Barriers and Facilitators to Optimal Fluoride Varnish Application

Abstract

Objective:

National guidelines recommend that all children under age six receive fluoride varnish (FV) in medical settings. However, application rates remain low. This study aimed to update understanding of barriers and facilitators to guideline concordant FV application.

Methods:

We conducted virtual semi-structured interviews with a purposive sample (e.g., FV application rates, geographic location, practice size and type) of pediatric primary care clinicians and medical assistants in Massachusetts between February 1 and June 30, 2022. The Consolidated Framework for Implementation Research served as the study’s theoretical framework and data were analyzed using a modified grounded theory approach.

Results:

Of the 31 participants, 90% identified as White and 81% as female. Major themes, which linked to four CFIR domains, included: Variation in perceived adequacy of reimbursement; Differences in FV application across practice types; Variation in processes, protocols, and priorities; External accountability for quality of care; and Potential levers for change. Important subthemes included challenges for small practices; role of quality measures in delivering guideline-concordant preventive oral healthcare; and desire for preventive care coordination with dentists.

Conclusions:

This study suggests that potential barriers and facilitators to guideline concordant FV application exist at multiple levels that may warrant further study. Examples include testing the effectiveness of quality measures for FV application and testing strategies for implementing consistent processes and protocols for improving FV application rates.

Introduction

Dental caries is one of the most prevalent preventable diseases of childhood.^1,2 Overall, an estimated 23% of U.S. children aged 2 to 5 years have caries and 46% of children in this age group in Latinx and Black populations, despite evidence-based preventive interventions, such as fluoride varnish (FV).^3–5 Dental caries causes pain, school absences, and contributes to lifelong poor oral health.⁶ The U.S. Preventive Services Task Force (USPSTF) recommends that all children under age six receive FV in pediatric primary care settings starting with first tooth eruption, regardless of oral health risk or dental access (Grade B 2014, renewed 2021).^7,8 Professional guidelines, including those from the American Academy of Pediatrics (AAP), recommend application every three to six months.⁹

Although system- and practice-level interventions have improved FV application rates somewhat in pediatric primary care settings, rates remain low.^10–12 Fewer than 20% of respondents to the 2018 AAP Periodic Survey provided FV to children aged three and under.¹³ Furthermore, only an estimated 9.4% of Medicaid-insured¹⁴ and 4.8% of commercially insured eligible children receive FV.¹⁵ Prior studies have found that barriers and facilitators for providing FV in pediatric primary care practices included clinician self-efficacy, competing priorities, and variation in reimbursement by insurers.^16–18 However, less is known about how barriers and facilitators may have changed in recent years following major policy changes and the COVID pandemic. For example, Medicaid programs were typically the only insurers paying for FV during medical visits until the preventive services mandate in the Affordable Care Act (ACA) required nearly all private medical insurers to cover it in 2015.¹⁹ Additionally, the COVID pandemic upended the delivery of medical services - further highlighting the need to determine contemporary barriers and facilitators to providing FV.

This qualitative study aimed to characterize contemporary barriers and facilitators to providing FV in pediatric primary care practices at multiple levels and to identify strategies that may lead to more effective implementation and sustainment of FV use in pediatric primary care settings.

Methods

We conducted virtual semi-structured interviews with a purposive sample of pediatric primary care clinicians and medical assistants in Massachusetts between February 1 and June 30, 2022. We estimated FV rates using publicly available Medicaid data and sampled practices from the top and bottom quartiles. We then used publicly available information to achieve variation in practice size, geographic location, specialty, and practice type within these quartiles to elicit a range of perspectives.

The interview guide (Appendix) was developed using the Consolidated Framework for Implementation Research (CFIR)²⁰ as a conceptual framework for data collection and analysis. The CFIR consists of five domains (Intervention Characteristics, Outer Setting, Inner Setting, Characteristics of Individuals, and Process) with associated constructs that map to factors that can influence implementation at multiple levels (e.g., policy, practice, community, individual). The CFIR was selected as a framework because its domains and constructs can identify factors at multiple levels that may inhibit or promote implementation of evidence-based interventions.²⁰ The interview guide was pre-tested with the first three study participants; data from these interviews were included in the analysis. Changes following pre-testing included moving demographic questions to a pre-interview survey and consolidating some questions to shorten the duration of the interview. The manuscript adheres to the Consolidated Criteria for Reporting Qualitative Research²¹ and was approved by the [redacted] Human Subjects Protection Committee.

