Adoption and Implementation of Policies to Support Preventive Dentistry Initiatives for Physicians: A National Survey of Medicaid Programs

Abstract

Objectives. We determined the number of state Medicaid programs adopting initiatives to support preventive dental services provision by nondental health care professionals, their perceived attributes, and implementation barriers.

Methods. We used Qualtrics to conduct a cross-sectional survey in 2008 of Medicaid dental program managers to determine organizational stage of adoption classified according to the Transtheoretical Model of Behavior Change with 3-year follow-up. We assessed perceptions of the influence of 18 initiative attributes on the decision to adopt, drawn from Roger’s diffusion of innovations theory. Stage and date of adoption are presented descriptively. Attributes and barriers were analyzed by stage of adoption by using analyses of variance or χ² statistics.

Results. By 2011, 42 states had adopted a policy. Only 9 states included a comprehensive set of preventive services, the most common being fluoride varnish. Adoption was affected by perceived initiative simplicity and its compatibility with other Medicaid programs. Administrative barriers were the most common among the 15 studied.

Conclusions. State Medicaid policies to reimburse nondental providers for preventive dental services are becoming widespread. Interventions are needed to ensure oral health services delivery at the practice level.

Dental caries among preschool-aged children has emerged as a major public health issue,^1–3 with numerous strategies created by state Medicaid programs to alleviate the oral health needs of low-income populations.^4–6 One innovative approach has been to expand the role of primary health care providers in the delivery of preventive dental services, particularly screening for pediatric dental disease, providing fluoride therapy, and counseling parents.^7–9 These programs were developed to increase access to preventive care, reduce the burden of dental disease in very young children, and improve dental referrals based on detection of risk factors or disease.

The American Academy of Pediatrics (AAP) reports implementation of Medicaid preventive oral health programs in medical settings in a number of states.¹⁰ They have intuitive appeal, because most children receive primary medical care very early in life but not dental care.¹¹ Although these programs are based on a strong rationale, little is known about them beyond some limited information regarding the benefits included in insurance coverage.

The purpose of this study was to determine (1) the number of state Medicaid programs with policies supporting the delivery of preventive dental services by nondental primary health care providers, their stage of adoption at the time of a baseline assessment, and subsequent adoption in a 3-year follow-up period; (2) the characteristics and perceived attributes of these initiatives and their relationship to baseline stage of adoption; and (3) the barriers Medicaid programs have experienced in developing and implementing these initiatives. Information on perceptions of initiative attributes held by program managers and barriers to implementation will benefit our understanding about the diffusion of these programs in the United States and ways to improve them.

METHODS

We invited Medicaid dental directors in all states and the District of Columbia to participate in a Web-based survey in 2008 designed to assess the diffusion of preventive dentistry programs in nondental settings. We defined program or initiative as any preventive dentistry activity by the Medicaid program from

adding a single benefit, such as reimbursement of fluoride varnish, to more comprehensive preventive dentistry initiatives that include reimbursement for additional services, such as oral health risk assessments, parental counseling about oral health, or training of physicians.

For those without an initiative at the time of the survey, we determined subsequent adoption of any preventive dentistry innovation between completion of the survey and the end of 2011 by using secondary data sources.

Identification of Sample Frame and Survey Development

We identified and confirmed the sample frame of Medicaid dental directors by using the Centers of Medicare and Medicaid Services directory and the Medicaid–Children’s Health Insurance Program State Dental Association membership list. We resolved discrepancies between the 2 lists by contacting individuals knowledgeable about the states in question.

We used the diffusion of innovation theory as the study framework.¹² This theory explains how a new service, product, or idea is dispersed within a group, and evaluates factors that are related to adoption and implementation. Medicaid policies to support delivery of preventive dental services by physicians is considered an “innovation” in this study.

