JDE
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

J Dent Educ. 71(3): 403-418 2007
© 2007 American Dental Education Association
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Davidson, P. L.
Right arrow Articles by Andersen, R. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Davidson, P. L.
Right arrow Articles by Andersen, R. M.

Milieu in Dental School and Practice

Influence of Contextual Environment and Community-Based Dental Education on Practice Plans of Graduating Seniors

Pamela L. Davidson, Ph.D.; Daisy C. Carreon, M.P.H.; Sebastian E. Baumeister, Ph.D.; Terry T. Nakazono, M.A.; John J. Gutierrez, B.A.; Abdelmonem A. Afifi, Ph.D.; Ronald M. Andersen, Ph.D.

Key words: dental practice intention, dental education, access to dental care

Submitted for publication 07/31/06; accepted 11/07/06


   Abstract
Top
 Abstract
Literature review
 Materials and Methods
 Results
 Discussion
 Conclusions
 References
 
This study investigated senior dental students’ plans to provide care to underserved racial/ethnic minority populations. Three sets of determinants were analyzed: contextual environment, community-based dental education (CBDE), and student characteristics. We analyzed data from the ADEA Survey of Dental School Seniors and administrative data sources to construct contextual variables. Multivariable results show three contextual variables predicted practice plans: greater numbers of federally qualified health centers, higher percentages of underrepresented minorities, and attending a California Pipeline dental school. Regarding CBDE predictors, it was alarming to find seniors who viewed the cultural competency curriculum as inadequate and perceived themselves as less prepared to provide oral health care to diverse populations were also those most likely to serve minority patients. Significant student characteristics included racial/ethnic minority, female gender, older age, lower parent’s income, and socially conscious orientation. The study provides evidence that contextual environment, CBDE, and student characteristics were significantly associated with plans to care for underserved patients. Findings suggest if the Pipeline initiative is successful in stimulating reform in U.S. dental schools, future students will develop greater awareness regarding critical access problems and the competencies required to effectively care for diverse populations. In the long term, addressing the problem of dental care access will require the creation of policy, financial, and structural interventions to motivate providers to care for the underserved.

The nation’s dental care safety net is inadequate to serve the population in need, and access problems are likely to intensify.1,2 Access barriers are created on several fronts not the least of which are the following: 1) so few underrepresented minorities enter dentistry; 2) a general apathy regarding societal responsibility to care for underserved populations among dental schools, faculties, and the students they historically recruit; 3) little or no government reimbursement to provide care to low-income, uninsured populations; and 4) sizeable educational debt incurred by dental school seniors, which immediately accrues interest upon graduation. Minority representation in the dental and other health professions remains a concern.3–5 The shortage of dentists is particularly critical in African American and Hispanic communities.6–8

This study analyzed dental school senior students’ plans to provide care to underserved racial/ethnic minority populations. The Robert Wood Johnson Foundation (RWJF) and The California Endowment (TCE) funded the Pipeline, Profession, and Practice: Community-Based Dental Education program to address the critical shortage of oral health care services for underserved and disadvantaged populations by changing dental education in the United States. RWJF funded competitive grants proposed by eleven of the fifty-six accredited U.S. dental schools, and one year later, TCE funded four additional dental schools in California to design and implement a Pipeline program. Additionally, TCE required all five of California’s dental schools to develop a regional recruitment program for underrepresented minorities and a collaborative statewide health policy effort to sustain the Pipeline initiative after funding ends.

The three Pipeline objectives are: 1) increase recruitment and retention of underrepresented minority and low-income students; 2) revise didactic and clinical curricula to support community-based educational programs; and 3) establish community-based clinical education programs that provide dental students and residents with sixty days of experience in a patient care environment.9,10 This study examines student characteristics (e.g., race/ethnicity), community-based dental education (e.g., curriculum and clinical rotations), and contextual environment (e.g., number of federally qualified health centers in the county) on a longer-term outcome, the practice plans of senior dental students.

Data were analyzed from the 2003 American Dental Education Association (ADEA) Survey of Dental School Seniors and a set of contextual variables. To our knowledge, this study is the first to include individual and contextual-level variables to investigate plans to provide care to underserved minority patients upon graduation; no other studies were found in the literature combining these levels of data. From a methodological perspective this study advances the use of contextual variables in dental education and dental health services research and provides detailed information on constructing contextual variables.

The study examines baseline measures before the Pipeline program was implemented in academic year 2002–03. Future evaluation research will analyze data collected in the 2007 ADEA survey, when foundation funding culminates, to examine the impact of the Pipeline program on practice plans of graduating seniors.


   Literature Review
Top
 Abstract
 Literature review
Materials and Methods
 Results
 Discussion
 Conclusions
 References
 
Articles were identified from medical and health services research journals using PUBMED database and the following keywords: plans upon graduation, early and long-term career decisions, practice decisions, preferences, motivations, and interests. Research has emphasized different aspects of practice plans, such as practice location,11–14 practice arrangement (solo versus group practice),4,15–17 and postdoctoral education and specialization.18,19 Six studies specifically investigated factors influencing plans to provide care to underserved and minority populations.4,20–24 Several researchers focused on the personal characteristics of students or dentists, such as gender and race/ethnicity;4,15–18,24 only a few explored variables that influence the individual, i.e., educational program and economic factors.12,21,22 The following sections summarize findings from the research literature on determinants of practice plans including student characteristics, community-based dental education, and contextual variables.

