| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Critical Issues in Dental Education |
Key words: dental schools, recruitment of minority students, diversity
Submitted for publication 01/10/07; accepted 05/14/07
 |
Abstract |
|---|
 Top Abstract Literature reviewMethodsResultsDiscussionReferences |
|---|
Racial and ethnic minority groups experience a disproportionately higher level of oral health problems and have limited access to dental care.3 Increasing minority representation in the dental profession may be one way to address the problem of access. Members of minority groups are more willing to seek care if that care is rendered by someone with whom they can identify.4,5 Minority patients also prefer ethnic- and language-concordant health care providers.6,7 Other studies show that minority dental students plan to treat higher percentages of minority and low-income patients and patients from rural and inner-city communities than do white students.4,8–10
Despite efforts by dental schools to recruit URM, the number of African American, Hispanic, and Native American students in dental schools remains low.8,11,12 A report issued by ADEA found that, in 2004, URM comprised 12.4 percent of the applicants and 11.6 percent of first-year enrollees.13 Asian/Pacific Islanders and whites comprised 69.7 percent of applicants and 71.1 percent of first-year enrollees. The proportion of URM applying and enrolling in U.S. dental schools is far less than the proportion of URM in the communities served by the dental school. For example, during the 2003–04 academic year, 7 percent of dental students enrolled at UCLA and USC were Hispanic,8 while 46.5 percent of the Los Angeles population were Hispanic.14 Also in 2003–04, total African American enrollment at all U.S. dental schools was 5.41 percent,8 while 12.8 percent of the U.S. population were black.14 The proportion of URM dentists also remains significantly lower than the proportion of URM in the U.S. population.4,8,15–17 Currently, about 6.8 percent of professionally active dentists are URM, while 27.9 percent of the U.S. population are URM.14
|
Literature Review |
|---|
|
TopAbstractLiterature reviewMethodsResultsDiscussionReferences |
|---|
Surveys conducted at the University of Pittsburgh School of Dental Medicine identified dental school characteristics associated with matriculation.18 For all 1994–95 applicants, reputation of the dental school was the number one factor influencing the decision to attend the institution; among the 2000–01 applicants, location was ranked highest. Curriculum, tuition costs, and facilities were also important. Less important in the ranking system were class size, financial aid, and research opportunities. Additionally, a study at the University of Pennsylvania School of Dental Medicine reported that its successful minority recruitment program involved major efforts around areas of leadership, financial support, and institutional commitment to create an inclusive environment for diverse cultures.19 The recruitment program included an administration and mission commitment to a diverse student body, scholarships and financial support, partnerships with two historically black colleges, and a program directed at URM high school seniors. With the implementation of this program, the total URM enrollment increased from eleven in 1989 to fifty-six in 2002.
Factors affecting recruitment at medical and other health professions schools have also been examined. Similar to dental schools, the availability of financial aid, lower tuition costs, and participation in pre-enrollment programs were identified.20,21 Agrawal et al. surveyed deans and faculty at 144 U.S. accredited medical schools and found that lack of minority faculty and minority role models was a commonly cited barrier to URM recruitment.22 However, recruiting and enrolling URM in dental schools are more problematic than in medical schools. During much of the 1990s, medicine had a higher proportion of URM students than all health professions other than public health.23 URM enrollment in U.S. medical schools increased 43 percent between 1986 and 1994.12 For allopathic medicine, the percentage of total URM enrollees in 1999–2000 was 13.8 percent compared to 9.7 percent for dentistry.23 In 2003–04, the Association of American Medical Colleges continued to note an increase in minority applications to medical schools.17 Further, the proportion of dentists who are URM is lower than the proportion of physicians who are URM. In 1999–2000, 4.1 percent of the dentists were URM compared to 9.4 percent of the physicians.17
Although recruitment factors have been previously studied, much of the literature focuses on individual and school characteristics. Characteristics of the communities where schools exist have been largely ignored. Our study uses a comprehensive framework to investigate community, dental school, and community-based dental education characteristics that may influence the proportion of URM in dental schools. The study also contributes to previous work by examining whether graduating senior students’ perceptions of the curriculum and extramural clinical rotation programs influence the decision to attend an institution. Understanding the reasons why some schools attract more URM students may lead to improved URM recruitment strategies for dental schools.