We used Qualtrics version 11924 (Qualtrics Labs Inc, Provo, UT) to design the Internet-based survey and manage its distribution and collection. For states with a preventive dentistry initiative, the online survey contained 45 questions in 5 domains: status of preventive dental initiatives (familiarity, stage of adoption, attributes), barriers to adoption and implementation (knowledge, provider support, legal issues, administrative concerns), characteristics of the initiative (services, reimbursement amounts, benefit specifications), training of medical providers (requirements, didactic method, content, support materials), and initiative performance and outcomes.

We used Stages of Change, a core construct specified in the Transtheoretical Model (TTM) of Behavior Change,¹³ as a guide for development of our measure of adoption. As theorized in the TTM for the process of individual behavior change, the stages include precontemplation (not thinking about making changes), contemplation (thinking about making changes but not immediately), preparation (planning to make changes within a short time or already making small changes), action (initiated the change recently), and maintenance (maintained the change for a while and working to prevent relapse). Prochaska et al.,^14–16 who developed the TTM, and others¹⁷ have suggested recently that this theory can be applied to organizations to determine their readiness to change, with resulting information being helpful in designing interventions to change organizations.

We classified Medicaid program adoption according to 1 of 6 categories of adoption on the basis of the following question:

Does your state Medicaid program have an initiative in which nondental primary health care providers (e.g., physicians, physicians’ assistants, nurses) deliver preventive dental services in medical care settings?

Those who answered “no” and that they did not intend to implement one in the next 12 months were considered to be in the precontemplation stage (not ready). Those who answered “no,” but intended to implement one in the next 12 months or 6 months were considered to be in the contemplation or preparation stage, respectively. States with initiatives that had been in place for less than or more than 12 months were considered to be in the action and maintenance stages, respectively. We added a final category, relapse, usually not considered a stage itself, but a re-cycling through one of the previous stages, for programs that might have implemented an initiative but discontinued it before the survey release.

Attributes from the diffusion of innovation theory that could contribute to adoption of this initiative by state Medicaid programs included relative advantage (5 items), complexity (3 items), compatibility (3 items), observability (4 items), and trialability (3 items.¹² Content of these 18 items was based on first-hand knowledge of a long-standing preventive dental program developed for nondental health care providers in North Carolina,¹⁸ review of the literature, and descriptions of these initiatives that have been made at scientific meetings. The format for these items was based on an instrument developed by Moore and Benbasat¹⁹ for a study of adoption of an information technology innovation. Items were rated by respondents through a 5-point scale ranging from “strongly influenced the decision” to “did not influence the decision” to implement the innovation. We included a “don’t know” response option.

We included 15 items in the questionnaire that asked respondents the extent to which each factor affected their ability to implement a preventive dental program. Development of these items was based on 3 studies of barriers to the adoption and implementation of preventive dental services reported by primary medical care providers.^8,20,21 Items in the other 3 domains included single closed, multiple closed, and open-ended options. Dental services that could be part of a preventive dental initiative were included as response options for some questions. The options were based on recommended guidelines for physicians' involvement in oral health practices.^22–25 We pilot tested the questionnaire for clarity with 8 individuals, including 4 Medicaid dental managers, and refined it on the basis of this feedback before dissemination.

Data Collection and Analysis

We sent an invitation soliciting the participation of dental program managers via e-mail directing respondents to a secure Web site at the University of North Carolina at Chapel Hill where they completed the electronic questionnaire by using Qualtrics. We distributed weekly follow-up e-mails to nonrespondents beginning the week after the initial invitation for a maximum of 7 cycles. After the seventh cycle, we made a 1-time follow-up phone call to nonrespondents who could be contacted, followed by a copy of the survey as an attachment to an e-mail if they could not be contacted by phone. We determined adoption of a preventive dentistry initiative for nonrespondents to the baseline survey and during the follow-up period through an Internet search of state Medicaid and health department Web sites.

We downloaded survey responses and merged them with another database of follow-up adoption status. Information on stage and date of adoption is presented descriptively. We tested differences in dental program managers’ perceptions of attributes and barriers according to stage of adoption of preventive dental initiatives at baseline by using analyses of variance or χ² statistics depending on whether the variable was continuous or categorical. We also tested the association of baseline readiness to change and adoption during the follow-up period. We completed all analyses with SAS software, version 9.2 (SAS Institute, Cary, NC).

