- Email: [email protected]
Early orthodontic treatment as a means to increase access for children enrolled in Medicaid in Washington state
T
R
E
N
D
S
Early orthodontic treatment as a means to increase access for children enrolled in Medicaid in Washington state Gregory J. King, DMD, DMSc; Charles V. Hall, DDS, MSD; Peter Milgrom, DDS; David E. Grembowski, PhD
JADA, Vol. 137
http://jada.ada.org
✷
D
A
✷
A
T
IO N
Background. The authors assessed the likelihood that interceptive orthodontic Medicaid programs C would increase access to care for Washington children. N U I U NG ED A 4 Methods. The authors surveyed 210 Washington RT ICLE state orthodontists, including questions on demographics, attitudes toward early treatment, use of innovations and perceptions of Medicaid. Respondents were either Medicaid participants or nonparticipants. Results. Fifty of 159 respondents were Medicaid participants. Most respondents perceived early orthodontic treatment as beneficial. Medicaid participants were more willing to participate in Medicaid earlytreatment programs, had slightly fewer patients in the “other insurance” category, provided more discounted fees, received more Medicaid inquiries, practiced in rural areas with lower household incomes, reported feeling overworked and experienced fewer Medicaid problems. The principal problem reported with the Medicaid system was low fee reimbursement. Conclusions. Programs offering early orthodontic treatment could increase access. Important barriers would be low fees and unfamiliarity with Medicaid. Practice Implications. Medicaid should design programs aimed at early treatment with reasonable reimbursement and an educational component. Key Words. Access; health care disparities; interceptive orthodontics; early orthodontic treatment; Medicaid. JADA 2006;137:86-94. I
86
A
J
T
A
ABSTRACT CON
ccess to orthodontic services for children from low-income families who are enrolled in Medicaid is limited nationwide. Low participation by orthodontists in Medicaid programs is an important contributing factor. The status of Medicaid-funded orthodontics in Washington state is typical. More than 500,000 children statewide were eligible for dental treatment under the Medicaid program in 1999. Of these, less than 1 percent received any orthodontic care in 1999 (Washington State Department of Social Health Services, written communication, 1999), despite estimates of the incidence of malocclusion being between 15 and 60 percent.1-5 Given that most Medicaid orthodontic programs target the most severe types of malocclusions, these estimates of unmet need may be high, but they remain significant. Approximately onequarter of practicing orthodontists in Washington state participated in the Medicaid program in 1999 but most treated only a few patients enrolled in Medicaid (Washington State Department of Social Health Services, Olympia, Wash., unpublished data, 1999). Ten orthodontists provided approximately 81 percent of the orthodontic treatment statewide, excluding those for patients with cleft lip and palate. Orthodontists may not provide treatment to patients enrolled in
Dr. King is the Moore Riedel professor and the chair, Department of Orthodontics, University of Washington School of Dentistry, Seattle, and an investigator, Northwest/Alaska Center to Reduce Oral Health Disparities, University of Washington, Seattle. Address reprint requests to Dr. King at Department of Orthodontics, University of Washington, School of Dentistry, D569 Health Sciences Center, Box 357446, Seattle, Wash. 98195-7446, e-mail “[email protected]”. Dr. Hall was a graduate student, University of Washington School of Dentistry, when this study was conducted. He now maintains a private practice in orthodontics, Stanwood, Wash. Dr. Milgrom is a professor, Department of Health Services and Dental Public Health Sciences, University of Washington School of Dentistry, Seattle, and is the director, Northwest/Alaska Center to Reduce Oral Health Disparities, University of Washington, Seattle. Dr. Grembowski is a professor, Department of Health Services and Dental Public Health Sciences, University of Washington School of Dentistry, Seattle.
January 2006
Copyright ©2006 American Dental Association. All rights reserved.
T R E N D S
Medicaid because of actual or perceived problems with the Medicaid program and its clients. These problems may include low fee reimbursement, uncooperative patients and high rates of noncompliance. A few studies have examined the perceptions of general dentists and pediatric dentists regarding the treatment of patients enrolled in Medicaid.6-8 Results from these studies show that the most commonly reported deterrent to being a Medicaid participant is low fees. In a study conducted in Washington state, participants reported patient behavioral issues to be of only minor importance.8 Despite both pediatric and general dentists’ reporting dissatisfaction with the allowable fees, they had significant differences in their attitudes toward the Medicaid program. Pediatric dentists were more likely to accept and treat patients enrolled in Medicaid than were general dentists, and they saw treatment of children as more important than did general dentists.7,9 No similar data exist on barriers to Medicaid participation by orthodontists. In addition to low participation by dentists, there are other possible explanations for this disparity between the need and the availability of service for low-income children. For example, there may be fundamental differences in malocclusion patterns in children from low-income families. Cultural, socioeconomic or behavioral differences also may act as barriers to seeking orthodontic treatment. Most publicly funded programs ration orthodontic treatment by limiting it to only the patients with the most severe malocclusions. This approach has profound effects on the level of participation by orthodontists because it selects for patients who will require more resources and who often are at risk of having poorer outcomes even under ideal conditions. This issue is compounded by reimbursement rates of about 60 to 65 percent of the customary fee. An alternative rationing strategy would be to select patients who may be at risk of developing severe malocclusions and intervene early using approaches that have reduced complexity and expense compared with full treatment in the permanent dentition. Orthodontists consider early interventions to be partial treatments that may require a second phase in the permanent dentition to completely alleviate both functional and esthetic problems.10 Orthodontists recognize that the approach to early orthodontic treatment today does not provide finished results and, ideally, should be followed by elective finishing in a
second phase of treatment. These two-phase approaches often are more costly than one-phase treatment in late adolescence. Patients, however, benefit significantly from early treatment, which usually is less complex and costly than secondphase finishing or one-phase treatments during adolescence. As a public health measure, early intervention—without a publicly funded elective second phase of finishing—would provide substantial benefit at a reduced cost per patient compared with full treatment in the permanent dentition. Although most orthodontists may accept this approach, no researchers have examined its usefulness as a means for increasing access to orthodontic services for low-income families. We hypothesize that orthodontists would perceive early orthodontic treatment as a useful means of reducing the severity of malocclusion, that orthodontists who favor early intervention also would favor other innovative orthodontic treatments, that the practice demographic of orthodontists who are Medicaid participants would differ from that of orthodontists who are not, and that these perceptions and practices would predict acceptance of a Medicaid program focusing on early orthodontic treatment in children at risk of developing severe malocclusions. We conducted a study to assess the likelihood that programs focusing on early interception of malocclusions could increase Medicaid participation by orthodontists and, thereby, increase access for low-income families. SUBJECTS, MATERIALS AND METHODS
The institutional review board of the University of Washington approved our study. Study respondents. We identified and surveyed the 210 members of the American Association of Orthodontists who practiced in Washington state in 1999. They represented most of the orthodontists practicing in the state in 1999. Survey. The survey included questions in four general categories that represent the important factors affecting acceptance of a Medicaid program of early orthodontic treatment: practice demographics, attitudes toward early orthodontic treatment, acceptance of innovative orthodontic approaches and perceptions of the Medicaid system. Practice demographics. We designed the first section of the survey to determine if any differences existed between the demographics of Medic-
JADA, Vol. 137 http://jada.ada.org Copyright ©2006 American Dental Association. All rights reserved.