|
Methods |
|---|
|
TopAbstractLiterature reviewMethodsResultsDiscussionReferences |
|---|
presents a comprehensive framework for understanding factors that might influence the proportion of URM students in dental schools. The framework suggests that the community or larger environment in which the dental school is located, characteristics of the dental school itself, and the nature of community-based dental education (CBDE) within the dental school will all be associated with URM recruitment. The dental school has the least control over community characteristics, somewhat more control over some characteristics of the school itself, and most control over the CBDE curriculum. The selection of a dental school is jointly made by the student and the school. Students choose a school on the basis of its environment or location, the characteristics of the school, and the curriculum that they expect to experience. Schools choose students based on their characteristics and how they will fit with the schools’ characteristics and curriculum.
|
. While Table 1 is lengthy, it is important to the development of a comprehensive approach to the study of URM recruitment and provides some rationale for variable selection that might help in the interpretation of the findings.
|
Dental school-level variables also come from a variety of data sources and include characteristics of the dental school and CBDE characteristics. Pipeline status (national, California, or non-Pipeline) was derived from the National Program Office that sponsors and coordinates the Pipeline program. Schools were also categorized as either primarily privately or publicly supported institutions according to the ADA. Two variables were constructed to represent the school’s mission statement (commitment to recruit URM and commitment to provide care to URM). After the mission statements were collected from dental school websites, two researchers reviewed and coded the statements independently. Educational costs, including first-year resident tuition, total financial aid awarded, and first-year total costs come from the ADA survey of predoctoral dental education, conducted annually and completed by dental school administrators. The percent of URM clinical faculty also comes from the ADA.
In addition to general dental school characteristics, we investigated a set of CBDE variables possibly related to the school’s recruiting of URM students. All of these variables were derived from the 2003 ADEA survey of dental school seniors. This survey is distributed annually to graduating seniors at all accredited U.S. dental schools in order to obtain information about the financing of dental education, indebtedness, practice and postdoctoral education plans upon graduation, and opinions about adequacy of time devoted to instruction and preparedness in various areas of predoctoral education. Weaver et al. provide a detailed description of the methodology.25 Fifty-two of the fifty-six accredited dental schools returned ADEA surveys in 2003. Two schools did not respond, and two other schools (the Arizona School of Dentistry & Oral Health and the University of Nevada, Las Vegas) did not have graduating seniors in 2003. We constructed these variables by calculating the school-level mean score of graduating student responses to the ADEA survey to questions concerning the following: 1) CBDE curriculum—adequacy of time devoted to instruction of cultural competency and social and behavioral sciences; humanistic treatment of patients at the main school clinics and at the extramural clinical site; and preparedness to provide oral health care for racial, ethnic, and culturally diverse patients, to accept and respect diverse patients, and to integrate cultural differences into treatment planning; and 2) extramural clinical rotations—mean number of weeks in extramural clinical rotations; adequacy of time spent in extramural clinical rotations; the influence of extramural experience for improving ability to care for diverse groups; and perception of experiences in extramural clinical rotations. A variable measuring the dental school’s aggregate cultural and social environment also comes from an ADEA question: "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 scale from strongly disagree (1) to strongly agree (4).
The data source for the outcome variables was the 2003 ADA survey of predoctoral education, which provides the ethnicity of the first-year enrolled dental students. There are three outcome variables measuring each dental school’s recruitment of URM: ratios of first-year URM students, first-year Hispanic students, and first-year African American students to all first-year students for the year 2002–03. Because ratios of Hispanic students and African American students were highly skewed, we conducted our analysis after transformations using the square root function, which made the variables nearly normal (refer to Table 1
). We were also interested in recruitment of Native Americans, but since the majority of schools recruited no Native Americans, the analysis used to predict ratios is not possible. For the year 2002–03, only ten schools recruited first-year Native American students. The total number of these students was twenty-three. Two schools accounted for a substantial proportion of Native American students. The University of Oklahoma had six students, and the University of North Carolina had five students.
The unit of analysis was the dental school. Three dental schools (Howard University, Meharry Medical College, and University of Puerto Rico) had large majorities of URM students. Howard University had an entering class in 2002 that included 75.6 percent African Americans, and Meharry University had 88.9 percent African Americans. The University of Puerto Rico’s entering class had 100 percent Hispanic students. While these schools are central to the education of URM students, these large proportions of URM prevent us from including them in the ratio analyses for this project because the Pipeline program is concerned with increasing minority enrollment in the other schools. Thus, forty-nine dental schools responding to the ADEA survey in 2003 were used in the analysis. The overall student response rate for the forty-nine schools was 87 percent.