E-mail invitations were sent to medical directors, practice owners, or other designees identified through phone calls to practice managers. One trained team member led the virtual interview while another took field notes. Interviews lasted approximately 30 to 60 minutes, were audio-recorded with the participant’s permission, and professionally transcribed verbatim. Participants were given a $100 e-gift certificate in appreciation of their time.

Analysis

Charmaz’s constructivist approach to grounded theory served as the analytic framework because it uses grounded theory methods while recognizing existing literature on a subject.²² We created an a priori codebook using the semi-structured interview guide and the CFIR domains and constructs, allowing for identification of unanticipated themes that might fall outside of the framework. The first several transcripts were reviewed by the primary analytic team (ED, CG, SG, KW) and the codebook was refined based on this review. Each team member then coded these transcripts independently and codes were compared to develop consistency in coding.

Open coding was conducted by paired team members in an iterative process using Dedoose qualitative software (Dedoose Version 9.0.17). Each pair independently coded assigned transcripts and met to resolve discrepancies by consensus; each pair achieved >80% agreement. The full team met weekly to discuss and resolve residual coding questions and refine the code book. Memos were used to identify coding questions and emerging themes and an audit trail of analytic decisions was maintained. Axial coding proceeded as open coding progressed, linking emerging themes to the different levels of potential influence both deductively and inductively, and identifying pertinent themes that fell outside of the CFIR framework. Interviews were conducted until data saturation was reached, defined as no new themes identified across three consecutive interviews with a minimum of 20 interviews.^23,24 A summary of themes and subthemes was sent to all interview participants for review and commentary (member checking).

Reflexive considerations included: SG is a primary care pediatrician and knew some of the participants professionally; ED is the parent of a toddler and had recent personal experience related to pediatric FV application; KW and CG are student research assistants in SG’s research group and may have been less likely to vigorously debate coding decisions given the potential power dynamic.

Results

Interviews were conducted with 31 physicians, nurse practitioners, and medical assistants, at 22 of the 37 (59%) practices contacted. Pediatric, family practice, and combined internal medicine/pediatric practices were included in the sample. Most participants were White (90%), female (81%), and had finished training more than 10 years prior (68%). Four of the practices were resident training sites; additional demographics are in Table 1. All participants felt FV should be applied in pediatric primary care practices, including participants whose practices did not offer FV.

Table 1:

Participant and Practice Characteristics

Participants n=31	%
Years in Practice
<5	28%
6–10	13%
11–20	31%
>20	28%
Gender
Female	81%
Male	19%
Race
White	90%
Black	3%
Other	7%
Role in Practice
Clinician (MD, DO, and NP)	87%
Nurse	6.5%
Medical Assistant	6.5%
Practices n=22*
Type of practice
Private	59%
*Community health center (including federally qualified)	32%
University/college health center	9%
*4 community health centers are also residency training sites
Practice Specialty
Pediatrics	59%
Family Practice (includes Med-Peds)	41%
Size (number of primary care clinicians)
<5	32%
6–10	13%
11–20	32%
>20	23%
Geographic Location (Massachusetts regions)
West	50%
Central	14%
Northeast	4%
Metropolitan West	14%
Southeast	0%
Boston	18%
Estimated percentage of pediatric patients in practice
0–25%	23%
26–50%	9%
51–75%	9%
76–100%	59%
Estimated percentage of pediatric patients insured by Medicaid
0–25%	32%
26–50%	27%
51–75%	14%
76–100%	27%
Offered Fluoride Varnish
Yes	68%
No	32%
Estimated percent of eligible children who receive fluoride varnish for practices that offer it (n=15)
0–25%	13%
26–50%	20%
51–75%	27%
76–100%	40%

Five major themes related to barriers and facilitators to applying FV in pediatric primary care settings emerged from data analysis: Variation in perceived adequacy of reimbursement; Differences in FV application across practice types; Variation in processes, protocols, and priorities; External accountability for quality of care; Potential levers for change. The themes were associated with four CFIR domains/levels of influence (Outer Setting; Inner Setting; Characteristics of Individuals; and Intervention Characteristics) (Figure 1). Illustrative quotes corresponding to each theme are provided in Table 2.