January 2006
87
T R E N D S
aid participants and nonparticipants that could predict acceptance of an early treatment program. We categorized as Medicaid participants only those respondents who treated at least one patient enrolled in Medicaid during the previous year. This section asked for the respondents’ number of years in practice, number of years at their location, total number of patients they started seeing in the past six months, the ZIP code for their main office, the number of patients who they provided discounted fees and who they started seeing in the past six months and the number of Medicaid inquiries in a typical month. We collected the respondent’s ZIP codes so we could assess sociodemographic data such as median family income and population density. First, we determined the county for each ZIP code,11 and then we recorded the median family income12 and population density13 we obtained from 1999 U.S. census data. We also asked respondents what their average fee was for a 24month case, how busy their practice was during the past year and what their practice arrangement was (that is, self-employed, partner, employee). Attitudes toward early orthodontic treatment. We designed this section of the survey to determine if orthodontists have a common vision of the acceptability of early treatment approaches. We asked orthodontists how much benefit patients derived from early treatment and to rank the effectiveness of early treatment for 10 malocclusions on a five-point scale, with 1 being “never effective” and 5 being “always effective.” We designed a subsection to determine if early treatment without a second phase was acceptable. Our rationale for this question stemmed from our belief that an early treatment program under Medicaid may not include funding for a second phase of treatment. Acceptance of innovative orthodontic approaches. We designed the third section of the survey to determine if acceptance of innovative approaches to treatment would predict the likelihood of participation in a proposed new Medicaid program. We asked respondents how many cases involving indirect bonding and how many involving orthognathic surgery they started in the past six months. We also asked them whether they had used self-ligating brackets, self-etching primers, lingual braces or aligners to treat patients. Perceptions of the Medicaid system. We 88
JADA, Vol. 137
http://jada.ada.org
designed the fourth section of the survey to explore how orthodontists felt about the Medicaid system and a proposed program aimed at increasing participation through early treatment of patients. We asked the orthodontists to provide information regarding their patients’ methods of payment during the past six months. Specifically, we asked them what percentages of their patients had Washington Dental Service (WDS) insurance (WDS [Seattle] is the Delta Dental provider in the state), other private insurance, Medicaid, no insurance or other. We listed 10 commonly cited problems with the Medicaid system and patients enrolled in Medicaid. We asked respondents to indicate if they perceived each of these problems to be a “major problem,” “minor problem” or “not a problem” or to respond “don’t know.” We asked this to determine if any of these perceived problems were barriers to participation in Medicaid programs. Study design. Two orthodontists pretested the survey. One had considerable experience with patients enrolled in Medicaid, while the other’s experience with this population was limited. We chose only two orthodontists to avoid any reduction in the sample size or in significantly biasing the sample. One of the orthodontists (the Medicaid participant) was in the process of retiring, so we did not include him in the survey population. We included the other orthodontist in the final survey population. After the pretesting was completed, we revised the survey instrument. We collected data through a mail survey using the Total Design Method.14 First, we sent a letter on University of Washington School of Dentistry stationery that described the study and identified us. Two weeks later, we mailed the survey, a cover letter and a postage-paid return envelope to all 210 members of the American Association of Orthodontists practicing in Washington state. We did not include an incentive with the survey. We mailed a reminder postcard to the nonrespondents six weeks later followed by a replacement copy of the survey after eight more weeks. The total data collection period was 19 weeks. We kept the survey responses confidential and destroyed links between the contact information and the actual questionnaires after we received the surveys. Data analysis. We analyzed the data using Statistical Package for the Social Sciences (Version 8.0, SPSS, Chicago). All of the respondents did not answer every question. Therefore, we cal-
January 2006
Copyright ©2006 American Dental Association. All rights reserved.
T R E N D S
TABLE 1 culated percentages based only on the number of orthodontists who responded Acceptance of proposed Medicaid to a particular question. program and innovative techniques. We compared orthodontists’ percepVARIABLE n NONPARTICIPANTS PARTICIPANTS tions of the benefits of early treatment Attitude Toward and the frequency of presenting twoEarly Orthodontic phase treatment options using 2 × 4 Treatment tables (tables with two columns and four (% Yes [n]) 58.3 (28) 44.4 (48) 156 rows) and χ2 tests. We calculated, Consider ranked and compared the mean 28.6* (14) 13.9* (15) 157 Participate response scores for the respondents’ ratWilling to Adopt ings of the perceived effectiveness of Innovative Techearly treatment for 10 malocclusions niques (% Yes [n]) using χ2 tests. 55.1 (27) 44.0 (48) 158 Self-ligating brackets We separated respondents into two 53.1 (26) 42.2 (46) Self-etching primers 158 groups for each of the six “innovative” orthodontic techniques shown in Table 26.5 (13) 22.9 (25) Lingual braces 158 1, according to whether they used that † 59.2† (29) 42.2 (46) Aligners 158 particular technique. We defined a user 14.7 (16) 16.3 (8) Indirect bonding 158 of indirect bonding and orthognathic surgery techniques as anyone who had 93.5 (101) 93.6 (44) Orthognathic surgery 155 used these techniques during the past * P < .05, Pearson χ test value = 4.825. six months. We defined a user of the † P < .05, Pearson χ test value = 3.909. other techniques as anyone who had ever used that particular technique. We future Medicaid programs, we separated the then compared the differences between Medicaid participants and nonparticipants using individual respondents into two groups based on their mean perceived problem scores, discussed previously. 2 × 2 tables and χ2 tests. We placed respondents with a mean problem We compared demographic and response frequency differences between Medicaid participants score of 0 to 1 in the “low” group and those with a score of 1.1 to 2 in the “high” group. We then comand nonparticipants using a series of 2 × 2, 2 × 3 2 pared their differences in willingness to particiand 2 × 4 tables and χ tests. We calculated pate in the two Medicaid programs using 2 × 3 respondents’ means and standard deviations tables and χ2 tests. Next, we compared the mean (SDs) for the number of patients they offered discounted fees, number of Medicaid inquiries, perceived problem scores for those willing to parnumber of years in practice, number of years at ticipate in each of the programs with the mean location, percentage of each payment method, scores for those who were not willing or were not population density and median family incomes. sure if they would participate. We performed the We analyzed mean differences between particiMann-Whitney test to determine significance. pants and nonparticipants using nonparametric To determine if orthodontists’ use of innovative tests because the data were not distributed techniques is related to willingness to participate normally. in future Medicaid programs, we created an index We also calculated a mean response score for of each respondent’s propensity toward innovaeach of the perceived problems with the Medicaid tion by adding the number of innovative techsystem. We gave the response of “not a problem” a niques each respondent used. Possible scores value of 0, the response of “minor problem” a ranged from 0 to 6. We then separated respondvalue of 1 and the response of “major problem” a ents into high and low innovation groups. We value of 2. We compared mean scores using the analyzed various cutoffs for each group and anaMann-Whitney test. We excluded respondents lyzed the data using 2 × 3 tables and χ2 tests. who chose the “don’t know” response from our RESULTS calculation. To compare perceptions of Medicaid problems Response rate. A total of 159 of 210 surveys with a respondent’s willingness to participate in were completed. One survey was returned 2 2
JADA, Vol. 137 http://jada.ada.org Copyright ©2006 American Dental Association. All rights reserved.
January 2006
89
T R E N D S
respondents had a mean number of years Comparison of respondents’ financial variables. in practice of 15.9 years with no differences FINANCIAL VARIABLE n NONPARTICIPANTS PARTICIPANTS based on participant status. This is consis0† (0) 129 6.1† (15.8) % Patients Enrolled in Medicaid (SD *) tent with the mean of 16.2 (SD = 10.2) years 157 Full Treatment Fee in practice for all 6.4 (7) 14.6 (7) — % < $4,000 (n) American Association of 74.3 (81) 75.0 (36) — % $4,000 > $5,000 (n) Orthodontists members in Washington state. 19.3 (21) 10.4 (5) — % > $5,000 (n) Medicaid participants 36.0 (19.1) 39.6 (15.4) 129 % Private Funding (SD) more often were located 28.2† (15.7) 21.1† (11.0) 129 % Washington Dental in rural counties or Service Insurance (SD) counties with lower pop33.7 (15.2) 31.4 (13.3) 129 % Other Insurance (SD) ulation densities and with lower median 2.3 (6.5) 2.4 (9.1) 129 % Other Funding (SD) family incomes. Med3.01† (5.6) 5.18† (7.0) 140 No. of Discounted Fees icaid participants also (SD) started seeing fewer 1.05 (1.7) 1.16 (1.7) 146 No. of No Fees (SD) new patients per year 4.91‡ (7.1) 11.8‡ (12.4) 140 No. of Medicaid Inquiries but were more likely to Per Month (SD) report feeling over* SD: Standard deviation. worked, while nonpar† P < .05, Mann-Whitney test. ticipants were more ‡ P < .001, Mann-Whitney test. § Not significant, χ test. likely to report feeling busy enough but not overworked. Attitudes toward early orthodontic treatwithout a forwarding address, and one was ment. There were no significant differences returned unfilled because the practice was in the between Medicaid participants and nonparticiprocess of closing. We considered the response pants regarding perceptions of early orthodontic rate of 76 percent (159/208) to be an adequate treatment. A total of 99.4 percent of respondents representation of the orthodontists in said that there was at least some benefit to early Washington state. orthodontic treatment (Table 4, page 92). A total Practice demographics. Fifty of the 159 of 53.9 percent of the respondents reported they respondents (31.4 percent) reported some level of “often” or “always” offer an early treatment or a Medicaid billing during the past six months. A two-phase treatment option to their patients. total of 6.1 percent of patients in these practices The ratings of the effectiveness (on a scale of were enrolled in Medicaid (Table 2). There were 1-5, with 5 representing greater effectiveness) of no differences in the mean fees charged by parearly treatment for 10 malocclusions generally ticipants and nonparticipants. Overall, responwere favorable. Rated as always effective were dents reported that 60.3 percent of their patients anterior crossbite with a shift (4.6, SD = 0.57), had some form of private insurance. Medicaid posterior crossbite with a shift (4.5, SD = 0.58) participants had slightly fewer patients with priand posterior crossbite without a shift (4.0, SD = vate insurance, treated more patients at dis0.83). Early orthodontic treatments rated as modcounted fees and received more than twice as erately effective were anterior crossbite without a many inquiries from patients enrolled in Medshift (3.9, SD = 0.95), Class II with overjet greater icaid compared with nonparticipants. than or equal to 6 millimeters (3.8, SD = 0.58), We saw no differences in practice arrangeClass II with overjet less than 6 mm (3.6, SD = ments between Medicaid participants and non0.79), arch length deficiency less than 4 mm (3.5, participants. Most respondents were selfSD = 1.08) and arch length deficiency 4 to 8 mm employed without partners (Table 3). The TABLE 2
§
2
90
JADA, Vol. 137
http://jada.ada.org
January 2006
Copyright ©2006 American Dental Association. All rights reserved.