Data Analysis
Descriptive statistics for our study variables are included in Table 1. Bivariate correlation coefficients were used to describe the unadjusted relationships between our outcome and explanatory variables, as shown in Table 2.
|
. To select the predictors shown in our final models, we first examined a correlation matrix between the outcome and all candidate predictors, selecting only those predictors whose correlations with their outcomes were significant at the p<.20 level, as shown in Table 2. From this set of predictors, we used the best subset regression procedure based on the Cp criterion26 to choose the variables to be used in the OLS final model shown in Table 3.
|
|
Results |
|---|
|
TopAbstractLiterature reviewMethodsResultsDiscussionReferences |
|---|
presents the mean and range for independent and dependent variables. For the community policy variables, 12 percent of the schools are located in states with no adult Medicaid benefits, while at the other extreme, 31 percent are located in states with full Medicaid dental benefits. The states where the schools are located have on average 15 percent URM state legislators, with a range of 0 to 28 percent. The population characteristics of the counties in which the schools are located shows that, on average, a third of the residents are URM, with a range of 4.6 percent to 70.0 percent. The median percent of blacks is 15.2 percent, for Hispanics 6.7 percent, for Asian/Pacific Islanders 3.1 percent, and for foreign-born population 9.0 percent. The counties had an average low income population of 32 percent. Almost one fifth of the residents had less than a high school education, and almost all reside in urban areas. The dentist to population ratio averaged 6.1 per 10,000 residents, ranging from 3.9 to 8.2.
The dental schools in this study had the following characteristics: one quarter of the schools are participating in the Pipeline program; a third are privately supported, while two-thirds are publicly supported. When we explored the mission statements of the schools, we found that 22 percent included a specific commitment to recruit URM students and 18 percent mentioned providing care to URM patients. Educational costs for first-year residents in the 2002–03 school year averaged $18,173 for tuition and $26,985 overall. On average, 7.1 percent of the schools’ clinical faculty are URM, with a range from 1.0 percent to 18.8 percent. Graduating students’ agreement that their school promoted acceptance and respect of different races and cultures (on a four-point scale from strongly agree [4] to strongly disagree [1]) averaged 3.2.
Based on a three-point scale of inadequate (1), appropriate (2), excessive (3), graduating students on average rated time devoted to cultural competency at 1.8 (range 1.5–2.2) and time devoted to social and behavioral determinants of health at 1.9 (range 1.8–2.1). Graduating students were also asked to rate the quality of care provided on a five-point scale from excellent to very poor. Students, on average, rated the humanistic treatment of patients as close to "very good" in both the school clinics (3.9) and extramural clinics (3.8) (4 on a five-point scale is described in the questionnaire as "very good").
Further, graduating students were asked how well prepared they were in three areas and answered on a five-point scale from 1 (not well enough prepared) to 5 (well prepared). The average scores across schools for preparedness for oral health care for diverse groups (3.4), to accept and respect diverse patients (3.4), and to integrate cultural differences in treatment planning (3.2) all suggest students rated their preparedness on average as slightly above prepared (3 on a five-point scale is described in the questionnaire as "prepared"). Graduating students reported on average 6.6 weeks spent on clinical rotations, and that time was, on average, appropriate (1.8 on a three-point scale representing excessive, appropriate, and inadequate). In judging the influence of their extramural clinical rotation experience, graduating students’ average score across schools for improving ability to care for diverse groups was 3.1 (on a five-point scale from not at all to very much); on a five-point scale rating the nature of the experience from very negative to very positive, the average score was 3.8, that is, close to positive.
Our outcome variables show that the average percentage of total URM in the first-year class of dental schools was about 9 percent, with a range from 0 to 21 percent. The average percent of first-year Hispanic students was 4.6 percent, with a range of 0 to 18 percent, while first-year African Americans averaged 3.3 percent and ranged from 1 to 14 percent.
Determinants of Recruitment
Percent URM First-Year Students.
Table 2 shows that, among community characteristics, schools located in states with emergency Medicaid dental benefits recruited proportionately more URM. Also, schools located in counties with a lower ratio of practicing dentists recruited more URM in the first-year class. Among dental school characteristics, schools that had lower first-year resident tuition, lower total financial aid awarded, and lower first-year total costs recruited proportionately more URM. Schools with a higher percent of URM clinical faculty were also found to recruit more URM students. One clinical rotation characteristic was significantly correlated with URM recruitment: schools where graduating students felt that their extramural experience improved their ability to care for diverse groups had higher URM recruitment.