RESULTS

Respondents in 48 of the 51 states (response rate = 94%) completed questionnaires between December 2008 and April 2009. Forty surveys were completed online and 8 were self-administered by using the abbreviated questionnaire distributed by e-mail. Of respondents, 10.9% had been employed in their current Medicaid position for less than 1 year, 23.9% for 1 to 2 years, 21.7% for 3 to 4 years, 30.4% for 5 to 9 years, and 13.0% for 10 or more years. We confirmed baseline stage of adoption for the 3 states that did not complete the survey by using information available online or by telephone communication with Medicaid staff. We obtained 3-year follow-up information on implementation of a preventive dental initiative for those states that had not adopted one at the time of the baseline survey for all states from online resources.

Adoption of Innovation and Readiness to Change

Twenty-nine (57%) states had an oral health initiative with nondental primary health care providers at the time of the survey (Table 1; Figure 1). Another 9 states (18%) planned to start an initiative. A total of 42 states (82%) had implemented a program by the end of 2011, most (56%) doing so in 2007 through 2009. The percentage of programs that implemented an initiative during the follow-up period among those with plans at baseline (77.7%) was greater than the percentage without plans (46.1%), but the difference was not statistically significant (2-tailed P = .231, Fisher’s exact test).

TABLE 1—

Stage of Adoption of Oral Health Initiatives in Medicaid Programs in 2008 and Subsequent Implementation: United States

State	Baseline Stage of Adoption in 2008			Implemented 2009–2011		Date of Implementation
Initiative in Place	Plans to Implement in 12 Months	No Plans to Be Implemented in 12 Months	Within 12 Months	After 12 Months
Alabama		X		X		Jan 2009
Alaska		X			X	Jul 2010
Arizona			X			…a
Arkansas			X			…a
California	X					Jun 2006
Colorado		X		X		Jul 2009
Connecticut	X					Nov 2008
Delaware			X			…a
District of Columbia		X				…a
Florida	X					Apr 2008
Georgia			X		X	Aug 2010
Hawaii		X				…a
Idaho	X					Nov 2002
Illinois	X					Jul 2007b
Indiana			X			…a
Iowa	X					Jan 2001
Kansas	X					Sep 2005
Kentucky	X					Jul 2007
Louisiana			X		X	Dec 2011
Maine	X					Sep 2008
Maryland		X		X		Jul 2009
Massachusetts	X					Oct 2008
Michigan	X					Jan 2008
Minnesota	X					Mar 2003
Mississippi			X		X	Jul 2010
Missouri	X					Nov 2008
Montana	X					Oct 2008
Nebraska		X		X		Apr 2009
Nevada	X					Jan 2007
New Hampshire			X			…a
New Jersey			X			…a
New Mexico			X	X		Jul 2009
New York			X	X		Oct 2009
North Carolina	X					Feb 2000
North Dakota	X					Jan 2008
Ohio	X					Jul 2006
Oklahoma		X			X	Jul 2011
Oregon	X					Jul 2002
Pennsylvania		X			X	Apr 2010
Rhode Island	X					Nov 2008
South Carolina	X					Aug 2007
South Dakota	X					Jan 2007
Tennessee			X		X	Jul 2011
Texas	X					Sep 2008
Utah	X					Oct 2006
Vermont	X					Jan 2008
Virginia	X					Jan 2008
Washington	X					Apr 1998
West Virginia			X			…a
Wisconsin	X					Feb 2004
Wyoming	X					Jan 2007

^aNo initiative implemented.

^bJuly implementation month assumed.

FIGURE 1—

Cumulative percentage adoption curve for fluoride varnish: state Medicaid programs, United States, 1998–2011.

Note. Nonadopting states were AZ, AR, DC, DE, HI, IN, NH, NJ, and WV.