Figure 1: Major Themes and Subthemes.

Table 2:

Illustrative Quotes

Theme	Quote
Variation in perceived adequacy of reimbursement	“I wanted to bring [FV] back at some point… The reason why providers said, ‘I’m not doing this,’ [was] because we were not getting any compensation from doing it [pre-ACA]. ‘I’m taking some of my time disrupting my whole visit, and I’m not seeing anything in return.’" (11–1; physician; 5–10 PCPs; 51–75% MassHealth; community health center) In reference to unpredictability of MassHealth payment rates: “You have an added twist of the state every now and then deciding that for budgetary reasons, their contracted rates [for all services] might need to change. And then they tell you in March that they made the new rates effective as of January retroactively, so you owe them money for people. […] So you can either write us [MassHealth] a check or just see people for free until you pay it off. That was the final straw. I closed to all MassHealth for peds and adults as did the other two [providers] pretty soon after that happened.” (20–1; physician; <5 PCPs; 0–25% MassHealth; privately owned) “Q: What would you consider the single most important factor that might encourage the rest of your practice to get started with the fluoride varnish? A: Resources I think would probably be up there right now, just because we are kind of running. […] We’re hanging in there but definitely could benefit from one or two more [nurses or MAs]. And then money. […] I don’t know the cost of bringing it into the practice or the reimbursement and how that... but those are probably the two things that I think.” (8–1; physician; <5 PCPs; 25–50% MassHealth; privately owned) “Anecdotally I have heard that at times the [insurers] don’t pay the same. I actually have seen that they paid more in some cases but sometimes they’re not paying as much. Early on we had some situations where they weren’t paying and we actually had to connect with the AAP to have the AAP contact that ensure to get them paying. That seems to be a non-issue now, but I don’t know enough about the actual rates.” (2–1; physician; >20 PCPs; 51–75% MassHealth; community health center) “It just seemed like it was something where the practice could do this. Financially, it was advantageous, and it was the right thing clinically, so all of those things lined up. (1–1; physician, 11–20 PCPs; 26–50% MassHealth; privately owned)” “The most important [factor that encourages FV application] is because [we] want to do the right thing and we know it works and patients can benefit from it. […] Number two is some sort of compensation for the practice, and that is in place and it’s paying well.” (11–1; physician; 5–10 PCPs; 51–75% MassHealth; community health center)
Differences in FV application across practice types  Pediatric patient volume  Community health centers	“[FV] still is not happening. … [Those eligible] is a small subset of our population. [Practice name] is not primarily pediatric population.” (19–2; physician; 11–20 PCPs; 0–25% MassHealth; university health center) “[FV] is difficult because it’s another task to have to do. It’s more time. It doesn’t take a lot of time, but it can add up, and it becomes just another administrative burden for me to try to deal with it. … If I had [pediatric] partners, I think it would be easier to kind of share this a little bit and probably put it into better practice. That’s really the bottom line.” (20–1; physician; <5 PCPs; 0–25% MassHealth; privately owned) “The way I frame it is that your kid is here for [their well] child appointment and this is a part of routine care that you know we recommend to prevent cavities… I think [one] thing that make[s] it easy is that there’s a relatively high acceptance rate among our patients of what’s viewed as routine medical procedures.… Because the health center is so much situated in the community and people are so used to everybody they know getting care there. I think there’s a decent level of trust to that we wouldn’t recommend something that wasn’t […] needed or important.” (21–1, physician; 5–10 PCPs; >75% MassHealth; community health center) “We usually start [applying FV] around six months of age and that’s a good time because of the first eruption of teeth. And I mean most of our kids won’t be seeing dentists until I [would] say two years of age, And so we just offer every six months from then on. ” (18–1, physician, 11–20 PCPs; >75% MassHealth; community health center) We started varnishing again… almost a year ago, maybe a little longer… routinely unless the kiddo had some sort of COVID symptoms…and was there for the well-child check, unless…they’re actively coughing and spewing on you. … My practice has been pretty aggressive in COVID, meaning that we have definitely been the outliers in the area because we never stopped seeing kids.” (6–1; physician; 11–20 PCPs; >75% MassHealth; community health center) Q: Can you tell us a little bit about what happened during COVID with fluoride varnish? “I think it totally went downhill… especially after the first six months, we’re just survivors trying to get kids in to get their vaccines, so we can get everyone updated. I think… it was just one of the last things we were thinking of trying to get people up to date [on],.. I think unfortunately it just became on the back burner… I think it’s [FV application rate] still lower. I don’t think we came back to where we had been before the pandemic. (14–1; <5 PCPs; <25% MassHealth; private practice)