Student Characteristics
In part, the Pipeline initiative was promulgated on the belief that workforce diversity may help to alleviate disparities in oral health care for low-income and underserved populations. Minority providers may be more culturally sensitive to their minority patients’ needs. Our review yielded only a few studies examining the relationship between race/ethnicity and plans to provide care for minority patients. One study showed African Americans compared to whites were more likely to establish practices in underserved communities, provide care to uninsured and Medicaid beneficiaries, and continue service after participation in the National Health Service Corps.4,15,21 In studying the supply of dentists in California, Hayes-Bautista et al.5 found Latino dentists comprised 4.6 percent of the total dental supply in 2000. Although Latinos comprise about a third of the state’s population, only one out of every twenty dentists was Latino. Our search yielded no literature on practice plans of American Indians. Clearly, more research is needed to understand recruitment of underrepresented minorities to dentistry and implications for practice upon graduation.

The root problem for the underrepresented minority groups (African American, American Indian, and Hispanic) appears to be that so few apply for and are accepted into dental school. A report issued by ADEA found underrepresented minorities comprised 12.4 percent of the applicant pool and 11.6 percent of first-year enrollees in 2004.25 Asian/Pacific Islanders and whites comprised 69.7 percent of applicants and 71.1 percent of first-year enrollees (see Table 3 in Weaver et al. for more detailed information).26

Other student characteristics, such as gender and age, have been found to influence dental students’ preferences for type and specialty of practice.15,19,27,28 For instance, males have consistently rated the solo owner practice arrangement more favorably than females.15–17,29 Age significantly influences the decision to enter into an academic career, with younger individuals finding income of an academic dentist to be a deterrent.19 However, the search yielded no studies indicating gender or age was associated with providing care to the underserved.

Our search yielded few studies showing the relationship between attitudes and beliefs and practice plans. Medical students when compared to dental students were found to demonstrate greater altruism and a sense of intellectual challenge as motivating factors in career choice; dental students demonstrated more commitment to personal and financial gain.18 In studying attitudes of family physicians, Eliason et al.24 found an association between universalism values (i.e., motivation to enhance and protect all people) and the number of indigent patients served. Li et al.23 found primary care providers who had a "strong sense of service to humanity" were more satisfied with their work.

Some believe inequities are compounded in dentistry more than medicine due to tensions within the dental profession between the moral values traditionally identified with the health professions and the commercial values of practice achieved through entrepreneurial self-interest.30 In summary, although the literature is limited, research suggests certain student characteristics (race/ethnicity, attitudes, and beliefs) do predict practice plans.

Community-Based Dental Education
In a descriptive study, Smith et al.20 found a positive relationship between curricular emphasis on treating patients from diverse backgrounds and student and alumni intentions to care for these patients in their practices. Findings from another study showed greater time spent in rotations was a significant predictor of perceived ability to provide care to diverse groups.31 In a third study, Mofidi et al.21 found 46 percent of alumni who participated in the National Health Service Corps continued to provide care to underserved groups. In contrast, DeCastro et al.22 found no significant differences in alumni attitudes towards practicing in underserved areas or accepting Medicaid patients between a community-oriented dental education program and a traditional program. Although results are somewhat equivocal, the literature does suggest some correlation between preparation in the academic programs (didactic and clinical rotations) and the extent to which care is provided to underserved patients in future practice.

Contextual Environment
Contextual variables represent the social, economic, structural, and public policy environment influencing access to care.32,33 As noted, much of the literature on practice plans focuses on the characteristics of the decision maker (student or dentist) and a few on the educational program. Our search yielded only one study testing the effects of contextual variables on plans to provide care to underserved patients. Beazoglou et al.12 found size of population, per capita disposable income, and cost of operating a dental practice were significantly associated with number of practicing dentists in 140 Connecticut townships. The study examined distribution of dentists in the state, but did not analyze the influence of individual characteristics or academic program on the decision-making process; only contextual variables were analyzed in this study.


   Materials and Methods
Top
 Abstract
 Literature review
 Materials and Methods
Results
 Discussion
 Conclusions
 References
 
The following sources of data were used to investigate senior dental students’ plans to provide care to underserved racial/ethnic minority populations.

From the ADEA Annual Survey of Dental School Seniors, questionnaire items were identified that best represent student characteristics and components of community-based dental education using a conceptual and analytical model to guide variable selection (Figure 1Go). The ADEA survey administers an annual questionnaire to graduating seniors in accredited dental schools. The survey collects information about social and demographic characteristics, educational financing, indebtedness, adequacy of time in predoctoral instruction, preparedness for practice, and practice and postdoctoral plans.25 Additionally, starting in 2003 the survey began collecting information related to the Pipeline initiative and community-based dental education, e.g., recruitment, curriculum, and extramural clinical rotations. Each school uses its own survey distribution and collection system. Surveys are returned annually to ADEA for reporting and analysis. Additionally, a set of constructed contextual variables came from administrative data sources measuring policy, population, dental care delivery system, and dental school environment. (See Table 1Go for a detailed list of data sources and references used to construct contextual variables.)


Figure 1
View larger version (21K):
[in this window]
[in a new window]

  		Figure 1. Measurement model for predicting plans to care for underserved minority patients

 

View this table:
[in this window]
[in a new window]

  		Table 1. Description and distribution of independent and dependent variables
 
Figure 1Go presents a measurement model for predicting plans to care for underserved minority patients. The model posits contextual environment, community-based dental education (CBDE), and student characteristics influence practice plans of graduating seniors. The variables selected to measure the various constructs, along with their definitions and distributions, are provided on Table 1Go.