Table 3 shows the results of the multiple regression after the subset regression method selected three variables as being the "best" model. This analysis confirmed the bivariate results in that the most important factors predicting higher proportions of URM were a higher percent of URM clinical faculty and graduating students’ perception that their extramural clinical experience improved their ability to care for diverse groups.
Percent Hispanic First-Year Students.
Turning to the analysis of specific ethnic groups making up the URM, Table 2 shows that, among community characteristics, schools located in states with a higher percent of URM legislative members recruited proportionately more Hispanics in the first-year class. Also, schools located in counties with a higher percent of Hispanic, Asian/Pacific Islander, and foreign-born persons recruited higher proportions of Hispanic students. Further, schools located in urban counties had a significantly positive impact on Hispanic recruitment. Among the dental school characteristics, we found that schools having a mission statement that specifically states a commitment to recruit URM students was positively correlated to Hispanic recruitment. Also positively correlated were a higher percent of URM clinical faculty and greater number of weeks the graduating students spent in extramural rotations. The bivariate correlation analysis also showed an inverse relationship between graduating students’ feelings of preparedness to care for diverse groups and Hispanic recruitment.
Table 3 shows that the best subset regression model included four variables. It showed a higher percent of URM clinical faculty, graduating students’ perceiving that they were less well prepared to treat diverse groups, and greater number of weeks students spent in rotations significantly predicted higher proportions of Hispanic recruitment.
Percent African American First-Year Students.
The final column in Table 2 shows correlations between explanatory variables and the proportion of African American students in the first-year class. Similar to percent URM in the first year, schools located in states with emergency Medicaid dental benefits recruited proportionately more African Americans. Schools located in counties with a higher percent of blacks and a lower percent of Hispanics were significantly associated with first-year African American enrollment. Among dental school characteristics, schools that were part of the National Pipeline program recruited more African Americans, while schools of California Pipeline status recruited fewer African Americans. Further, schools with lower total financial aid awarded recruited greater proportions of African Americans in the first year. Similar to percent URM and percent Hispanic in the first year, schools with a higher percent of URM clinical faculty were found to recruit proportionately more African Americans. Finally, schools whose graduating dental students perceived their school’s cultural and social environment as accepting and respectful of diverse groups recruited more African Americans.
Among the CBDE factors, several variables were significantly correlated with percent African Americans in the first year. Schools where graduating students perceived the time devoted to instruction in cultural competency and social and behavioral determinants was adequate recruited fewer African American students. Alternatively, schools where graduating students perceived the patients at the main school clinic received humanistic treatment were positively correlated with African American recruitment. Also, schools whose graduating students felt prepared for oral health care for diverse groups, to accept and respect diverse patients, and to integrate cultural differences into treatment planning were positively correlated with proportions of African American students in the first-year class.
Table 3 shows that the subset regression procedure selected four predictors. Similar to the bivariate correlations, percent blacks in the county was found to be a significant predictor of African American recruitment in the final model. Also positively associated with African American recruitment were being a National Pipeline school and graduating students feeling more prepared to integrate cultural differences into treatment planning. Negatively associated with recruitment of African Americans in the first year was total financial aid awarded by the school.
|
Discussion |
|---|
|
TopAbstractLiterature reviewMethodsResultsDiscussionReferences |
|---|
Multivariable results also confirmed that certain dental school characteristics and CBDE characteristics may increase recruitment of underrepresented minorities. Among dental school characteristics especially important was the finding that a higher percent of URM clinical faculty predicted greater URM and Hispanic recruitment. Percent URM clinical faculty was not selected for the final model predicting selection of African Americans; however, it was found to be a predictor in the bivariate correlations. These findings support previous suggestions that increased presence of URM clinical faculty is particularly attractive to URM student recruits. We acknowledge the problem regarding the general difficulty of recruiting dentists to faculty positions. The problem is greater for recruiting URM faculty members when we consider the small number of URM dentists in the country. However, some Pipeline schools have successfully recruited URM faculty. There are also several federal programs that exist to increase URM faculty including the Faculty Loan Repayment Programs and Faculty Development Fellowships.27 Contrary to our expectations, less financial aid was associated with greater recruitment of African American students. Apparently, African Americans are less likely to attend the private high cost schools that also tend to offer the largest financial aid packages.
Greater number of weeks spent in extramural clinical rotations, graduating student perceptions that their extramural experiences improved their ability to care for diverse patients, and graduating student perceptions of preparedness to integrate cultural differences into treatment planning were all positive CBDE recruitment factors. These are factors that the schools control and can modify to increase enrollment of URM. URM applicants appear attracted to schools that provide more community experience for their students, as well as schools that provide current students with experiences that make students feel more prepared to integrate cultural differences into treatment planning. It may also be that the schools emphasizing community-based dental education are more effective in recruiting URM students for reasons not necessarily specified in our model.