Variation in processes, protocols, and priorities Processes and protocols  Electronic health records (EHR)  Consistent processes & workflows  Consistent protocols for training Good communication and teamwork Priorities	“We changed our [EHR] to have quick visit templates that automatically have [FV application] as a standing order checkbox. That upped our rates of fluoride varnish dramatically, like I would say probably tripled it, if we look at the billing data. So, once we had an easy check box that the provider could just check off as they’re ordering all their vaccines as opposed to having to type it in as a separate order. Our fluoride varnish rates went way, way up.” (6–1; physician; 11–20 PCPs; >75% MassHealth; community health center) “We have things called smart sets which populate what a typical physical has in terms of immunizations that they would typically have, […] the counseling codes and fluoride varnish [is] opted in. So if the child refuse[s], I have to check, uncheck it.” (16–2; provider; 11–20 PCPs; 25–50% MassHealth; privately owned) “A lot of our [medical] assistants are very well organized, and they look to the day ahead… what shots am I giving, what ages are eligible [to receive FV]… and they try to plan out their day, too, so they know who they’re going to be talking to about it as well.” (1–2; physician; 11–20 PCPs; 26–50% MassHealth; privately owned) “What makes it easier is making sure [FV supplies are] always in stock, and also being in the routine of it. When we were in the routine of it, it was always happening and there were very few times that I would… check-in with the family and it hadn’t been already asked or done.” (21–2; physician; 5–10 PCPs; >75% MassHealth; community health center) “The [FV] kits are easy to use. That helps. So, you have what you need that’s available to when you do it.” (5–1; physician; >20 PCPs; >75% MassHealth; community health center, community health center) “We have a nursing supervisor at our office. … she oversees the training, but then we have a few seasoned MAs and nurses that do the training. Some people are better at training than others …or feel more comfortable doing it,…. so we have a few of our staff that regularly do the training for our new hires.” (1–2; physician; 11–20 PCPs; 26–50% MassHealth; privately owned) “Some of the new people coming on board… might not be certified [to apply FV], but we teach them. Those of us that are certified, we just automatically teach them what to do. Like I said, everybody has a little bit different process, but the end result is the same.” (13–1; nurse; 11–20 PCPs; 26–50% MassHealth; privately owned) “There’s no real, particular, logistical barriers in our clinic because thankfully, we have a great team to do it. If our clinic maybe didn’t have the staffing, or our staff wasn’t properly trained, or those might be barriers, but thankfully, I really don’t experience any of that. We have a pretty great, great team.” (4–2; nurse practitioner; 11–20 PCPs; >75% MassHealth; community health center) “Dental caries and dental issues [are] a huge problem… Parents aren’t always able to get kids to dental appointments or something happens or whatever. Sometimes they don’t see dentists for several years… So, I think anything we can do as an intervention of primary care especially if it takes like three seconds. It’s something we should do, because having bad teeth leads to other problems… and [FV] just seems like a very easy intervention.” (21–2; physician; 5–10 PCPs; >75% MassHealth; community health center) “I think it’s mostly just our own, busy nature of what we do and just sometimes forgetting-- sometimes just time gets in the way. There are other things… whether it’s a really critically ill patient versus… a lot of issues that are just happening with that family, and teeth are the last thing at that moment, unfortunately, because they have so many other things.” (1–2; physician; 11–20 PCPs; 26–50% MassHealth; privately owned) “When COVID hit… we just stopped doing [FV]. And then since then, it’s actually just a couple weeks ago, I was like, ‘Hey, are we doing fluoride?’ And the staff was like, ‘Oh we don’t have any.’ I was like, ‘What? We don’t have any? …’ So, then we ordered more. …then just this week I was like, ‘Hey, are we doing fluoride varnish?’ Like, ‘We don’t have any.’ I’m like, ‘What?’ So, I went to the person who orders and she’s like, ‘We have it. We just don’t want to do it.’” (21–2; physician; 5–10 PCPs; >75% MassHealth; community health center) “I remember in residency how it really seemed like we were trained in everything except oral health. It was just this black box that didn’t belong to us, and so I think it is nice that we are having some more overlap and awareness and ownership of making sure that’s an important part for kids to keep healthy too. So, I feel really lucky to be at a practice that really tries to stay at the forefront of things as they evolve and as we learn more about just in general how to keep kids healthy, that we really try to provide whatever is available to our families. We try to be… early adopters of new resources like that. So that’s been great for our families and they’re appreciative as well.” (16–1; physician; 11–20 PCPs; 26–50% MassHealth; privately owned)