All but 1 responding state of those with complete questionnaire information (n = 23) included fluoride varnish applications as part of its identified preventive dental initiative. Most respondents also included referral for follow-up (n = 16; 70%) and counseling on oral health (n = 14; 61%). Fewer than half of the states reported that their initiative provided for clinical screening and risk assessments (n = 10; 43%). Only 9 (39%) states indicated initiatives with a comprehensive set of services (i.e., oral evaluation, referral for follow-up, counseling, and fluoride varnish). Prescription of dietary fluoride supplements was identified as a component of the initiative by 5 states (22%). Infants and toddlers up to 36 or 42 months of age were eligible for services in half of the states, children up to 6 years of age in 2 states, and all children eligible for State Children’s Health Insurance Program or Medicaid services in 8 states.

All but 6 states included an oral evaluation or counseling along with fluoride varnish application as part of their initiative, but most (n = 18; 78.2%) only reimbursed for fluoride varnish. The reimbursement rate for fluoride varnish ranged from $12.00 to $53.30 (mean = $24.11). The total reimbursement for all services for the 4 states that reimburse for an oral health evaluation or counseling at the same visit as fluoride varnish ranged from $45.00 to $74.68 (mean = $56.96). One state did not include fluoride varnish in its initiative but reimbursed $39.00 for an oral evaluation and counseling.

Eight (35%) of the states have guidelines that provide recommendations to providers for delivering preventive dental services. These guidelines included forms to document results of an oral evaluation (n = 6; 26%), risk assessment (n = 2; 8%), or referral to a dentist (n = 7; 30%); patient educational materials (n = 4; 17%); and use of an artificial light source (n = 1) or mouth mirror (n = 2) for conducting the oral evaluation. One third of states require the physician to conduct the oral evaluation and 11 (50%) require provider training before reimbursement. Few states provided any information for the section of the questionnaire requesting estimates for the number of participating providers (1 respondent; n = 75), practices (2 respondents; min = 75; max = 450) or children receiving services (7 respondents; min = 9474; max = 385 592).

Perceived Attributes of Innovation

The influence of perceived attributes of the innovation on the decision to adopt a program differed by domain (Table 2). The mean score for 4 of the 5 attributes in the “relative advantage” domain fell in the “moderate” to “strong” category (overall score ≥ 3.0). One item from “complexity,” “compatibility,” and “observability” fell in this range, but the only negatively worded item in the 18 attributes suggests that program complexity did not have a strong influence on adoption decisions (mean score = 1.44 for “program too complex to implement”). The ability to test the initiative before wide-scale adoption did not seem to be an important attribute to Medicaid managers (mean score = 1.54 for “we could try the program … without fully committing to it”).

TABLE 2—

Influence of Individual Attributes on Oral Health Initiative Adoption Decision, by Program Adoption Status at Baseline: United States, 2008–2009

Attributes by Domain Category	Overall (n = 42), Mean Likert Scale Score	Existing (n = 25), Mean Likert Scale Score	Plans (n = 9), Mean Likert Scale Score	None (n = 8), Mean Likert Scale Score
Relative advantagea
The program is effective in improving the oral health of children in Medicaid	3.68	3.80	3.63	3.38
The program is effective in improving access to preventive dental services	3.52	3.64	3.56	3.13
The program is based on the latest evidence for caries prevention	3.48	3.60	3.22	3.38
The program would improve the quality of preventive dental services provided for children in our Medicaid program	3.37	3.42	3.44	3.13
The program would have a small impact on the Medicaid budget	2.90	3.00	2.38	3.13
Overall	3.39	3.49	3.11	3.56
Complexityb
The dental procedures are easy for medical providers to learn	3.00	3.40	2.44	2.38**
The procedures are easy for medical providers to integrate into their practices	2.83	3.28	2.22	2.13**
The program was too complex to implement	1.44	1.13	1.22	2.63**
Overall	2.44	2.63	2.22	2.00
Compatibilityc
The program fits with our organization’s mission, goals, or practices	3.36	3.52	3.56	2.63*
The program is compatible with other dental programs that we have for children	2.65	2.84	1.75	3.00*
We did not have to make many changes in our organization	2.25	2.38	1.78	2.43
Overall	2.75	2.93	2.54	2.33
Trialabilityd
Providers could easily try the intervention and stop it if they did not like it	1.89	2.18	1.22	1.83
For us, committing to the program was not an all-or-nothing decision	1.86	2.00	1.25	2.20
We could try the program in local areas of the state without fully committing to it	1.54	1.46	1.50	1.86
Overall	1.79	1.87	1.36	2.33
Observabilitye
Improvements in access to preventive dental services resulting from the program could be observed	3.18	3.26	3.25	2.86
The results of the program would be apparent to me and others in Medicaid	2.93	3.08	2.78	2.57
The results of the program would be apparent to policymakers and health care leaders in the state	2.93	3.08	2.78	2.57
Satisfaction of providers could be easily gauged	2.29	2.30	2.50	2.00
Overall	2.82	2.92	2.68	2.60
Overall mean score	2.75	2.87	2.48	2.66