On the other hand, our results showed that schools in which senior students felt more prepared to treat diverse groups recruited relatively fewer Hispanics. It is unclear why concerns about being less prepared to treat diverse patients would predict more Hispanic recruitment. According to research conducted by Hewlett et al., racial and ethnic groups may value preparedness to care for diverse groups more highly than their white counterparts.28 It may be that URM applicants are attracted to schools where the current students are particularly concerned about being prepared to care for diverse groups and are thus seeking even more experiences that would increase this preparedness.
Limitations of our study include a relatively few degrees of freedom (forty-nine dental schools) and a relatively large number of possible independent variables (twenty-six). While we used the Bonferroni adjustment in our bivariate analyses to limit the number of significant findings we might observe due to chance and screening procedures to limit the number of independent variables in our multivariable analyses, we still had limited ability to detect the importance of the independent variables in our comprehensive model. Furthermore, due to the cross-sectional nature of our study, we cannot posit causality with certainty given our research design. Since our study is limited to students matriculating in dental schools, we were not able to compare the characteristics of rejected applicants with successful ones. Future studies concerned with minority recruitment might survey all students, including the unsuccessful ones, who applied for dental school. The effect of the racial composition of the local population on the proportion of URM in the dental school might also be attributable to other unobserved factors, possibly the political culture that comes along with an ethnically diverse neighborhood. Neighborhood ethnic diversity might also be related to the location of the school. Students are not only choosing a dental school; they are also looking for an interesting and diversified place to live, which is usually a larger city that attracts a mix of racial, ethnic, and cultural groups. Reputation of the dental school is another important factor that was not included in our models. Although reputation or perception of reputation might be difficult to measure, it is expected to be highly predictive for at least the student’s primary choice of dental school.
Despite its limitations, this study advances our understanding of URM recruitment in dental schools. It presented a comprehensive framework for studying the links among community context, school-level factors, and proportion of URM students enrolled. Second, to the best of our knowledge, this is the first study of the relationship among population characteristics, state-level policy factors (e.g., Medicaid dental benefits, legislative representation), and the proportion of URM students in the dental school. It provides evidence that contextual factors are relevant to the number of URM students recruited to dental schools.
Community characteristics (e.g., URM in the population), dental school characteristics (e.g., presence of URM faculty), and community-based dental education characteristics (e.g., emphasis on community rotations) can all play a role in the recruitment and selection process of underrepresented minority students. A more global view of these factors may add to the ability to attract more minority students to dentistry and, consequently, better meet the needs of our minority populations who continue to lack access to care.
|
Footnotes |
|---|
This work is supported by the Robert Wood Johnson Foundation (RWJF) and The California Endowment (TCE). Ronald Andersen received support from the UCLA/DREW Project EXPORT, NCMHD, P20MD000148/P20MD000182.
|
REFERENCES |
|---|
|
TopAbstractLiterature reviewMethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. C. Sinkford, J. F. West, R. G. Weaver, and R. W. Valachovic Modeling Mentoring: Early Lessons from the W.K. Kellogg/ADEA Minority Dental Faculty Development Program J Dent Educ., June 1, 2009; 73(6): 753 - 763. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
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]
|
|
|
|
|
|
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]
|
|
|
|
|
|
R. M. Andersen, J.-A. Friedman, D. C. Carreon, J. Bai, T. T. Nakazono, A. Afifi, and J. J. Gutierrez Recruitment and Retention of Underrepresented Minority and Low-Income Dental Students: Effects of the Pipeline Program J Dent Educ., February 1, 2009; 73(2_suppl): S238 - 258. [Full Text] [PDF]
|
|
|
|
|
|
A. Thind, K. A. Atchison, R. M. Andersen, T. T. Nakazono, and J. J. Gutierrez Reforming Dental Education: Faculty Members' Perceptions on the Continuation of Pipeline Program Changes J Dent Educ., December 1, 2008; 72(12): 1472 - 1480. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Sewell, N. Hawley, K. Kingsley, S. O'Malley, and C. C. Ancajas Recent Admissions Trends at UNLV-SDM: Perspectives on Recruitment of Female and Minority Students at a New Dental School J Dent Educ., November 1, 2008; 72(11): 1261 - 1267. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