	“One of the founders of our practice was …the one who was instrumental in getting fluoride in the water supply in [practice location] many, many years ago. So, fluoride has always been one of those things that we’ve kind of talked about and you know, it’s kind of been around in discussions in our office for prevention. When we heard this discussion about…fluoride varnish, it obviously sounded like a really great idea.” (1–2; physician; 11–20 PCPs; 26–50% MassHealth; privately owned) “That was my request when I joined the practice--they had to be willing to apply fluoride varnish for me to join.” (3–1; physician; 5–10 PCPs; 26–50% MassHealth; privately owned) “Good evidence [shows] that [FV] really has had an impact in preventing early childhood caries. We do try to stay up to date on best practices and evidence-based medicine, and when there’s something that has a clear benefit, then we try to get on board and get that into our system as quickly as we can.” (16–1; physician; 11–20 PCPs; 26–50% MassHealth; privately owned) “I want to say like one of the big factors that made the doctors decide to start doing it is the fact that like, some people don’t have access to the dentist… they don’t have a ride or they’re low income and they just can’t get to one for whatever reason. That was a big step of why we chose to do it.” (13–2; medical assistant; 11–20 PCPs; 26–50% MassHealth; privately owned)
External accountability for quality of care	“[Our membership in an independent physician organization] encouraged [FV]. I think it was a positive relationship. They certainly encourage people to do it.” (1–2; physician; 11–20 PCPs; 26–50% MassHealth; privately owned) “I would say being part of [independent physician organization] has really spearheaded us doing [FV] right because: A, it’s a measure, two, they’ve given us information about it, three, they’ve made it so that it pops up on EPIC so it’s in your face every moment.” (12–1; physician; 11–20 PCPS; 0–25% MassHealth; privately owned) “Well it’s risky, actually because [Organization name] is self-insured, it’s in a bubble and actually at this point remarkably there is no data quality measurement that happens, and hasn’t happened in two years, because there’s no accountability because it’s a self-funded shop.” (19–1; physician; <5 PCPs; <25% MassHealth; university/college health center)
Potential levers for change  Quality measures  Oral health education during  training  Interprofessional learning and collaboration with dentists	[In response to a question about FV rates as a quality measure] “There is nothing that drives doctors that are all these good doobies is like giving you the report card. We all want ‘A’ work.” (22–1; physician; <5 PCPs; 0–25% MassHealth; privately owned) “I do think that [FV] is going to be a quality measure in the nearer future for [Practice Name]. … We try really hard to be compliant with our measures. And I think it’s good practice. So, I think if that was the driving force, then it would definitely make an impact.” (17–1; nurse practitioner; 5–10 PCPs; 26–50% MassHealth; privately owned) “I would definitely agree with [adding oral health into training]. I went to [institution name] which has a pretty good established nursing program. […] I feel like [oral health interventions] could be easily applied.” (8–2; nurse practioner; <5 PCPs; 25–50% MassHealth; privately owned) “I feel like some formal thing with new hires even or I think something more formal and routine because there’s a lot of turnovers. So like having it twice a year even or maybe even who knows would be helpful to keep it in everybody’s mind.” (21–2; physician; 5–10 PCPs; >75% MassHealth; community health center) “Sometimes we’ll invite specialists to come to speak with our group during a provider meeting over the lunch hour. […] I think a pediatric dentist would be a great person to have to really review fluoride systemic versus varnish. […] I think there’s also a lot more that they do though than just caries prevention and then [post-op] rehab. I mean we all had families that come in for pre-op visits and then go into the OR for their kids’ dental work. And I would love to know what that process looks like and how they approach that and if it’s child friendly and if child life involved, I think that would be really nice to be able to counsel the families on what to expect.” (16–1; physician; 11–20 PCPs; 26–50% MassHealth; privately owned)