The dependent variable for students’ practice plans was measured using the following ADEA questionnaire item: "When you enter practice, about what percent of your patients do you expect will be from underserved racial/ethnic minority populations?" The variable included five response categories (or cut points): 0 percent (n=177, 4.8 percent), 1–10 percent (n=1460, 39.5 percent), 11–24 percent (n=1246, 33.7 percent), 25–50 percent (n=291, 7.9 percent), and greater than 50 percent (n=234, 6.3). We combined the 25–50 percent and greater than 50 percent categories for the analysis (n=525, 14.2 percent). A practice plans intervening variable was included and constructed into a categorical measure indicating primary activity immediately upon graduation: 1) private practice (49.9 percent), 2) community clinic or government service (9.9 percent), and 3) postdoctoral or academic appointment (40.2 percent). Independent variables in the measurement model are discussed in the sections below: contextual environment, CBDE, and student characteristics (Figure 1Go).

The contextual environment included policy, population, dental care delivery system, and dental school characteristics (Table 1Go). Federal, state, and local health policies influence dental care financing and the percent of the population with public insurance. Racial/ethnic representation in the state legislature can influence resources for medical and dental education and the availability of services for vulnerable populations. Two state policy variables were measured: percent underrepresented minorities in the state legislature (mean=16.8, range 0–100 percent because the study includes Puerto Rico); and adult Medicaid dental coverage: no benefits (15 percent), emergency only (18.2 percent), partial coverage (39.4 percent), and full coverage (27.3 percent).

Contextual variables can be used to measure population characteristics and their collective effect on access. For example, when large numbers of low-income, racial/ethnic minority groups and/or uninsured persons reside in a geographic area, access barriers are magnified for individuals competing for limited services and resources.32 When dental students have the opportunity to train in community-based settings, potentially they will become more aware of access barriers and better trained to respond to population oral health needs. Two county-level population variables were measured: percent underrepresented minorities (ranging from 4.6 to 98.3 percent because the study includes Puerto Rico); and percent population with income less than 200 percent of the federal poverty level (ranging from 15.8 percent to 63.6 percent).

Three contextual variables measured the dental care delivery system: 1) practicing dentists per 10,000 population for each state and District of Columbia (ranging from 3.9 to 12.6); 2) number of federally qualified health centers (FQHCs) in the county providing dental care (ranging from 0–10); and 3) number of federally qualified health centers (FQHCs) in the county per 100,000 low-income residents (ranging from 0 to 81.7).

The final set of contextual variables measured dental school environment (Table 1Go). Data collected by the American Dental Association (ADA) were used to show university ownership (public or private). Two variables were constructed from dental school mission statements: commitment to recruiting diverse students and providing health care to underserved populations. Two school-level variables were measured: percent underrepresented minority (URM) dental students (years 1–5); and average total educational cost for first-year students. The final contextual variable measured the dental school’s aggregate cultural and social environment using average values at the school level from the following ADEA questionnaire item: "The cultural and social environment of your school promotes acceptance and respect of students and patients of different races, ethnicities, and cultures," measuring level of agreement using a four-point Likert scale.

Additionally, the measurement model (Figure 1Go) included a critical program evaluation measure showing Pipeline program status among the accredited dental schools: 1) "National Pipeline" schools funded by the Robert Wood Johnson Foundation; 2) "California Pipeline" schools funded by The California Endowment; and 3) non-Pipeline dental schools. The University of California, San Francisco (UCSF) was funded by both foundations. However, we included UCSF in the "California Pipeline" category because all schools in the state of California received Pipeline program funding and are engaged in collaborative statewide recruitment and health policy initiatives.

In addition to contextual environment, we investigated a set of community-based dental education (CBDE) variables hypothesized to influence practice plans (Figure 1Go). Specifically, we examined variables representing 1) recruitment of underrepresented minority students, 2) CBDE curricula, and 3) extramural clinical rotations. Recruitment measures tested included "Importance of the following factors in influencing the decision to pursue dentistry as a career": a) high school or college counselor, b) recruitment by a dental school, c) pre- or post-baccalaureate dental career program, and d) awareness of workforce supply and demand trends in dentistry, using a five-point Likert scale ranging from 1 (low) to 5 (high). Two curricular measures were selected for the study: 1) adequacy of time devoted to your instruction in cultural competency was measured using three response categories: inadequate (25.1 percent), appropriate (68.5 percent), and excessive (6.4 percent); and 2) level of preparedness for providing "oral health care for racial, ethnic, and culturally diverse groups" was measured using a five-point Likert scale ranging from not well enough prepared to well prepared (mean=3.4, s.d.=0.96). Extramural clinical rotation measures included the following: 1) extramural clinical experiences influenced your practice location plans, using a five-point Likert scale ranging from not at all to very much; 2) "were your extramural clinical rotations positive or negative experiences in your dental education?" using a five-point scale ranging from very negative to very positive; and 3) "number of weeks providing dental care at extramural clinics," providing an open-ended space for seniors to indicate a response.