Note. Mean of Likert scores: 1 = did not influence decision; 2 = slightly influenced decision; 3 = moderately influenced decision; 4 = strongly influenced decision. “Don’t know” response excluded from mean scores. Minimum, maximum % “don’t know” by stage of adoption: existing program = 5%, 13%; plans = 0%, 7%; none = 20%, 40%. Item order within domains from largest to smallest influence on adoption.

^aThe degree to which an innovation is more advantageous than current practice.

^bMeasures whether the innovation is difficult to understand or is complicated to use.

^cSeeks to find out whether the innovation fits with the adopter’s existing values, past experiences, and needs.

^dMeasures whether the results are visible to others; it also explores the outcomes of having the new innovation.

^eMeasures whether the innovation can be tested or tried on a limited basis before wide-scale adoption.

*P < .05; **P < .01 (statistical significance for analyses of variance comparison of means by adoption status).

Mean scores by baseline adoption status differed at statistically significant levels for all attributes in the “complexity” domain and 2 of the 3 items in the “compatibility” domain. The perceived ease with which medical providers could learn the dental procedures or integrate them into their practice was less important to those who had no plans to adopt than those who had already made the decision to adopt. They also were influenced less by the fit of the initiative with their organization’s mission, goals, and practices. Those without plans were influenced more by the complexity of the program than those who had already adopted or were planning to do so.

Barriers to Implementation of Innovation

The extent to which each of 15 barriers was thought by survey respondents to affect program implementation is displayed in Table 3. The most commonly reported barriers were administrative, including limited Medicaid budgets (80%), concerns about reimbursement codes (78%) and claims forms (66%), and lack of personnel (66%). More than half of states reported lack of interest from physicians (61%) and opposition from the dental profession (54%) as barriers to implementation. In general, Medicaid programs that already had an initiative or plans for one had lower mean Likert scores for the extent to which respondents considered an item to be a barrier than did programs with no plans. But with the exception of 1 barrier (required changes in administrative rules or state plan), scores did not differ at a statistically significant level.

TABLE 3—

Percentage of States Reporting Factor to Be a Barrier to Oral Health Initiative Implementation and Mean Extent Score Among Those Reporting a Barrier, by Stage of Adoption at Baseline: United States, 2008–2009