T R E N D S
(3.4, SD = 0.84). The TABLE 3 effectiveness of early Comparison of respondents’ environmental variables. orthodontic treatments for arch length defiPARTICIPANTS n NONPARTICIPANTS ENVIRONMENTAL VARIABLE ciency greater than 8 mm (3.0, SD = 1.08) 158 Practice Arrangement and tooth 75.5 (37) 77.1 (84) — % Self-employed without agenesis (3.0, SD = partners (n) 0.89) was rated low. % Self-employed without 6.1 (3) 2.8 (3) — Acceptance of partners but with expense innovative orthosharing (n) dontic techniques. % Self-employed as a partner, 8.2 (4) 11.0 (12) — Medicaid participants sharing income and expenses (n) were more likely to express a willingness to % Employed by another 6.1 (3) 5.5 (6) — orthodontist (n) participate in a Medicaid-sponsored pro% Other (n) 4.1 (2) 3.7 (4) — gram of early orthoDemographics dontic treatment (Table 157 Mean population density (no. 377.9† (296.4) 567.9† (286.8) 1). Both Medicaid parper square mile) (SD*) ticipants and nonpar157 Mean family income ($1,000) 44.0† (10.1) 50.5† (8.3) ticipants were similar (SD) in their use of innova137 Mean no. of case starts (SD) 109.6 (60.8) 117.2 (75.1) tions in orthodontics, with the only difference Mean years in practice (SD) 158 15.8 (10.9) 15.9 (9.8) being that participants 158 Mean years at current location 12.5 (9.9) 12.6 (9.3) reported using aligners (SD) more than did than 153 Busyness nonparticipants. We — % Too busy (n) 0 (0) 2.8 (3) found that orthodon‡ tists’ perceptions of — % Overworked (n) 19.1‡ (9) 6.6 (7) early treatment and — % Not overworked (n) 44.7§ (21) 67.9§ (72) their use of innovative — % Not busy (n) 36.2 (17) 22.6 (24) approaches to treatment did not predict a * SD: Standard deviation. † P < .001, Mann-Whitney test. willingness to partici‡ P < .05, χ test. pate in the proposed Medicaid program. Perceptions of the Medicaid system. Medfor appointments,” “difficulty collecting from Medicaid participants had fewer problems with Medicaid,” “delays in receiving payment” and “patients icaid in only two (“getting billing questions are often late.” Most respondents in both groups answered” and “need for prior authorization”) of considered “unruly or uncooperative patients,” the 10 issues cited in the survey (Table 5, page 93). “loss of coverage during treatment” and “patients Medicaid respondents reported that these were cancel at the last minute” issues considered not to only minor problems, whereas the majority of nonbe significant. participants saw them as major problems. “Fee DISCUSSION reimbursement too low” was the largest problem The data from our study are limited to a single for both participants and nonparticipants. Sevstate, a fact that one should consider when interenty-nine percent of all respondents reported that preting the results. Some aspects of the data may low fees were a major problem (data not shown). be unique to the environment in Washington state Other issues considered to be of some significance (for example, demographics, fees and modes of by both participant groups were, in descending practice) and, therefore, not easily generalizable. order of importance, “patient may fail to show up 2
JADA, Vol. 137 http://jada.ada.org Copyright ©2006 American Dental Association. All rights reserved.
January 2006
91
T R E N D S
TABLE 4
of the level, would likely be different from those who do Respondents’ perceptions of early orthodontic not. treatment.* Practice demographics. PERCEPTION NONPARTICIPANTS (% [n]) PARTICIPANTS (% [n]) Our failure to find differences based on participant status in Beneficial practice arrangement, fees, 55.9 (57) 43.8 (21) A lot number of new cases started, 43.1 (44) 56.3 (27) Some years in practice and practice location indicates that these 1.0 (1) 0 (0) Very little are not important determi0 (0) 0 (0) None nants in the likelihood of a Need for Phase 2 Treatment clinician’s participating in a Medicaid program. However, 10.4 (11) 6.3 (3) Always Medicaid participants pro45.3 (48) 43.8 (21) Often vided more discounted fees 43.4 (46) 50.0 (24) Occasionally than and received twice as many inquiries from patients 0.9 (1) 0 (0) Never enrolled in Medicaid as did * P = nonsignificant, χ test. nonparticipants. The former may be due to a higher level of altruism or greater exposure to low-income famiOther aspects (for example, orthodontists’ perceplies. The latter suggests that reduced access and tions of the usefulness of early treatment and Medreferral patterns tend to funnel patients enrolled icaid) are likely to be more widely applicable. in Medicaid to practitioners who are known to The low level of Medicaid participation in genaccept them. Medicaid participants treated fewer eral by orthodontists in Washington state (approxiprivately insured patients than did nonparticimately 25 percent) and the poor distribution of pants. This may reflect fewer privately insured those providing these services (10 orthodontists patients in the areas in which Medicaid participrovided approximately 81 percent of the orthopants are located. It also was interesting to note dontic treatment statewide) resulted in a low that Medicaid participants tended to feel more average percentage of patients enrolled in Medoverworked than nonparticipants, despite having icaid per practice (6.1 percent). This raises the consimilar caseloads to nonparticipants. This may be cern that the comparisons of participants and nona reflection of differing practice management participants based on acceptance of any patients styles, such as less use of auxiliary personnel. enrolled in Medicaid may have minimal value Our survey shows that orthodontists with pracbecause of the small difference between levels of tices in areas with lower population densities participation in the two groups. An alternative (rural areas) and lower median family incomes approach to this study design would have been to were more likely to treat patients enrolled in Medcompare the 10 orthodontists who provided the icaid than were orthodontists with practices in majority of the treatment with the remainder of more densely populated areas with higher median the orthodontists in the state. This could have profamily incomes. This differs from an earlier report9 vided clearer differences in participation, but it may not have been representative. Because of that found no difference in Medicaid participation these differences, we felt that programs designed rate based on practice location. However, the preto increase access should target the rank and file vious study categorized practices as downtown/city, orthodontists, not the practitioners who already suburban or rural, whereas our study categorized are doing the work. The value of us using a more practices strictly on the basis of county census liberal definition of Medicaid participation in this data. The finding that orthodontists practicing in study was that it provided us with a large enough rural areas treated more patients enrolled in Medgroup of practitioners so we could have some confiicaid has a few potential interpretations. Orthodence that it was representative. Moreover, we felt dontists who choose to practice in rural counties that the characteristics of practitioners who elect may be more community-oriented and may have to participate in the Medicaid program, regardless more inclusive practices.15 Rural environments also 2
92
JADA, Vol. 137
http://jada.ada.org
January 2006
Copyright ©2006 American Dental Association. All rights reserved.
T R E N D S
TABLE 5 may provide more pressure not to deny treatment to patients Rankings of respondents’ perceived problems with enrolled in Medicaid because Medicaid. there would be fewer other NONPARTICIPANTS PARTICIPANTS (MEAN PROBLEM n orthodontists to whom to refer (MEAN RANKING* [SD†]) RANKING* [SD]) these patients. This difference Fee Reimbursement Too 134 1.7 (0.5) 1.8 (0.6) also may relate to fees.9 Our Low data showed that the average 123 1.5 (0.7) fee for rural practices was lower Patients May Fail to 1.4 (0.8) Show Up for than that for urban practices, Appointments 113 1.5 (0.6) and, therefore, the difference 1.4 (0.8) Difficulty Collecting From Medicaid between the Medicaid reim104 1.5 (0.7) bursement and the standard fee Delays in Receiving 1.4 (0.7) Payment may not have been sufficient to 119 1.4 (0.7) influence an orthodontist’s 1.2 (0.8) Patients Are Often Late acceptance of patients. This 119 1.4 (0.7) 1.2 (0.8) Patients Cancel at the finding also may be unique to Last Minute the demographics of Wash112 1.4 (0.7)‡ 1.0 (0.8)‡ Getting Billing Questions ington state and not generalizAnswered able to other states because the 1.3 (0.8)§ 1.0 (0.8)§ 119 Need for Prior rural areas of the state are priAuthorization marily agricultural with large 1.2 (0.8) 1.0 (0.8) 125 Unruly or Uncooperative numbers of low-income migrant Patients farm workers. 1.2 (0.7) 0.9 (0.8) 99 Loss of Coverage During Attitudes toward early Treatment orthodontic treatment. 1.4 (0.4)¶ 1.2 (0.3)¶ 83 Overall Mean Problem Almost all respondents reported Score that there was at least some * Ranking scale: 0 = No problem, 1 = minor problem, 2 = major problem. Respondents who did not answer the question or responded “don’t know” were excluded from the mean calculation. benefit to early orthodontic † SD: Standard deviation. treatment. This finding sup‡ P < .01, Mann-Whitney test = 1010. ports our hypothesis that ortho- ¶§ P < .05, Mann-Whitney test = 1182. P < .05, Mann-Whitney test = 120.5. dontists perceive early orthodontic treatment as a useful means of reducing the severity of malocclusion. The generally favorable perceptions of the effecHowever, about one-half of the respondents always tiveness of early treatment for most malocclusions or often tell patients there is a need for a twosuggest that orthodontists feel comfortable using phase treatment plan, suggesting that many view the approach for a wide variety of conditions. This early treatment as only partial treatment. Our further suggests that a program aimed at early conclusions are supported by frequent reports of orthodontic treatments could affect most types of malocclusions. success of early treatment in reducing the severity Acceptance of innovative orthodontic of several types of malocclusion in the mixed dentiapproaches. For a new program to improve access, tion.16-21 Our failure to find significant differences between Medicaid participants and nonpartician important requirement would be that a substanpants regarding their perceptions of early treatment tial number of nonparticipating orthodontists would effectiveness suggests that negative perceptions of find it appealing enough to become Medicaid particithe effectiveness of early treatment would not be a pants. Our survey failed to show this trend with barrier to acceptance of such a program by orthorespect to early orthodontic treatment. The greater dontists. However, a better understanding of the acceptance of a Medicaid-sponsored program of frequency of elective phase 2 treatments would be early orthodontic treatment by participants sugimportant because orthodontists may be uncomfortgests that practitioners’ familiarity with the Medable with accepting patients for publicly funded icaid system and its clients may be an important phase 1 treatment without assurances that needed prerequisite for such a program to improve access. phase 2 treatment would be available. Educational initiatives designed to familiarize non-
JADA, Vol. 137 http://jada.ada.org Copyright ©2006 American Dental Association. All rights reserved.
January 2006
93
T R E N D S
participants and their staff members with Medicaid and the new program would be essential for these programs to affect access much. Our study failed to show differences in the use of innovative techniques based on participant status, except with respect to aligners. The failure to find associations between orthodontists’ perceptions of early treatment, their use of innovative or new techniques and their willingness to participate in the proposed Medicaid program does not support our hypothesis that openness to innovative orthodontic treatments predicts a willingness to participate in new Medicaid programs. Perceptions of the Medicaid system. Medicaid’s low reimbursement rate is an important problem for both participants and nonparticipants.9,22 However, the impact of low fees on access is unclear, because we do not know if nonparticipating dentists would be willing to treat patients enrolled in Medicaid without increasing fees. An evaluation of the effect of fees on participation by Medicaid participants would be required to confirm this. A program targeting early treatment has the potential to reduce the costs per patient because these procedures tend to be less complex and, therefore, less expensive. For the same orthodontic budget, this approach would permit more patient treatments, but it may not increase the level of participation or access. Bureaucratic hurdles seem to be less of a problem, and the differences between participants and nonparticipants suggest that these hurdles could be overcome by more familiarity with the program. CONCLUSIONS
We found that orthodontists perceive early treatment as acceptable and having wide applicability. Clinicians who are already Medicaid participants would most readily accept a Medicaid-sponsored program of early orthodontic treatment. Therefore, such a program needs to be accompanied by initiatives to educate nonparticipants. We also found that Medicaid participants had fewer patients with private insurance, provided more discounted fees, received more inquiries from patients enrolled in Medicaid, practiced in more rural areas with lower family incomes, were more likely to feel overworked and had fewer problems
94
JADA, Vol. 137
http://jada.ada.org
with the Medicaid system. The most significant problem with Medicaid cited by both participants and nonparticipants was low fee reimbursement. ■ This work was supported, in part, by National Institute of Dental and Craniofacial Research/National Center on Minority Health and Health Disparities/National Institutes of Health grant U54-14254 to the Northwest/Alaska Center to Reduce Oral Health Disparities. Dr. Hall’s work was completed in partial fulfillment of a master of science in dentistry degree in orthodontics. 1. Bowden DE, Davies AP. Inter- and intraexaminer variability in assessment of orthodontic treatment need. Community Dent Oral Epidemiol 1975;3(4):198-200. 2. Brunelle JA, Bhat M, Lipton JA. Prevalence and distribution of selected occlusal characteristics in the US population, 1988-1991. J Dent Res 1996;75(special number):706-13. 3. Helm S, Kreiborg S, Barlebo J, et al. Estimates of orthodontic treatment need in Danish schoolchildren. Community Dent Oral Epidemiol 1975;3(3):136-42. 4. Foster TD, Day AJ. A survey of malocclusion and the need for orthodontic treatment in a Shropshire school population. Br J Orthod 1974;1(3):73-8. 5. Pietila T, Pietila I, Vaataja P. Early screening for orthodontic treatment: differences in assessments made by a consultant orthodontist and three public health dentists. Community Dent Oral Epidemiol 1992;20(4):208-13. 6. Lang WP, Weintraub JA. Comparison of Medicaid and non-Medicaid dental providers. J Public Health Dent 1986;46(4):207-11. 7. Venezie RD, Vann WF Jr. Pediatric dentists’ participation in the North Carolina Medicaid program. Pediatr Dent 1993;15(3):175-81. 8. Milgrom P, Riedy C. Survey of Medicaid child dental services in Washington state: preparation for a marketing program. JADA 1998;129(6):75363. 9. Nainar SM, Tinanoff N. Effect of Medicaid reimbursement rates on children’s access to dental care. Pediatr Dent 1997;19(5):315-6. 10. Kluemper GT, Beeman CS, Hicks EP. Early orthodontic treatment: what are the imperatives? JADA 2000;131(5):613-20. 11. Washington State Department of Ecology. Washington Zip code maps. Available at: “www.ecy.wa.gov/services/gis/maps/county/ zipcode/zipco.htm”. Accessed Nov. 30, 2005. 12. U.S. Census Bureau. State & County QuickFacts. Available at: “http://quickfacts.census.gov/qfd/states/53000.html”. Accessed Nov. 30, 2005. 13. Washington State Office of Financial Management. Population density. Available at: “www.ofm.wa.gov/popden/index.htm“. Accessed Sept. 13, 2001. 14. Salant P, Dillman DA. How to conduct your own survey. New York: Wiley; 1994. 15. Lam M, Riedy CA, Milgrom P. Improving access for Medicaid-insured children: focus on front-office personnel. JADA 1999;130(3):365-73. 16. Yang EY, Kiyak HA. Orthodontic treatment timing: a survey of orthodontists. Am J Orthod Dentofacial Orthop 1998;113(1):96-103. 17. King GJ, Wheeler TT, McGorray SP, Aiosa LS, Bloom RM, Taylor MG. Orthodontists’ perceptions of the impact of phase 1 treatment for Class II malocclusion on phase 2 needs. J Dent Res 1999;78(11):1745-53. 18. Keeling SD, Wheeler TT, King GJ, et al. Anteroposterior skeletal and dental changes after early Class II treatment with bionators and headgear. Am J Orthod Dentofacial Orthop 1998;113(1):40-50. 19. Tulloch JF, Proffit WR, Phillips C. Influences on the outcome of early treatment for Class II malocclusion. Am J Orthod Dentofacial Orthop 1997;111(5):533-42. 20. Tulloch JF, Phillips C, Koch G, Proffit WR. The effect of early intervention on skeletal pattern in Class II malocclusion: a randomized clinical trial. Am J Orthod Dentofacial Orthop 1997;111(4):391-400. 21. Ghafari J, King GJ, Tulloch JF. Early treatment of Class II, division 1 malocclusion: comparison of alternative treatment modalities. Clin Orthod Res 1998;1(2):107-17. 22. Giorgetti PJ Jr, Machen DE. Fee payment for Medicaid patients. J Clin Orthod 1993;27(10):561-2.