Student characteristics were categorized as demographic, socioeconomic, educational expenses and debt, and attitudes and beliefs regarding dental practice and access (Table 1Go). Student characteristics included gender, age, race/ethnicity, marital status, parents’ annual household income (as a proxy for student resources), father’s educational attainment, participation in a loan repayment program, and debt upon graduation. For the race/ethnicity variable, African Americans, Hispanics, and American Indians were combined into an "underrepresented minority" group (10 percent), and Asian/Pacific Islanders (26.3 percent) and whites (63.7 percent) were kept separate. Parents’ income was divided into three categories: 1) less than $30,000 (14.7 percent); 2) $30,001–$50,000 (13.2 percent); and 3) greater than $50,000 (72.1 percent). Since the Pipeline program is targeting low-income students, we chose relatively lower income thresholds to examine differences among the lowest income families. According to the U.S. Department of Health and Human Services 2003 HHS Poverty Guidelines, the 200 percent Federal Poverty Level income threshold for a family of three was roughly $30,000; we categorized $30,001–$50,000 as a moderate income level; the >$50,001 comprised the rest of the sample.

Four scales were constructed from the ADEA questionnaire to measure attitudes and beliefs: service orientation, social consciousness, entrepreneurial scale, and cultural awareness. Thind et al.31 developed the "service orientation" scale from three of nine reasons for selecting dentistry as a career (service to others, service to my own race or ethnic group, and opportunity to serve vulnerable and low-income populations) and the "socially conscious" scale from questions related to access (access to care is a societal good and right; access to oral health care is a major problem in the United States; assuring and providing care to all segments of society is an ethical and professional obligation; and everyone is entitled to receive basic oral health care regardless of his or her ability to pay). The "entrepreneurial scale" was developed from two items (opportunity for self-employment and high income potential) by Baumeister et al.34 The "cultural awareness" scale was constructed from questions related to preparedness to accept and respect patients of different races, ethnicities, and cultures; preparedness to integrate knowledge regarding cultural differences into treatment planning and care delivery; and the cultural and social environment of your school promotes acceptance and respect of students and patients of different races, ethnicities, and cultures.

In summary, the model posits that student characteristics, community-based dental education, and contextual variables influence practice plans of dental school seniors.

The dependent variable is ordinal and thus requires a multivariable model appropriate for ordered data. The ordered logit model is typically applied in this situation; however, this model relaxes the restrictive proportional odds assumption (also called the parallel regression assumption),35 which requires that the effects of the covariates in the log-odds of observing a score on a dependent variable are invariant to the cut point (or four categories) of the dependent variable. To evaluate the proportional odds assumption for the multivariable model, we performed a Brant test,36 which indicates the assumption did not hold for some covariates. Consequently, we used a generalized ordered logit model,37,38 which relaxes the proportional odds assumption. The generalized ordered logit model allows the effects of explanatory variables to change in addition to allowing for different intercepts. Variables that do not fulfill the proportional odds assumption can have varying effects on the dependent variable. (For more information on the generalized ordered logit model, refer to Long and Freese, Agresti, and McCullagh and Nelder35,39,40) We used the Wald {chi}2 test for assessing equality of odds ratios obtained from the generalized ordered logit model.

Data analyses took into account lack of independence in student reports from dental schools. This was necessary as students from the same schools are likely to have correlated measures due to shared environments resulting in intracluster correlation. Analyses that assume independence of the observations will generally underestimate the true variance and lead to test statistics with inflated Type I errors.41 Bivariate statistics were calculated with the SVY procedures in the Stata software package. These procedures were developed to analyze data from complex sampling structures including clustered samples,42 The generalized ordered logit models were fitted using Rogers’s generalization of the robust variance estimation procedure developed by Huber.43,44 Zero-order correlations among the covariates and collinearity diagnostics (i.e., a variance inflation factor <3) were examined as a check for multicollinearity.


   Results
Top
 Abstract
 Literature review
 Materials and Methods
 Results
Discussion
 Conclusions
 References
 
Fifty-two of the fifty-four accredited dental schools had graduating seniors in 2003 and returned ADEA surveys, resulting in an overall student response rate of 83.2 percent in 2003. This response rate is based on the total number of seniors who completed the survey from the fifty-two schools returning surveys divided by the total number of seniors graduating from the fifty-four schools (refer to Weaver et al.25 for detailed information). For the fifty-two schools included in the 2003 ADEA survey, we conducted a bias analysis to compare differences in race/ethnicity using the ADA statistics for dental senior students. As a cautionary note, we found some discrepancies in the total number of seniors reported by ADEA and the ADA, and some individuals failed to report race/ethnicity, so these findings are limited. Nevertheless, among the estimated 543 nonresponders, results showed slightly higher nonresponses among African Americans (ADA 6.3 percent versus ADEA 3.7 percent) and Hispanics (ADA 6.8 percent versus ADEA 5.3 percent) in the ADEA survey. Additionally, more ADEA respondents reported themselves as American Indian dental seniors than reported in the ADA, but the number and percentages of American Indian students nationally were extremely low (<1 percent in the 2003 ADEA survey).

Table 1Go describes candidate independent variables organized by construct specified in the measurement model (Figure 1Go) along with definitions and distributions. All candidate independent variables were tested in preliminary stepwise regression analysis (data not shown). Variables were tested in conceptually cohesive blocks as shown in Figure 1Go to identify significant predictors for the final model and to resolve multicollinearity concerns. Independent variables were trimmed from the final model if they did not reach significance (p<.20) or substituted due to collinearity concerns. For example, percent URM in the dental school had a variance inflation factor >3. This collinearity disappeared after removing percent URM in state legislature. Contextual variables were then added from distal to more proximal to the dependent variable: contextual environment, community-based dental education, and individual student-level variables. The regression model reported on Table 2Go shows only variables remaining significant in the final multivariable analysis.