Barrier	Overall (n = 40),a % (Mean Likert Scale Scoreb)	Existing Program (n = 26), % (Mean Likert Scale Scoreb)	Plans for Program (n = 9), % (Mean Likert Scale Scoreb)	No Plans (n = 5), % (Mean Likert Scale Scoreb)
Limited budget to add new benefits	80 (2.1)	68 (1.9)	100 (1.8)	100 (3.0)
Concerns about reimbursement codes	78 (1.7)	74 (1.6)	89 (1.8)	80 (2.0)
Concerns about type of claim form to use	66 (1.6)	58 (1.4)	89 (1.8)	60 (2.0)
Lack of personnel to implement or oversee program	66 (1.8)	67 (1.8)	78 (1.4)	40 (3.0)
Low reimbursement rates for preventive dental procedures	63 (1.9)	71 (1.9)	50 (1.8)	33 (2.0)
Lack of interest from physicians	61 (2.0)	64 (2.1)	71 (1.4)	25 (3.0)
Opposition from organized dentistry	54 (1.5)	58 (1.6)	50 (1.3)	40 (1.5)
Lack of access to information about starting a program	50 (1.7)	52 (1.5)	50 (1.8)	40 (2.5)
Required changes in Medicaid administrative rules or state plan	47 (1.7)	32 (1.4)	88 (1.7)	60 (2.7)**
Need to update computer systems to accommodate preventive dental program	43 (1.8)	35 (1.4)	67 (2.0)	40 (3.0)
Medical practice act issues	40 (1.6)	48 (1.5)	25 (1.5)	20 (3.0)
No carve out for preventive dentistry program	35 (1.5)	30 (1.1)	33 (1.3)	60 (2.3)
Lack of interest from medical societies	35 (1.8)	38 (1.9)	43 (1.7)	0 (NA)
Required off-label use of fluoride varnish	37 (1.6)	41 (1.3)	13 (3.0)	60 (2.0)
Lack of support from Medicaid administration	29 (2.2)	24 (2.0)	25 (2.5)	60 (2.3)

Note. NA = not applicable.

^aBarriers ranked according to overall percentage.

^bMean of Likert score for each item: 1 = not a barrier at all; 2 = somewhat of a barrier; 3 = very much a barrier.

**P ≤ .01 for all comparisons by adoption status within domains and overall with the Fisher exact χ² test.

DISCUSSION

The adoption of state Medicaid policies to support the provision of preventive dentistry services by medical providers is becoming widespread. At the time of this survey in 2008, approximately 75% of Medicaid programs had established an initiative or were planning one within the next 12 months. By the end of the follow-up period in 2011, all but 9 states were reimbursing medical providers for the delivery of preventive dental services. In January 2012 the West Virginia Medicaid program became the 43rd state program to reimburse primary care providers for the provision of preventive dental services.²⁶

The rate of adoption resembles an S-curve, with a few early and late adopting states, but most falling into the middle years (Figure 1). Although highly variable, the translation of research evidence into clinical practice can take an estimated 17 years on average.^27,28 The first state Medicaid program began reimbursing primary care providers for fluoride varnish in 1998, 4 years after the Food and Drug Administration approved it for use in the United States. A number of organizations began promoting the integration of medicine and dentistry after the US Surgeon General called for this strategy in 2000.^{23–25,29–32} By 2008 preventive oral health had become recommended practice in pediatric primary medical care by the AAP.²⁴ In 2011 the National Quality Forum included fluoride varnish application during primary child care visits as 1 of 43 national performance measures.³³ Reimbursement of fluoride varnish application by primary medical care providers in the state’s Medicaid program was 1 of 8 benchmarks recently used by the Pew Foundation to evaluate children’s dental health in each state.³⁴

The characteristics reported to be important to adoption were its contribution to the overall mission of the Medicaid program; its effectiveness in improving quality of care, access to care, and oral health; its low complexity; and its observable results, thus providing program accountability. These findings demonstrate the strong commitment of Medicaid program managers to improving oral health services for young children. Several of these attributes belong to the “relative advantage” domain, known to be among the best predictors of diffusion of innovations.^12,35

Our findings and their support in the literature emphasize the importance of having evidence available on the effectiveness of preventive programs delivered in primary care to help facilitate diffusion of this innovation. Evidence is beginning to accumulate about improvements in access to preventive services,^36,37 identification of children needing referral,³⁸ reductions in caries treatments,³⁹ better oral health status,⁴⁰ and averted hospitalizations⁴¹ resulting from these programs. But stronger evidence is needed to help ensure continued expansion and enhancement of these programs because most of the evidence generated so far is from observational studies.