January 2006
Copyright ©2006 American Dental Association. All rights reserved.
R
E
N
D
S
Early orthodontic treatment as a means to increase access for children enrolled in Medicaid in Washington state Gregory J. King, DMD, DMSc; Charles V. Hall, DDS, MSD; Peter Milgrom, DDS; David E. Grembowski, PhD
JADA, Vol. 137
http://jada.ada.org
✷
D
A
✷
A
T
IO N
Background. The authors assessed the likelihood that interceptive orthodontic Medicaid programs C would increase access to care for Washington children. N U I U NG ED A 4 Methods. The authors surveyed 210 Washington RT ICLE state orthodontists, including questions on demographics, attitudes toward early treatment, use of innovations and perceptions of Medicaid. Respondents were either Medicaid participants or nonparticipants. Results. Fifty of 159 respondents were Medicaid participants. Most respondents perceived early orthodontic treatment as beneficial. Medicaid participants were more willing to participate in Medicaid earlytreatment programs, had slightly fewer patients in the “other insurance” category, provided more discounted fees, received more Medicaid inquiries, practiced in rural areas with lower household incomes, reported feeling overworked and experienced fewer Medicaid problems. The principal problem reported with the Medicaid system was low fee reimbursement. Conclusions. Programs offering early orthodontic treatment could increase access. Important barriers would be low fees and unfamiliarity with Medicaid. Practice Implications. Medicaid should design programs aimed at early treatment with reasonable reimbursement and an educational component. Key Words. Access; health care disparities; interceptive orthodontics; early orthodontic treatment; Medicaid. JADA 2006;137:86-94. I
86
A
J
T
A
ABSTRACT CON
ccess to orthodontic services for children from low-income families who are enrolled in Medicaid is limited nationwide. Low participation by orthodontists in Medicaid programs is an important contributing factor. The status of Medicaid-funded orthodontics in Washington state is typical. More than 500,000 children statewide were eligible for dental treatment under the Medicaid program in 1999. Of these, less than 1 percent received any orthodontic care in 1999 (Washington State Department of Social Health Services, written communication, 1999), despite estimates of the incidence of malocclusion being between 15 and 60 percent.1-5 Given that most Medicaid orthodontic programs target the most severe types of malocclusions, these estimates of unmet need may be high, but they remain significant. Approximately onequarter of practicing orthodontists in Washington state participated in the Medicaid program in 1999 but most treated only a few patients enrolled in Medicaid (Washington State Department of Social Health Services, Olympia, Wash., unpublished data, 1999). Ten orthodontists provided approximately 81 percent of the orthodontic treatment statewide, excluding those for patients with cleft lip and palate. Orthodontists may not provide treatment to patients enrolled in
Dr. King is the Moore Riedel professor and the chair, Department of Orthodontics, University of Washington School of Dentistry, Seattle, and an investigator, Northwest/Alaska Center to Reduce Oral Health Disparities, University of Washington, Seattle. Address reprint requests to Dr. King at Department of Orthodontics, University of Washington, School of Dentistry, D569 Health Sciences Center, Box 357446, Seattle, Wash. 98195-7446, e-mail “[email protected]”. Dr. Hall was a graduate student, University of Washington School of Dentistry, when this study was conducted. He now maintains a private practice in orthodontics, Stanwood, Wash. Dr. Milgrom is a professor, Department of Health Services and Dental Public Health Sciences, University of Washington School of Dentistry, Seattle, and is the director, Northwest/Alaska Center to Reduce Oral Health Disparities, University of Washington, Seattle. Dr. Grembowski is a professor, Department of Health Services and Dental Public Health Sciences, University of Washington School of Dentistry, Seattle.
January 2006
Copyright ©2006 American Dental Association. All rights reserved.
T R E N D S
Medicaid because of actual or perceived problems with the Medicaid program and its clients. These problems may include low fee reimbursement, uncooperative patients and high rates of noncompliance. A few studies have examined the perceptions of general dentists and pediatric dentists regarding the treatment of patients enrolled in Medicaid.6-8 Results from these studies show that the most commonly reported deterrent to being a Medicaid participant is low fees. In a study conducted in Washington state, participants reported patient behavioral issues to be of only minor importance.8 Despite both pediatric and general dentists’ reporting dissatisfaction with the allowable fees, they had significant differences in their attitudes toward the Medicaid program. Pediatric dentists were more likely to accept and treat patients enrolled in Medicaid than were general dentists, and they saw treatment of children as more important than did general dentists.7,9 No similar data exist on barriers to Medicaid participation by orthodontists. In addition to low participation by dentists, there are other possible explanations for this disparity between the need and the availability of service for low-income children. For example, there may be fundamental differences in malocclusion patterns in children from low-income families. Cultural, socioeconomic or behavioral differences also may act as barriers to seeking orthodontic treatment. Most publicly funded programs ration orthodontic treatment by limiting it to only the patients with the most severe malocclusions. This approach has profound effects on the level of participation by orthodontists because it selects for patients who will require more resources and who often are at risk of having poorer outcomes even under ideal conditions. This issue is compounded by reimbursement rates of about 60 to 65 percent of the customary fee. An alternative rationing strategy would be to select patients who may be at risk of developing severe malocclusions and intervene early using approaches that have reduced complexity and expense compared with full treatment in the permanent dentition. Orthodontists consider early interventions to be partial treatments that may require a second phase in the permanent dentition to completely alleviate both functional and esthetic problems.10 Orthodontists recognize that the approach to early orthodontic treatment today does not provide finished results and, ideally, should be followed by elective finishing in a
second phase of treatment. These two-phase approaches often are more costly than one-phase treatment in late adolescence. Patients, however, benefit significantly from early treatment, which usually is less complex and costly than secondphase finishing or one-phase treatments during adolescence. As a public health measure, early intervention—without a publicly funded elective second phase of finishing—would provide substantial benefit at a reduced cost per patient compared with full treatment in the permanent dentition. Although most orthodontists may accept this approach, no researchers have examined its usefulness as a means for increasing access to orthodontic services for low-income families. We hypothesize that orthodontists would perceive early orthodontic treatment as a useful means of reducing the severity of malocclusion, that orthodontists who favor early intervention also would favor other innovative orthodontic treatments, that the practice demographic of orthodontists who are Medicaid participants would differ from that of orthodontists who are not, and that these perceptions and practices would predict acceptance of a Medicaid program focusing on early orthodontic treatment in children at risk of developing severe malocclusions. We conducted a study to assess the likelihood that programs focusing on early interception of malocclusions could increase Medicaid participation by orthodontists and, thereby, increase access for low-income families. SUBJECTS, MATERIALS AND METHODS
The institutional review board of the University of Washington approved our study. Study respondents. We identified and surveyed the 210 members of the American Association of Orthodontists who practiced in Washington state in 1999. They represented most of the orthodontists practicing in the state in 1999. Survey. The survey included questions in four general categories that represent the important factors affecting acceptance of a Medicaid program of early orthodontic treatment: practice demographics, attitudes toward early orthodontic treatment, acceptance of innovative orthodontic approaches and perceptions of the Medicaid system. Practice demographics. We designed the first section of the survey to determine if any differences existed between the demographics of Medic-
JADA, Vol. 137 http://jada.ada.org Copyright ©2006 American Dental Association. All rights reserved.
January 2006
87
T R E N D S
aid participants and nonparticipants that could predict acceptance of an early treatment program. We categorized as Medicaid participants only those respondents who treated at least one patient enrolled in Medicaid during the previous year. This section asked for the respondents’ number of years in practice, number of years at their location, total number of patients they started seeing in the past six months, the ZIP code for their main office, the number of patients who they provided discounted fees and who they started seeing in the past six months and the number of Medicaid inquiries in a typical month. We collected the respondent’s ZIP codes so we could assess sociodemographic data such as median family income and population density. First, we determined the county for each ZIP code,11 and then we recorded the median family income12 and population density13 we obtained from 1999 U.S. census data. We also asked respondents what their average fee was for a 24month case, how busy their practice was during the past year and what their practice arrangement was (that is, self-employed, partner, employee). Attitudes toward early orthodontic treatment. We designed this section of the survey to determine if orthodontists have a common vision of the acceptability of early treatment approaches. We asked orthodontists how much benefit patients derived from early treatment and to rank the effectiveness of early treatment for 10 malocclusions on a five-point scale, with 1 being “never effective” and 5 being “always effective.” We designed a subsection to determine if early treatment without a second phase was acceptable. Our rationale for this question stemmed from our belief that an early treatment program under Medicaid may not include funding for a second phase of treatment. Acceptance of innovative orthodontic approaches. We designed the third section of the survey to determine if acceptance of innovative approaches to treatment would predict the likelihood of participation in a proposed new Medicaid program. We asked respondents how many cases involving indirect bonding and how many involving orthognathic surgery they started in the past six months. We also asked them whether they had used self-ligating brackets, self-etching primers, lingual braces or aligners to treat patients. Perceptions of the Medicaid system. We 88
JADA, Vol. 137
http://jada.ada.org
designed the fourth section of the survey to explore how orthodontists felt about the Medicaid system and a proposed program aimed at increasing participation through early treatment of patients. We asked the orthodontists to provide information regarding their patients’ methods of payment during the past six months. Specifically, we asked them what percentages of their patients had Washington Dental Service (WDS) insurance (WDS [Seattle] is the Delta Dental provider in the state), other private insurance, Medicaid, no insurance or other. We listed 10 commonly cited problems with the Medicaid system and patients enrolled in Medicaid. We asked respondents to indicate if they perceived each of these problems to be a “major problem,” “minor problem” or “not a problem” or to respond “don’t know.” We asked this to determine if any of these perceived problems were barriers to participation in Medicaid programs. Study design. Two orthodontists pretested the survey. One had considerable experience with patients enrolled in Medicaid, while the other’s experience with this population was limited. We chose only two orthodontists to avoid any reduction in the sample size or in significantly biasing the sample. One of the orthodontists (the Medicaid participant) was in the process of retiring, so we did not include him in the survey population. We included the other orthodontist in the final survey population. After the pretesting was completed, we revised the survey instrument. We collected data through a mail survey using the Total Design Method.14 First, we sent a letter on University of Washington School of Dentistry stationery that described the study and identified us. Two weeks later, we mailed the survey, a cover letter and a postage-paid return envelope to all 210 members of the American Association of Orthodontists practicing in Washington state. We did not include an incentive with the survey. We mailed a reminder postcard to the nonrespondents six weeks later followed by a replacement copy of the survey after eight more weeks. The total data collection period was 19 weeks. We kept the survey responses confidential and destroyed links between the contact information and the actual questionnaires after we received the surveys. Data analysis. We analyzed the data using Statistical Package for the Social Sciences (Version 8.0, SPSS, Chicago). All of the respondents did not answer every question. Therefore, we cal-
January 2006
Copyright ©2006 American Dental Association. All rights reserved.
T R E N D S
TABLE 1 culated percentages based only on the number of orthodontists who responded Acceptance of proposed Medicaid to a particular question. program and innovative techniques. We compared orthodontists’ percepVARIABLE n NONPARTICIPANTS PARTICIPANTS tions of the benefits of early treatment Attitude Toward and the frequency of presenting twoEarly Orthodontic phase treatment options using 2 × 4 Treatment tables (tables with two columns and four (% Yes [n]) 58.3 (28) 44.4 (48) 156 rows) and χ2 tests. We calculated, Consider ranked and compared the mean 28.6* (14) 13.9* (15) 157 Participate response scores for the respondents’ ratWilling to Adopt ings of the perceived effectiveness of Innovative Techearly treatment for 10 malocclusions niques (% Yes [n]) using χ2 tests. 55.1 (27) 44.0 (48) 158 Self-ligating brackets We separated respondents into two 53.1 (26) 42.2 (46) Self-etching primers 158 groups for each of the six “innovative” orthodontic techniques shown in Table 26.5 (13) 22.9 (25) Lingual braces 158 1, according to whether they used that † 59.2† (29) 42.2 (46) Aligners 158 particular technique. We defined a user 14.7 (16) 16.3 (8) Indirect bonding 158 of indirect bonding and orthognathic surgery techniques as anyone who had 93.5 (101) 93.6 (44) Orthognathic surgery 155 used these techniques during the past * P < .05, Pearson χ test value = 4.825. six months. We defined a user of the † P < .05, Pearson χ test value = 3.909. other techniques as anyone who had ever used that particular technique. We future Medicaid programs, we separated the then compared the differences between Medicaid participants and nonparticipants using individual respondents into two groups based on their mean perceived problem scores, discussed previously. 2 × 2 tables and χ2 tests. We placed respondents with a mean problem We compared demographic and response frequency differences between Medicaid participants score of 0 to 1 in the “low” group and those with a score of 1.1 to 2 in the “high” group. We then comand nonparticipants using a series of 2 × 2, 2 × 3 2 pared their differences in willingness to particiand 2 × 4 tables and χ tests. We calculated pate in the two Medicaid programs using 2 × 3 respondents’ means and standard deviations tables and χ2 tests. Next, we compared the mean (SDs) for the number of patients they offered discounted fees, number of Medicaid inquiries, perceived problem scores for those willing to parnumber of years in practice, number of years at ticipate in each of the programs with the mean location, percentage of each payment method, scores for those who were not willing or were not population density and median family incomes. sure if they would participate. We performed the We analyzed mean differences between particiMann-Whitney test to determine significance. pants and nonparticipants using nonparametric To determine if orthodontists’ use of innovative tests because the data were not distributed techniques is related to willingness to participate normally. in future Medicaid programs, we created an index We also calculated a mean response score for of each respondent’s propensity toward innovaeach of the perceived problems with the Medicaid tion by adding the number of innovative techsystem. We gave the response of “not a problem” a niques each respondent used. Possible scores value of 0, the response of “minor problem” a ranged from 0 to 6. We then separated respondvalue of 1 and the response of “major problem” a ents into high and low innovation groups. We value of 2. We compared mean scores using the analyzed various cutoffs for each group and anaMann-Whitney test. We excluded respondents lyzed the data using 2 × 3 tables and χ2 tests. who chose the “don’t know” response from our RESULTS calculation. To compare perceptions of Medicaid problems Response rate. A total of 159 of 210 surveys with a respondent’s willingness to participate in were completed. One survey was returned 2 2
JADA, Vol. 137 http://jada.ada.org Copyright ©2006 American Dental Association. All rights reserved.
January 2006
89
T R E N D S
respondents had a mean number of years Comparison of respondents’ financial variables. in practice of 15.9 years with no differences FINANCIAL VARIABLE n NONPARTICIPANTS PARTICIPANTS based on participant status. This is consis0† (0) 129 6.1† (15.8) % Patients Enrolled in Medicaid (SD *) tent with the mean of 16.2 (SD = 10.2) years 157 Full Treatment Fee in practice for all 6.4 (7) 14.6 (7) — % < $4,000 (n) American Association of 74.3 (81) 75.0 (36) — % $4,000 > $5,000 (n) Orthodontists members in Washington state. 19.3 (21) 10.4 (5) — % > $5,000 (n) Medicaid participants 36.0 (19.1) 39.6 (15.4) 129 % Private Funding (SD) more often were located 28.2† (15.7) 21.1† (11.0) 129 % Washington Dental in rural counties or Service Insurance (SD) counties with lower pop33.7 (15.2) 31.4 (13.3) 129 % Other Insurance (SD) ulation densities and with lower median 2.3 (6.5) 2.4 (9.1) 129 % Other Funding (SD) family incomes. Med3.01† (5.6) 5.18† (7.0) 140 No. of Discounted Fees icaid participants also (SD) started seeing fewer 1.05 (1.7) 1.16 (1.7) 146 No. of No Fees (SD) new patients per year 4.91‡ (7.1) 11.8‡ (12.4) 140 No. of Medicaid Inquiries but were more likely to Per Month (SD) report feeling over* SD: Standard deviation. worked, while nonpar† P < .05, Mann-Whitney test. ticipants were more ‡ P < .001, Mann-Whitney test. § Not significant, χ test. likely to report feeling busy enough but not overworked. Attitudes toward early orthodontic treatwithout a forwarding address, and one was ment. There were no significant differences returned unfilled because the practice was in the between Medicaid participants and nonparticiprocess of closing. We considered the response pants regarding perceptions of early orthodontic rate of 76 percent (159/208) to be an adequate treatment. A total of 99.4 percent of respondents representation of the orthodontists in said that there was at least some benefit to early Washington state. orthodontic treatment (Table 4, page 92). A total Practice demographics. Fifty of the 159 of 53.9 percent of the respondents reported they respondents (31.4 percent) reported some level of “often” or “always” offer an early treatment or a Medicaid billing during the past six months. A two-phase treatment option to their patients. total of 6.1 percent of patients in these practices The ratings of the effectiveness (on a scale of were enrolled in Medicaid (Table 2). There were 1-5, with 5 representing greater effectiveness) of no differences in the mean fees charged by parearly treatment for 10 malocclusions generally ticipants and nonparticipants. Overall, responwere favorable. Rated as always effective were dents reported that 60.3 percent of their patients anterior crossbite with a shift (4.6, SD = 0.57), had some form of private insurance. Medicaid posterior crossbite with a shift (4.5, SD = 0.58) participants had slightly fewer patients with priand posterior crossbite without a shift (4.0, SD = vate insurance, treated more patients at dis0.83). Early orthodontic treatments rated as modcounted fees and received more than twice as erately effective were anterior crossbite without a many inquiries from patients enrolled in Medshift (3.9, SD = 0.95), Class II with overjet greater icaid compared with nonparticipants. than or equal to 6 millimeters (3.8, SD = 0.58), We saw no differences in practice arrangeClass II with overjet less than 6 mm (3.6, SD = ments between Medicaid participants and non0.79), arch length deficiency less than 4 mm (3.5, participants. Most respondents were selfSD = 1.08) and arch length deficiency 4 to 8 mm employed without partners (Table 3). The TABLE 2
§
2
90
JADA, Vol. 137
http://jada.ada.org
January 2006
Copyright ©2006 American Dental Association. All rights reserved.