View this table:
[in this window]
[in a new window]

  		Table 2. Variables predicting plans to care for underserved minority patients upon graduation
 
Table 2Go reports results using a generalized ordered logit model, which relaxes the proportional odds assumption. The far right column identifies the five variables reaching significance on the Brant test of proportional odds assumption: number of FQHCs in the county, preparedness for providing oral health care, URM, service orientation scale, and primary activity upon graduation (column 3). The effect of these variables is unequally distributed at each cut point of the dependent variable (0 percent; 1–10 percent; 11–24 percent), and an odds ratio is reported for each cut point. Additionally, we interpreted the results below. The overall Wald test was significant for all five variables, indicating each was significant in the multivariable model (data not shown). For variables not significant on the Brant test, the effect of the predictors on the dependent variable at each cut point was equally distributed and reporting a single odds ratio was sufficient (as shown in Table 2Go, column 2).

Three URM variables were tested in the model: percent URM in the county, percent URM in the dental school, and a variable representing individual URM students (reported under "student characteristics" below). "Percent URM in the school" was no longer significant when the student level URM variable was entered. However, the variable measuring "percent URM in the county" where the dental school was located remained a significant predictor. Seniors from dental schools in counties with greater percentages of underrepresented minorities were two times as likely to care for underserved patients upon graduation (OR=2.09, p<0.05).

The next predictor, "number of federally qualified health centers (FQHCs) in the county," was significant on the Brant test, which means it did not fulfill the proportional odds assumption and therefore had varying effects on each cut point of the dependent variable (column 3). Here the effect was only significant for the lowest cut point of the dependent variable (0% OR=0.92, p<.001). Attending a dental school with lower numbers of FQHCs in the county was associated with a decreased likelihood of providing care to minority patients. A final variable measured Pipeline school status. Compared to non-Pipeline schools, California seniors reported greater likelihood (OR=1.33, p<.01) of providing care to minority patients upon graduation (all five of the California schools are funded for a dental Pipeline program).

Turning to community-based dental education variables, the strongest predictors were "time devoted to cultural competency" and "preparedness for providing oral health care to racial, ethnic, and culturally diverse populations." Seniors who felt time devoted to cultural competency was "inadequate" had an increased likelihood of caring for underserved patients (OR=0.80, p<0.01). Preparedness for providing care to diverse groups was significant on the Brant test. Only one category of the dependent variable was significant (0% OR=1.36, p<.001) although data show a clear trend in the analysis. Individuals who perceive themselves as "less prepared" were more likely to provide care to minority patients upon graduation. More in-depth analysis revealed URM and Asian/PIs rated themselves as less prepared to care for racial, ethnic, and culturally diverse groups than white dental school seniors (chi square=19.0, p<.0001). However, parental income categories were not significantly different in regard to perceived preparedness (data not shown). Additionally, two predictors measuring the extramural clinical rotation experience showed a weak association with the practice plans dependent variable. Students who reported the "extramural clinical experience influenced practice location plans" and "extramural experience was positive" had a greater likelihood of serving minority patients upon graduation.

A final set of multivariable predictors indicated several student characteristics were associated with providing care to minority patients: URM and Asian/PI race/ethnicity (compared to whites), female gender, older age, lower parents’ income, social consciousness, service orientation, and entrepreneurial attitudes (Table 2Go). Two variables were significant on the Brant test: URM race/ethnicity and service orientation scale. Overall, the strongest predictor in the multivariable model showed URM seniors had a significantly greater likelihood of serving minority patients upon graduation (1–10% OR=2.43, p<0.001; 11–24% OR=2.97, p<0.001) compared to their white counterparts. URM dental school seniors were up to three times as likely to care for underserved minority patients upon graduation.

An intervening variable included in the model controlled for practice setting upon graduation: 1) private practice, 2) community clinic/government service, or 3) dental school/academic/postdoctoral residency. In terms of providing care to minority patients, individuals who were pursuing postdoctoral training or entering academic settings were not significantly different from those going into private practice. Not surprisingly, seniors planning to practice in community clinic or government settings had a significantly greater likelihood of serving minority patients.


   Discussion
Top
 Abstract
 Literature review
 Materials and Methods
 Results
 Discussion
Conclusions
 References
 
In this study, a comprehensive model was used to analyze contextual environment, community-based dental education, and student characteristics influencing practice plans of graduating seniors. Specifically, we examined variables significantly associated with plans to care for underserved minority patients. This baseline study analyzes 2003 ADEA data collected before the Pipeline, Profession, and Practice: Community-Based Dental Education program, sponsored by the Robert Wood Johnson Foundation and The California Endowment, was implemented in fifteen dental schools. A follow-up impact assessment will be conducted as part of the national evaluation study. We will analyze 2007 ADEA senior survey data and contextual variables to examine the impact of the Pipeline program on plans to care for underserved minority patients five years later.

Among the significant student characteristics, underrepresented minority (URM) race/ethnicity was the strongest predictor of plans to serve minority patients upon graduation. This is consistent with Pipeline program funding to increase recruitment and retention of URM students to dentistry, with the expectation that dental care access will increase in future years by virtue of their leadership and service.

Other significant predictors included female gender, older age (greater than twenty-nine years), Asian-Pacific Islander race/ethnicity, and lower parental income. These represent new research findings as our literature search yielded no studies indicating these variables were associated with providing care to the underserved. Recruiting increased numbers of URM and low-income students to dental schools is a major Pipeline program objective. Based on the findings of this study, perhaps the recruitment objective might be expanded to include other subgroups predisposed and motivated to respond to the critical shortage of dental care providers willing to care for underserved patients.

