronic referrals in a safety net health system. J Gen Intern Med 2009; 24(5):614– 9.