T R E N D S
(3.4, SD = 0.84). The TABLE 3 effectiveness of early Comparison of respondents’ environmental variables. orthodontic treatments for arch length defiPARTICIPANTS n NONPARTICIPANTS ENVIRONMENTAL VARIABLE ciency greater than 8 mm (3.0, SD = 1.08) 158 Practice Arrangement and tooth 75.5 (37) 77.1 (84) — % Self-employed without agenesis (3.0, SD = partners (n) 0.89) was rated low. % Self-employed without 6.1 (3) 2.8 (3) — Acceptance of partners but with expense innovative orthosharing (n) dontic techniques. % Self-employed as a partner, 8.2 (4) 11.0 (12) — Medicaid participants sharing income and expenses (n) were more likely to express a willingness to % Employed by another 6.1 (3) 5.5 (6) — orthodontist (n) participate in a Medicaid-sponsored pro% Other (n) 4.1 (2) 3.7 (4) — gram of early orthoDemographics dontic treatment (Table 157 Mean population density (no. 377.9† (296.4) 567.9† (286.8) 1). Both Medicaid parper square mile) (SD*) ticipants and nonpar157 Mean family income ($1,000) 44.0† (10.1) 50.5† (8.3) ticipants were similar (SD) in their use of innova137 Mean no. of case starts (SD) 109.6 (60.8) 117.2 (75.1) tions in orthodontics, with the only difference Mean years in practice (SD) 158 15.8 (10.9) 15.9 (9.8) being that participants 158 Mean years at current location 12.5 (9.9) 12.6 (9.3) reported using aligners (SD) more than did than 153 Busyness nonparticipants. We — % Too busy (n) 0 (0) 2.8 (3) found that orthodon‡ tists’ perceptions of — % Overworked (n) 19.1‡ (9) 6.6 (7) early treatment and — % Not overworked (n) 44.7§ (21) 67.9§ (72) their use of innovative — % Not busy (n) 36.2 (17) 22.6 (24) approaches to treatment did not predict a * SD: Standard deviation. † P < .001, Mann-Whitney test. willingness to partici‡ P < .05, χ test. pate in the proposed Medicaid program. Perceptions of the Medicaid system. Medfor appointments,” “difficulty collecting from Medicaid participants had fewer problems with Medicaid,” “delays in receiving payment” and “patients icaid in only two (“getting billing questions are often late.” Most respondents in both groups answered” and “need for prior authorization”) of considered “unruly or uncooperative patients,” the 10 issues cited in the survey (Table 5, page 93). “loss of coverage during treatment” and “patients Medicaid respondents reported that these were cancel at the last minute” issues considered not to only minor problems, whereas the majority of nonbe significant. participants saw them as major problems. “Fee DISCUSSION reimbursement too low” was the largest problem The data from our study are limited to a single for both participants and nonparticipants. Sevstate, a fact that one should consider when interenty-nine percent of all respondents reported that preting the results. Some aspects of the data may low fees were a major problem (data not shown). be unique to the environment in Washington state Other issues considered to be of some significance (for example, demographics, fees and modes of by both participant groups were, in descending practice) and, therefore, not easily generalizable. order of importance, “patient may fail to show up 2
JADA, Vol. 137 http://jada.ada.org Copyright ©2006 American Dental Association. All rights reserved.
January 2006
91
T R E N D S
TABLE 4
of the level, would likely be different from those who do Respondents’ perceptions of early orthodontic not. treatment.* Practice demographics. PERCEPTION NONPARTICIPANTS (% [n]) PARTICIPANTS (% [n]) Our failure to find differences based on participant status in Beneficial practice arrangement, fees, 55.9 (57) 43.8 (21) A lot number of new cases started, 43.1 (44) 56.3 (27) Some years in practice and practice location indicates that these 1.0 (1) 0 (0) Very little are not important determi0 (0) 0 (0) None nants in the likelihood of a Need for Phase 2 Treatment clinician’s participating in a Medicaid program. However, 10.4 (11) 6.3 (3) Always Medicaid participants pro45.3 (48) 43.8 (21) Often vided more discounted fees 43.4 (46) 50.0 (24) Occasionally than and received twice as many inquiries from patients 0.9 (1) 0 (0) Never enrolled in Medicaid as did * P = nonsignificant, χ test. nonparticipants. The former may be due to a higher level of altruism or greater exposure to low-income famiOther aspects (for example, orthodontists’ perceplies. The latter suggests that reduced access and tions of the usefulness of early treatment and Medreferral patterns tend to funnel patients enrolled icaid) are likely to be more widely applicable. in Medicaid to practitioners who are known to The low level of Medicaid participation in genaccept them. Medicaid participants treated fewer eral by orthodontists in Washington state (approxiprivately insured patients than did nonparticimately 25 percent) and the poor distribution of pants. This may reflect fewer privately insured those providing these services (10 orthodontists patients in the areas in which Medicaid participrovided approximately 81 percent of the orthopants are located. It also was interesting to note dontic treatment statewide) resulted in a low that Medicaid participants tended to feel more average percentage of patients enrolled in Medoverworked than nonparticipants, despite having icaid per practice (6.1 percent). This raises the consimilar caseloads to nonparticipants. This may be cern that the comparisons of participants and nona reflection of differing practice management participants based on acceptance of any patients styles, such as less use of auxiliary personnel. enrolled in Medicaid may have minimal value Our survey shows that orthodontists with pracbecause of the small difference between levels of tices in areas with lower population densities participation in the two groups. An alternative (rural areas) and lower median family incomes approach to this study design would have been to were more likely to treat patients enrolled in Medcompare the 10 orthodontists who provided the icaid than were orthodontists with practices in majority of the treatment with the remainder of more densely populated areas with higher median the orthodontists in the state. This could have profamily incomes. This differs from an earlier report9 vided clearer differences in participation, but it may not have been representative. Because of that found no difference in Medicaid participation these differences, we felt that programs designed rate based on practice location. However, the preto increase access should target the rank and file vious study categorized practices as downtown/city, orthodontists, not the practitioners who already suburban or rural, whereas our study categorized are doing the work. The value of us using a more practices strictly on the basis of county census liberal definition of Medicaid participation in this data. The finding that orthodontists practicing in study was that it provided us with a large enough rural areas treated more patients enrolled in Medgroup of practitioners so we could have some confiicaid has a few potential interpretations. Orthodence that it was representative. Moreover, we felt dontists who choose to practice in rural counties that the characteristics of practitioners who elect may be more community-oriented and may have to participate in the Medicaid program, regardless more inclusive practices.15 Rural environments also 2
92
JADA, Vol. 137
http://jada.ada.org
January 2006
Copyright ©2006 American Dental Association. All rights reserved.