For some, this seems to be the heart of the whole issue. Does the dental profession see itself as part of the "higher calling" professions: clergy, teachers, physicians, and soldiers who respond to a moral/societal service obligation, or as business people who respond to personal financial goals? Dental schools might consider placing greater emphasis on applicants’ statement of purpose, level of social responsibility, and record of community service as higher priority criteria for selecting among competing dental school applicants.

Educational debt incurred by students was not a significant predictor of plans to treat minority patients when we accounted for primary activity upon graduation (data not shown). Community clinic or government service (versus private practice) was highly significant and accounted for the variance explained by educational debt. Graduates entering these settings have opportunities for loan repayment through government programs, such as the National Health Service Corps and the State Loan Repayment Program, to provide care in designated shortage areas immediately upon graduation, although government funding for these programs has been cut in recent years.

Regarding community-based dental education (CBDE), two significant predictors were reported in this study: perceptions of "inadequate time devoted to cultural competency" and "less preparedness for providing oral health care to racial, ethnic, and culturally diverse populations." Smith et al. in a 2006 issue of the Journal of Dental Education reported the degree of curriculum focus on treating patients from diverse backgrounds was significantly correlated with student and alumni intentions to provide care to these patients in their practices.20 In contrast, we found students who perceived time devoted to cultural competency as "inadequate" were most likely to care for minority populations. It was also disturbing to find individuals who perceive themselves as "less prepared" to care for diverse populations are those more likely to care for these patients upon graduation. If the Pipeline initiative is successful in stimulating didactic and clinical curricular reforms in dental schools, hopefully, future students will develop greater competency for providing care to diverse populations. In this study, extramural clinical rotation predictors showed a weak but significant effect on practice plans, and we expect to see an increasing influence of these variables over time if the Pipeline program is effective.

Two county-level contextual variables were significant predictors of practice plans: greater numbers of URM in the county were associated with a greater likelihood of providing care to minority patients, and lower numbers of FQHCs were associated with a decreased likelihood to care for minority patients. Obviously both contextual factors are associated with opportunities for increased contact and experience with minority populations. The only dental school variable predictive of practice plans was attending a California Pipeline dental school. The state of California is a melting pot of cultures, race/ethnicities, and multilingual subgroups. In Los Angeles alone, over 100 different languages and dialects are spoken in the county schools.

The practice plans dependent variable is limited because it measures dental school seniors’ "intent" to provide care upon graduation. Decades of research conducted by Fishbein and Ajzen have found "behavioral intentions" are relatively strong predictors of behavior itself.45–47 Of course, a more costly longitudinal study would be necessary to follow seniors through time to investigate actual practice settings and dental care provided to underserved minority patients. Experience has shown a longitudinal study of this nature would be further complicated by university Institutional Review Board restrictions that make tracking students and/or minority patient medical care utilization prohibitive.

Another possible limitation to Pipeline program evaluation is "diffusion of intervention." The national program office is actively engaged in disseminating information to academic, practice, policy, and foundation stakeholders throughout the field of dentistry. For example, dental school accreditation requirements might change, or non-Pipeline dental schools might begin to transform their academic programs without foundation support. From a societal and foundation perspective, greater fieldwide movement towards CBDE would be an optimal scenario; however, from an evaluation perspective, it might limit our statistical conclusions regarding the impact of the Pipeline program. To be aware of changes in the contextual environment, evaluators will continue to monitor Pipeline and non-Pipeline schools, as well as national and regional activities to transform dental education in the United States.


   Conclusions
Top
 Abstract
 Literature review
 Materials and Methods
 Results
 Discussion
 Conclusions
References
 
Addressing the barriers to dental care access involves the creation of policy, financial, and structural interventions to motivate providers to care for the underserved. If our society finds the will to address the dental care access crisis, comprehensive changes will need to occur at the federal, state, university, and dental school levels.

This study provides evidence that student characteristics, community-based dental education, and contextual environment significantly predict plans to care for underserved populations upon graduation. Recruiting underrepresented minorities to the health professions and improving and sustaining a cultural competency curriculum and extramural community rotations through accreditation standards are strategies the profession and dental schools can implement to increase potential access to dental care.48 Stimulating partnerships among dental schools and federally qualified health centers will offer more opportunities to train dentists and improve access to oral health care in the states. Federal and state policies will be needed to sustain improvements rendered by the Pipeline program to 1) provide financing for clinical training in community-based settings, 2) offer attractive employment opportunities and benefits for those practicing in community clinics and government/public health settings, and 3) improve reimbursement of dental services through public insurance programs.


   Footnotes
 
Dr. Davidson is Associate Professor, Department of Health Services, School of Public Health, University of California, Los Angeles, and Project Director and Co-Investigator for the National Evaluation Team; Ms. Carreon is Research Associate, Department of Health Services, School of Public Health, University of California, Los Angeles; Dr. Baumeister is Research Epidemiologist, Institute of Epidemiology and Social Medicine, University of Greifswald, Germany; Mr. Nakazono is Programmer Analyst, Department of Health Services, School of Public Health, University of California, Los Angeles; Mr. Gutierrez is Project Manager, Department of Health Services, School of Public Health, University of California, Los Angeles; Dr. Afifi is Professor, Department of Biostatistics, School of Public Health, University of California, Los Angeles; and Dr. Andersen is Principal Investigator of the National Evaluation Team and Professor Emeritus, Department of Health Services, School of Public Health, University of California, Los Angeles. Direct correspondence and requests for reprints to Dr. Pamela L. Davidson, University of California, Los Angeles, Department of Health Services, CHS 31-293, 650 C.E. Young Drive South, Campus Box 951772, Los Angeles, CA 90092-1772; 310-825-7188 phone; 310-206-3566 fax; Davidson{at}ucla.edu.