^*Study ID#; role in practice; # of pediatric PCPs in practice; self-estimated % of patients in practice insured by MassHealth; practice type

Variation in perceived adequacy of reimbursement

Although healthcare providers in Massachusetts receive reimbursement for FV application, participants’ perceptions of the adequacy of reimbursement rates varied. In some practices that did not provide FV, participants expressed concerns about the time required to apply FV, potentially inadequate reimbursement by insurers, and effects of future theoretical reduction in reimbursement rates by MassHealth, the Medicaid and Children’s Health Insurance Plan program in Massachusetts. Conversely, some practices that did provide FV noted that, although the main reason they provided FV was because it is an evidence-based preventive practice, the ease of applying FV and insurance reimbursements made it easier to follow the guidelines.

Differences in FV application across practice types

Pediatric patient volume

Some participants in practices that saw both adults and children felt that having a relatively small proportion of pediatric patients influenced organizational decisions about FV. For example, a private practice clinician felt that it would take too much time to implement and sustain a FV program as the only pediatrician in the practice. Participants who practiced in college health centers felt that FV was not a high priority for practice administrators because the relatively lower percentage of young children in the practice meant there was little financial incentive to provide FV despite it being a recommended, evidence-based preventive intervention. Conversely, practices with only pediatric patients were more likely to offer FV as a regular part of their well-child visits.

Community Health Centers

Participants who worked in community health centers were more likely than participants who worked in other settings to describe FV application as an integral aspect of well-child visits. This difference was also noted when discussing the impact of the COVID-19 pandemic on well-child visits and oral healthcare. While all participants reported disruptions to FV application during the COVID-19 pandemic, the extent of disruption and time to return to pre-COVID application rates varied. Participants working in community health centers generally reported that their practice was closed for a short period of time during COVID, if it all, and resumed routine FV application quickly. In comparison, participants affiliated with other practice types reported resuming in-person well child visits later and either had not resumed FV application at pre-pandemic rates or had stopped offering FV completely. Reasons given for not re-starting FV application included concerns about exposure to COVID, even after vaccines and personal protective equipment were widely available, fluctuating rates of infection in the community, and ‘getting out of the routine’.

Variation in processes, protocols, and priorities

Processes and protocols

Participants described their practices’ process for providing FV and reflected on things they felt made it easier and harder to provide this service. Some participants found prompts through the electronic health record (EHR) helpful while others felt the EHR did not help because it could not differentiate children with higher and lower oral health risk. Several practices described consistent processes and workflows for well-child visits that included FV application. Practices with intentional inclusion of FV in workflows, such as a medical assistant always applying FV at the start of the visit, were more likely to report delivering FV to most eligible children. Following a consistent protocol for training new medical assistants, including how to apply FV, was also a process participants saw as a facilitator. Good communication and teamwork, such as having clinicians apply FV when medical assistants were running behind, were also cited as facilitators of FV application. In relation to COVID, having a formal ‘stop’ policy implemented by high-level administration with no formal ‘restart’ policy was seen as a barrier to resuming systematic application.

Priorities

Some participants reported that they were reluctant to apply FV and some perceived resistance from colleagues and staff. Some participants found applying FV to small children’s teeth challenging, which contributed to their reluctance. Some medical assistants felt that they already faced challenges completing other tasks and FV was a lower priority. Conversely, some participants felt that a Massachusetts policy that allows medical assistants to apply FV was the reason why FV application was offered consistently in their practice. Some of these participants felt that their practice’s medical assistants liked being responsible for the FV program because it was something they could do for families that would have a tangible positive effect.

Although all participants believed that FV was an effective intervention and belonged in pediatric primary care settings, variation in guideline concordance was noted. Examples included not offering FV to any patients, offering it only to children one to three years old, or preferentially offering it to children insured by Medicaid. Some reasons given for choices about which children were offered FV included feeling the practice was not ‘set up’ to provide it to all children and that, since it was not done consistently, it was most important to try to give it to the children who were likely to have higher risk for poor oral health or less likely to have a dentist. Some participants were not sure of FV guideline details.

Some participants felt that individual clinicians in their practice facilitated FV application. Examples included feeling that a clinician with a strong public health-orientation was instrumental in initiating and sustaining FV application and that having colleagues who made oral health a priority in the practice or had strong beliefs that FV should be a routine part of a well-child visit were facilitators to offering it to all eligible children.

External accountability for quality of care

Some participants indicated that factors external to the practice influenced FV application. For example, one participant felt that the resources for quality assurance provided by an independent practice organization (IPO), such as arranging FV application training, facilitated FV application. Conversely, some participants believed that being employed by an organization with a self-funded health plan (employer pays claims directly) was a potential barrier to implementing FV. The reason provided for this belief was that organizations with self-funded health plans are not always subject to the same oversight of quality of care that private and government-funded health plans require. FV application rates were not tracked by any of the participants’ practices as a quality measure, and none knew of any external groups, such as insurers, tracking their FV rates.

Potential levers for change

Participants made several suggestions about how to increase FV rates in pediatric primary care practices. Suggestions related to policy, education, and interprofessional training. For example, although participants whose practices were members of relatively new value-based Medicaid Accountable Care Organizations (ACO) did not think that ACO membership had influenced FV application, some believed that including FV application rates as a quality measure linked to payment or having a dashboard that reported practice-level rates would likely increase FV rates.

Some participants felt that better oral health education during training (e.g., residency, nursing school, and medical assistant training) would increase awareness of the importance of oral health and skills for promoting it, including FV application. Some participants said they had been trained to apply FV many years ago and that better access to current training tools for clinical teams might also help improve application rates. One participant recommended using social media to deliver training and to inform parents about oral health. Other recommendations included education about adequacy of reimbursement and ease of ordering supplies and billing.

There was substantial interest amongst participants in developing processes and infrastructure for interprofessional learning and collaboration with dentists related to FV application and oral healthcare in general. One participant described dental care as a ‘black box’ and some noted that communication between dental and pediatric practices was rare except when the pediatric clinician did a pre-operative visit for surgical treatment of caries. They felt the lack of coordination made it difficult to know whether a child had recently received FV from their dentist.

Some participants expressed frustration with conflicting messages parents received from dental providers, such as that the parent should not schedule an appointment until age three. One participant who worked in a federally qualified health center with co-located dentists felt it was beneficial to have dentists available on site to teach clinicians and patients/parents about oral health and to be able to send patients who came to the pediatric primary care practice with a dental issue directly to the dental clinic.

Discussion

This study explored contemporary facilitators and barriers to providing FV in pediatric primary care practices in the context of major policy changes and the COVID pandemic. Some barriers identified previously, such as resistance from staff and clinicians and inadequate training, have persisted and concerns about reimbursement remained for some.^16–18,25 Barriers related to COVID-19, having fewer pediatric patients, and perceived lack of accountability for quality of care, particularly for self-insured health plans, extend the literature on barriers to FV application.

Differences in participants’ perceived adequacy of reimbursement may warrant further exploration. Practices with a smaller percentage of pediatric patients expressed concern about reimbursement for FV more often than pediatric-only practices and fewer offered FV, raising questions about potential variation in receipt of recommended preventive care based on practice type and possible need for more support for this type of practice. This finding is consistent with prior studies that have suggested that pediatric primary care quality may vary based on organizational characteristics.^26,27 Information about reimbursement adequacy as experienced by pediatric practices that provide FV and limited time it takes to apply FV may help to ameliorate financial concerns.

Practice-level processes, protocols, and priorities also presented potential barriers and facilitators to FV application. Participants who were confident that their practice offered FV to a high percentage of their eligible patients also described processes that integrated FV application into other routine preventive interventions offered during well-child visits, suggesting that intentional strategies for implementing and sustaining the intervention might facilitate application for some practices. Variation in clinical guidelines for FV application⁹ may make it less clear to pediatric primary care clinicians which patients should receive FV, potentially contributing to variation in delivery of FV. Practice leadership and culture, such as public health orientation and teamwork, also appeared to influence FV application, consistent with other studies of healthcare quality.^28,29 Interventions to newly implement, sustain, or improve existing processes for delivering FV may benefit from considering these facilitators and barriers to implementation and sustainment as part of intervention design. The lower priority some participants placed on FV application may relate to sharing responsibility for children’s oral health with dental professionals given that some participants did not provide fluoride if they thought a child might have a dentist, or other factors. Although the ways in which clinicians prioritize recommended preventive services during visits have been studied in adult medicine,³⁰ less is known about how clinicians prioritize recommended services during busy well-child visits and may warrant further study.

Participants discussed potential levers for change to increase FV delivery in pediatric primary care settings, some of which may have policy implications. Some suggestions focused on clinician-level interventions, such as improving oral health training for pediatric primary care clinicians and staff, an intervention that has been effective with state-mandated training and non-mandated training pathways.^31,32 Some participants believed that accountability for FV rates such as through quality measures linked to payment, would improve FV rates, identifying a possible opportunity to use quality measures to increase compliance with FV guidelines. A prior study found that FV rates were higher in pediatric primary care practices that had preventive oral health quality improvement processes compared to practices that did not have such processes, supporting this belief.³³ FV rates have increasingly been included in sets of pediatric quality measures, including the 2022 Core Set of Children’s Health Care Quality Measures³⁴ and starting in 2023, Medicaid Accountable Care Organizations in Massachusetts will require FV application as part of the Tier 1 Quality Measures with which all ACOs must comply [communication with ACO practice], providing an opportunity to assess the effectiveness of these quality measures.

The COVID-19 pandemic presented a barrier to providing FV that may persist in some practices despite resumption of usual in-person care. Reduction or elimination of FV application during the pandemic may have been particularly impactful because access to dentists declined during this period.³⁵ Understanding the long-term effect of this gap in FV delivery on children’s oral health will be important as it may lead to increased dental caries, particularly among children in populations with higher oral health risk. Some participants noted it was difficult to resume FV application after it had stopped being part of the workflow, meaning it is possible that practices that stopped applying FV for an extended period may need to re-implement FV practices as if it were a new intervention for the practice.

This study should be considered in the context of its strengths and limitations. We intentionally selected participants from practices with varying characteristics, giving some confidence that a breadth of experiences and perspectives were included in the study. The data used to estimate FV application rates included only Medicaid data, but the sampling strategy achieved variation in self-reported estimated FV rates, including representation of practices that did not apply FV at all (pre-COVID or currently). We interviewed clinicians and medical assistants in Massachusetts; interviews with other people involved in children’s preventive oral healthcare (e.g., parents, insurers, dentists, and dental hygienists) and with clinicians in other states, could provide a more holistic view of FV application in pediatric medical settings. The USPSTF published a Grade B recommendation for fluoride varnish in 2014, the year before the ACA provisions went into effect, resulting in changes in reimbursement for private insurers. It is not possible to determine the effect of these changes independently with this study design, but the study is not intended to test associations or causality in relation to the ACA provisions. The qualitative methods used in the study are not intended to produce generalizable knowledge but can inform development and testing of strategies for implementing FV application in practices that do not currently offer it or interventions to improve existing processes in practices that offer it incompletely.

In conclusion, this study identified novel modifiable barriers and facilitators to providing guideline concordant FV application in pediatric primary care practices that may warrant further study. These studies might include developing and testing multi-level interventions to translate this evidence-based intervention into practice. Some participants felt that policy changes, such as making FV application rates a quality measure, could lead to an improvement in the consistency of guideline-concordant application of FV. Closer examination of differences in FV rates for children whose health plan is administered by a self-insured organization compared to private or Medicaid-sponsored health plans and for children in mixed-age practices with a low percentage of pediatric patients compared to practices with a higher percentage may be warranted. Testing the effectiveness of standardized processes and protocols for FV application and further assessment of the relationship between practice culture and leadership may provide additional insights into barriers and facilitators. Prioritizing development and testing of mechanisms to promote interprofessional communication, learning, and practice for dentists and medical team members may also be warranted.

What’s New:

• Variation in the likelihood of eligible children receiving guideline concordant fluoride varnish application by their primary care provider may be related to a practice’s size, patient age-mix, and provider awareness of reimbursements for fluoride varnish.

• Interprofessional training and improved care coordination with dentists might improve preventive oral health care for children.

• Practice shut-downs, concerns about infection, and loss of routine may have reduced fluoride varnish application rates during the peak of the pandemic and may have lasting negative effects on fluoride varnish delivery.