Implementation of a reimbursement policy in the Medicaid program does not ensure adoption and implementation among medical practices, and rates generally are low for the few programs that have reported that information. One year after implementing a policy in Massachusetts in 2008, only 5% of physicians self-reported providing fluoride varnish.⁴² Washington, the first state to implement a policy, reported 145 visits in 2000.²⁰ A subsequent demonstration in 2007 resulted in oral health services for 44% of children with well-child visits in 6 practices in Group Health Cooperative over 16 months. Wisconsin reported 17% of enrolled children aged 1 to 6 years of age had a visit in the 3 years after implementation of its policy in 2004.³⁷ North Carolina reported that 18% to 39% of enrolled 6- to 35-month-old children, depending on age, had an oral health visit in 2006, 7 years after implementation of its policy to reimburse nondental providers for the delivery of preventive dental services.³⁶

Implications for Practice-Level Dissemination and Implementation

Results from our study combined with practice-level barriers to adoption and implementation reported in the literature suggest a number of strategies that can be undertaken to increase the number of preventive dental services provided within medical practices once a reimbursement policy is in place within the Medicaid program. Practice-level barriers reported by medical practitioners include difficulties in integrating services into routine practice, staff resistance, and difficulties in locating a dentist for children found to need a dental referral.^8,20,21 These barriers are similar to those reported for the delivery of other preventive services in primary care and, thus, interventions can draw on those interventions proven to be effective for delivery of those services.^43–45

The AAP and other professional organizations recommend that children receive a number of age-specific preventive dental services in medical care settings.^22–25 But we found that many states do not have a comprehensive preventive dentistry program for medical settings and simply added a fluoride varnish benefit. Reimbursement for multiple services including screening and risk assessment, referral, and counseling benefits in addition to fluoride varnish application might help ensure that providers have the time and resources to devote to providing these services.

Although most states have implemented an oral health benefit, this strategy is not promoted widely, nor does the number of participating providers or patient recipients seem to be monitored closely. The attributes of these programs perceived by Medicaid program managers to be favorable, such as their relative advantage and low complexity, and annual estimates of number of recipients can be highlighted in communication with both providers and parents of Medicaid-enrolled children.

Medicaid programs should require training and certification of nondentist providers before allowing reimbursement for their services. The training can include educational meetings or workshops, but evidence suggests that this strategy alone is insufficient to change complex behaviors within medical practices.^46,47 It is more effective when combined with hands-on clinical training mixed with didactic formats, audit and feedback, quality improvement initiatives, or outreach visits. Use of guidelines and associated implementation tools along with recommendations about how to overcome some of the practice-level barriers also can improve outcomes.^48,49

Finally, Medicaid programs can work to overcome dentist workforce shortages, both real and perceived, that prevent some nondentists from providing screening and risk assessment services. States have developed innovative and successful models that can be used as a foundation for future progress. Strategies include increases in Medicaid reimbursement rates for dental services, use of community outreach and care coordination, changes in dental practice acts to allow alternative workforce models, financing innovations such as managed care carve-outs for dentistry, dentist training in the care of infants and toddlers, and improvements in the capacity of safety net programs.^32,50,51

Limitations

A major limitation of this study was the potential for response bias. Some respondents might have been asked to recall information about events that occurred before they were employed in their positions. They would not have been involved in developing their state policies or evaluating the dental initiatives’ implementation. Although the relatively early stages of diffusion of the innovation in most states helps to alleviate this concern, lack of knowledge about the initiative might have led respondents to choose the “don’t know” option.

Conclusions

A number of organizations support the delivery of preventive dental services by nondental health care providers as one way to improve access to oral health care for children. We found that a large number of state Medicaid programs are working to expand the practice of medicine to help address dental caries, and demonstrated a strong commitment to improving access to dental care for low-income children. These Medicaid programs provide an opportunity to increase access to basic oral health services. The next phase in their development is to ensure at the organizational level that these Medicaid programs are comprehensive and at the provider and practice levels that services are provided to most eligible children. Effective diffusion of this innovation will require multiple strategies, including monitoring and quality improvement initiatives in service delivery.