   REFERENCES
Top
 Abstract
 Literature review
 Materials and Methods
 Results
 Discussion
 Conclusions
 References
 

  1. American Dental Association Health Resources Center. Dental workforce model 2000–2020. Chicago: American Dental Association, 2002.
  2. Guay HG. Access to dental care: the triad of estimated factors in access-to-care programs. J Am Dent Assoc 2004; 135(6):779–85.[Abstract/Free Full Text]
  3. Valachovic RW. Dental workforce trends and children. Ambul Pediatr 2002; 2(2 suppl):154–61.[Medline]
  4. Butters JM, Winter PA. Professional motivation and career plan differences between African-American and Caucasian dental students: implications for improving workforce diversity. J Natl Med Assoc 2002; 94(6):492–504.[Medline]
  5. Hayes-Bautista DE, Kahramanian MI, Richardson EG, Hsu P, Sosa L, Gamboa C, et al. The rise and fall of the Latino dentist supply in California: implications for dental education. J Dent Educ 2007; 71(2):227–34.[Abstract/Free Full Text]
  6. Mertz E, O’Neil E. The growing challenge of providing oral health care services to all Americans. Health Aff (Millwood) 2002; 21(5):65–77.[Abstract/Free Full Text]
  7. Brown LJ, Lazar V. Trends in the dental health work force. J Am Dent Assoc 1999; 130(12):1743–9.[Abstract/Free Full Text]
  8. Douglass CW, Cole KO. The supply of dental manpower in the United States. J Dent Educ 1979; 43(5):287–302.[Abstract]
  9. Bailit HL, Formicola AJ, Herbert KD, Stavisky JS, Zamora G. The origins and design of the dental pipeline program. J Dent Educ 2005; 69(2):232–8.[Abstract/Free Full Text]
  10. Andersen RM, Davidson PL, Atchison KA, Hewlett E, Freed JR, Friedman JA, et al. Pipeline, profession, and practice program: evaluating change in dental education. J Dent Educ 2005; 69(2):239–48.[Abstract/Free Full Text]
  11. Lowell-Smith EG. Patterns of dentist location: Lansing, Michigan 1950–1991. Soc Sci Med 1993; 36(4):569–73.[Medline]
  12. Beazoglou TJ, Crakes GM, Doherty NJ, Heffley DR. Determinants of dentists‘ geographic distribution. J Dent Educ 1992; 56(11):735–40.[Abstract]
  13. Fyffe HE, Pitts NB. Origin, training, and subsequent practice location of Scotland‘s General and Community dentists. Community Dent Oral Epidemiol 1989; 17(6):325–9.[Medline]
  14. Hanes PJ, McKnight MC. Criteria for selection of a practice location among senior dental students at the University of Mississippi. J Dent Educ 1984; 48(2):102–4.[Medline]
  15. Butters JM, Winter PA. The effects of gender and race on practice pattern preferences of dental students. J Am Coll Dent 1999; 66(3):39–46.[Medline]
  16. Solomon ES, Hayes MJ. Gender and the transition into practice. J Dent Educ 1995; 59(8):836–40.[Abstract]
  17. Winter PA, Butters JM. An investigation of dental student practice preferences. J Dent Educ 1998; 62(8):565–71.[Abstract]
  18. Crossley ML, Mubarik A. A comparative investigation of dental and medical students’ motivation towards career choice. Br Dent J 2002; 193(8):471–3.[Medline]
  19. Schenkein HA, Best AM. Factors considered by new faculty in their decision to choose careers in academic dentistry. J Dent Educ 2001; 65(9):832–40.[Abstract]
  20. Smith CS, Ester TV, Inglehart MR. Dental education and care for underserved patients: an analysis of students‘ intentions and alumni behavior. J Dent Educ 2006; 70:398–408.[Abstract/Free Full Text]
  21. Mofidi M, Konrad TR, Portefield DS, Niska R, Wells B. Provisions of care to the underserved population by National Health Service Corps Alumni Dentists. Health Dent 2002; 62(2):102–8.
  22. DeCastro JE, Matheson PB, Panagakos FS, Stewart DC, Feldman CA. Alumni perspectives on community-based and traditional curricula. J Dent Educ 2003; 67(4): 418–26.[Abstract]
  23. Li LB, Williams SD, Scammon DL. Practicing with the urban underserved: a qualitative analysis of motivations, incentives, and disincentives. Arch Fam Med 1995; 4: 124–33.[Abstract/Free Full Text]
  24. Eliason BC, Guse C, Gottlieb MS. Personal values of family physicians, practice satisfaction, and service to the underserved. Arch Fam Med 2000; 9:228–32.[Abstract/Free Full Text]
  25. Weaver RG, Haden NK, Valachovic RW. Annual ADEA survey of dental school seniors: 2003 graduating class. J Dent Educ 2004; 68(9):1004–27.[Free Full Text]
  26. Weaver RG, Ramanna S, Haden NK, Valachovic RW. U.S. dental school applicants and enrollees: 2003 and 2004. J Dent Educ 2005; 69(9):1064–72.[Abstract/Free Full Text]
  27. Rosenberg HM, Cucchiara AJ, Helpin ML. Dental students’ attitude to gender roles. Soc Sci Med 1998; 47(11):1877–80.[Medline]
  28. Roberts MW, Lieff S, Seale NS. Factors that affect career choices of pediatric dentistry advanced education students. Pediatr Dent 1997; 19(5):317–20.[Medline]
  29. Butters JM, Winter PA. The effects of gender and race on practice pattern preferences of dental students. J Dent Educ 1998; 63(6):565–72.
  30. Dharamsi S, MacEntee MI. Dentistry and distributive justice. Soc Sci Med 2002; 55(2):323–9.[Medline]
  31. Thind A, Atchison K, Andersen R. What determines positive student perceptions of extramural clinical rotations? An analysis using 2003 ADEA senior survey data. J Dent Educ 2005; 69(3):355–62.[Abstract/Free Full Text]
  32. Davidson PL, Andersen RM, Wyn R, Brown ER. A framework for evaluating safety-net and other community-level factors on access for low-income populations. Inquiry 2004; 41(1):21–38.[Medline]
  33. Andersen RM, Davidson PL. Improving access to care in America: individual and contextual indicators. In: Andersen RM, Rice TH, Kominski GF, eds. Changing the American health care system: key issues in health services, policy and management. San Francisco: Jossey-Bass, 2006.
  34. Baumeister S, Davidson PL, Carreon DC, Nakazono TT, Gutierrez JJ, Andersen R. Improving access for special care populations: what determines interest in special care dentistry among graduating dental students in the U.S. Special Care Dent; in press.
  35. Long JS, Freese J. Regression models for categorical and limited dependent variables using Stata. College Station, TX: Stata Press, 2003.
  36. Brant R. Assessing proportionality in the proportional odds model for ordinal logistic regression. Biometrics 1990; 46(4):1171–8.[Medline]
  37. Kang Fu V. Estimating generalized ordered logit models. Stata Technical Bulletin 1998; 8:160–4.
  38. Williams R. Gologit2: generalized logistic regression/partial proportional odds model for ordinal variables, 2005. At: www.nd.edu/~rwilliam/gologit2/gologit2.pdf. Accessed: July 18, 2005.
  39. Agresti, A. Categorical data analysis. New York: Wiley, 1990.
  40. McCullagh P, Nelder JA. Generalized linear models. London: Chapman and Hall, 1983.
  41. Williams RL. A note on robust variance estimation for cluster-correlated data. Biometrics 2000; 56(2):645–6.[Medline]
  42. Stata Corp. Stata survey data reference manual, Release 9. College Station, TX: Stata Press, 2005.
  43. Rogers WH. Regression standard errors in clustered samples. Stata Technical Bulletin 1993; 13:19–23.
  44. Huber PJ. The behavior of maximum likelihood estimates under non-standard conditions. Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics and Probability. Berkeley: University of California Press, 1967.
  45. Fishbein M, Ajzen I. Belief, attitude, intention, and behavior: an introduction to theory and research. Reading, MA: Addison-Wesley, 1975.
  46. Ajzen I. The theory of planned behavior. Organizational Behav Hum Decision Processes 1991; 50:179–211.
  47. Goldthorpe JH. Rational action theory for sociology. Br J Sociol 1996; 49(2):167–92.
  48. Hewlett E, Davidson PL, Nakazono TT, Baumeister SE, Carreon DC, Freed JR. Dental school seniors’ perceived preparedness to care for racially and culturally diverse groups. Unpublished manuscript.