T R E N D S
TABLE 5 may provide more pressure not to deny treatment to patients Rankings of respondents’ perceived problems with enrolled in Medicaid because Medicaid. there would be fewer other NONPARTICIPANTS PARTICIPANTS (MEAN PROBLEM n orthodontists to whom to refer (MEAN RANKING* [SD†]) RANKING* [SD]) these patients. This difference Fee Reimbursement Too 134 1.7 (0.5) 1.8 (0.6) also may relate to fees.9 Our Low data showed that the average 123 1.5 (0.7) fee for rural practices was lower Patients May Fail to 1.4 (0.8) Show Up for than that for urban practices, Appointments 113 1.5 (0.6) and, therefore, the difference 1.4 (0.8) Difficulty Collecting From Medicaid between the Medicaid reim104 1.5 (0.7) bursement and the standard fee Delays in Receiving 1.4 (0.7) Payment may not have been sufficient to 119 1.4 (0.7) influence an orthodontist’s 1.2 (0.8) Patients Are Often Late acceptance of patients. This 119 1.4 (0.7) 1.2 (0.8) Patients Cancel at the finding also may be unique to Last Minute the demographics of Wash112 1.4 (0.7)‡ 1.0 (0.8)‡ Getting Billing Questions ington state and not generalizAnswered able to other states because the 1.3 (0.8)§ 1.0 (0.8)§ 119 Need for Prior rural areas of the state are priAuthorization marily agricultural with large 1.2 (0.8) 1.0 (0.8) 125 Unruly or Uncooperative numbers of low-income migrant Patients farm workers. 1.2 (0.7) 0.9 (0.8) 99 Loss of Coverage During Attitudes toward early Treatment orthodontic treatment. 1.4 (0.4)¶ 1.2 (0.3)¶ 83 Overall Mean Problem Almost all respondents reported Score that there was at least some * Ranking scale: 0 = No problem, 1 = minor problem, 2 = major problem. Respondents who did not answer the question or responded “don’t know” were excluded from the mean calculation. benefit to early orthodontic † SD: Standard deviation. treatment. This finding sup‡ P < .01, Mann-Whitney test = 1010. ports our hypothesis that ortho- ¶§ P < .05, Mann-Whitney test = 1182. P < .05, Mann-Whitney test = 120.5. dontists perceive early orthodontic treatment as a useful means of reducing the severity of malocclusion. The generally favorable perceptions of the effecHowever, about one-half of the respondents always tiveness of early treatment for most malocclusions or often tell patients there is a need for a twosuggest that orthodontists feel comfortable using phase treatment plan, suggesting that many view the approach for a wide variety of conditions. This early treatment as only partial treatment. Our further suggests that a program aimed at early conclusions are supported by frequent reports of orthodontic treatments could affect most types of malocclusions. success of early treatment in reducing the severity Acceptance of innovative orthodontic of several types of malocclusion in the mixed dentiapproaches. For a new program to improve access, tion.16-21 Our failure to find significant differences between Medicaid participants and nonpartician important requirement would be that a substanpants regarding their perceptions of early treatment tial number of nonparticipating orthodontists would effectiveness suggests that negative perceptions of find it appealing enough to become Medicaid particithe effectiveness of early treatment would not be a pants. Our survey failed to show this trend with barrier to acceptance of such a program by orthorespect to early orthodontic treatment. The greater dontists. However, a better understanding of the acceptance of a Medicaid-sponsored program of frequency of elective phase 2 treatments would be early orthodontic treatment by participants sugimportant because orthodontists may be uncomfortgests that practitioners’ familiarity with the Medable with accepting patients for publicly funded icaid system and its clients may be an important phase 1 treatment without assurances that needed prerequisite for such a program to improve access. phase 2 treatment would be available. Educational initiatives designed to familiarize non-
JADA, Vol. 137 http://jada.ada.org Copyright ©2006 American Dental Association. All rights reserved.
January 2006
93
T R E N D S
participants and their staff members with Medicaid and the new program would be essential for these programs to affect access much. Our study failed to show differences in the use of innovative techniques based on participant status, except with respect to aligners. The failure to find associations between orthodontists’ perceptions of early treatment, their use of innovative or new techniques and their willingness to participate in the proposed Medicaid program does not support our hypothesis that openness to innovative orthodontic treatments predicts a willingness to participate in new Medicaid programs. Perceptions of the Medicaid system. Medicaid’s low reimbursement rate is an important problem for both participants and nonparticipants.9,22 However, the impact of low fees on access is unclear, because we do not know if nonparticipating dentists would be willing to treat patients enrolled in Medicaid without increasing fees. An evaluation of the effect of fees on participation by Medicaid participants would be required to confirm this. A program targeting early treatment has the potential to reduce the costs per patient because these procedures tend to be less complex and, therefore, less expensive. For the same orthodontic budget, this approach would permit more patient treatments, but it may not increase the level of participation or access. Bureaucratic hurdles seem to be less of a problem, and the differences between participants and nonparticipants suggest that these hurdles could be overcome by more familiarity with the program. CONCLUSIONS
We found that orthodontists perceive early treatment as acceptable and having wide applicability. Clinicians who are already Medicaid participants would most readily accept a Medicaid-sponsored program of early orthodontic treatment. Therefore, such a program needs to be accompanied by initiatives to educate nonparticipants. We also found that Medicaid participants had fewer patients with private insurance, provided more discounted fees, received more inquiries from patients enrolled in Medicaid, practiced in more rural areas with lower family incomes, were more likely to feel overworked and had fewer problems
94
JADA, Vol. 137
http://jada.ada.org
with the Medicaid system. The most significant problem with Medicaid cited by both participants and nonparticipants was low fee reimbursement. ■ This work was supported, in part, by National Institute of Dental and Craniofacial Research/National Center on Minority Health and Health Disparities/National Institutes of Health grant U54-14254 to the Northwest/Alaska Center to Reduce Oral Health Disparities. Dr. Hall’s work was completed in partial fulfillment of a master of science in dentistry degree in orthodontics. 1. Bowden DE, Davies AP. Inter- and intraexaminer variability in assessment of orthodontic treatment need. Community Dent Oral Epidemiol 1975;3(4):198-200. 2. Brunelle JA, Bhat M, Lipton JA. Prevalence and distribution of selected occlusal characteristics in the US population, 1988-1991. J Dent Res 1996;75(special number):706-13. 3. Helm S, Kreiborg S, Barlebo J, et al. Estimates of orthodontic treatment need in Danish schoolchildren. Community Dent Oral Epidemiol 1975;3(3):136-42. 4. Foster TD, Day AJ. A survey of malocclusion and the need for orthodontic treatment in a Shropshire school population. Br J Orthod 1974;1(3):73-8. 5. Pietila T, Pietila I, Vaataja P. Early screening for orthodontic treatment: differences in assessments made by a consultant orthodontist and three public health dentists. Community Dent Oral Epidemiol 1992;20(4):208-13. 6. Lang WP, Weintraub JA. Comparison of Medicaid and non-Medicaid dental providers. J Public Health Dent 1986;46(4):207-11. 7. Venezie RD, Vann WF Jr. Pediatric dentists’ participation in the North Carolina Medicaid program. Pediatr Dent 1993;15(3):175-81. 8. Milgrom P, Riedy C. Survey of Medicaid child dental services in Washington state: preparation for a marketing program. JADA 1998;129(6):75363. 9. Nainar SM, Tinanoff N. Effect of Medicaid reimbursement rates on children’s access to dental care. Pediatr Dent 1997;19(5):315-6. 10. Kluemper GT, Beeman CS, Hicks EP. Early orthodontic treatment: what are the imperatives? JADA 2000;131(5):613-20. 11. Washington State Department of Ecology. Washington Zip code maps. Available at: “www.ecy.wa.gov/services/gis/maps/county/ zipcode/zipco.htm”. Accessed Nov. 30, 2005. 12. U.S. Census Bureau. State & County QuickFacts. Available at: “http://quickfacts.census.gov/qfd/states/53000.html”. Accessed Nov. 30, 2005. 13. Washington State Office of Financial Management. Population density. Available at: “www.ofm.wa.gov/popden/index.htm“. Accessed Sept. 13, 2001. 14. Salant P, Dillman DA. How to conduct your own survey. New York: Wiley; 1994. 15. Lam M, Riedy CA, Milgrom P. Improving access for Medicaid-insured children: focus on front-office personnel. JADA 1999;130(3):365-73. 16. Yang EY, Kiyak HA. Orthodontic treatment timing: a survey of orthodontists. Am J Orthod Dentofacial Orthop 1998;113(1):96-103. 17. King GJ, Wheeler TT, McGorray SP, Aiosa LS, Bloom RM, Taylor MG. Orthodontists’ perceptions of the impact of phase 1 treatment for Class II malocclusion on phase 2 needs. J Dent Res 1999;78(11):1745-53. 18. Keeling SD, Wheeler TT, King GJ, et al. Anteroposterior skeletal and dental changes after early Class II treatment with bionators and headgear. Am J Orthod Dentofacial Orthop 1998;113(1):40-50. 19. Tulloch JF, Proffit WR, Phillips C. Influences on the outcome of early treatment for Class II malocclusion. Am J Orthod Dentofacial Orthop 1997;111(5):533-42. 20. Tulloch JF, Phillips C, Koch G, Proffit WR. The effect of early intervention on skeletal pattern in Class II malocclusion: a randomized clinical trial. Am J Orthod Dentofacial Orthop 1997;111(4):391-400. 21. Ghafari J, King GJ, Tulloch JF. Early treatment of Class II, division 1 malocclusion: comparison of alternative treatment modalities. Clin Orthod Res 1998;1(2):107-17. 22. Giorgetti PJ Jr, Machen DE. Fee payment for Medicaid patients. J Clin Orthod 1993;27(10):561-2.
January 2006
Copyright ©2006 American Dental Association. All rights reserved.