This article has been cited by other articles:


Home page
 J Dent EducHome page
M. Mofidi and A. Gambrell
Community-Based Dental Partnerships: Improving Access to Dental Care for Persons Living with HIV/AIDS
J Dent Educ., November 1, 2009; 73(11): 1247 - 1259.
[Abstract] [Full Text] [PDF]

Home page
 J Dent EducHome page
L. C. Leviton
Foreword
J Dent Educ., February 1, 2009; 73(2_suppl): S5 - 7.
[Full Text] [PDF]

Home page
 J Dent EducHome page
R. M. Andersen and P. L. Davidson
Introduction to the Evaluating the Dental Pipeline Program Report
J Dent Educ., February 1, 2009; 73(2_suppl): S10 - 14.
[Full Text] [PDF]

Home page
 J Dent EducHome page
D. C. Carreon, P. L. Davidson, and R. M. Andersen
The Evaluation Framework for the Dental Pipeline Program with Literature Review
J Dent Educ., February 1, 2009; 73(2_suppl): S23 - 36.
[Full Text] [PDF]

Home page
 J Dent EducHome page
P. L. Davidson, T. T. Nakazono, A. Afifi, and J. J. Gutierrez
Methods for Evaluating Change in Community-Based Dental Education
J Dent Educ., February 1, 2009; 73(2_suppl): S37 - 51.
[Full Text] [PDF]

Home page
 J Dent EducHome page
P. L. Davidson, T. T. Nakazono, D. C. Carreon, J. Bai, and A. Afifi
Practice Plans of Dental School Graduating Seniors: Effects of the Pipeline Program
J Dent Educ., February 1, 2009; 73(2_suppl): S283 - 296.
[Full Text] [PDF]

Home page
 J Dent EducHome page
R. M. Andersen, D. C. Carreon, J.-A. Friedman, S. E. Baumeister, A. A. Afifi, T. T. Nakazono, and P. L. Davidson
What Enhances Underrepresented Minority Recruitment to Dental Schools?
J Dent Educ., August 1, 2007; 71(8): 994 - 1008.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Davidson, P. L.
Right arrow Articles by Andersen, R. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Davidson, P. L.
Right arrow Articles by Andersen, R. M.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS