- Top of page
Objectives: Most studies on the provision of dental procedures have focused on Medicaid enrollees known to have inadequate access to dental care. Little information on private insurance enrollees exists. This study documents the rates of preventive, restorative, endodontic, and surgical dental procedures provided to children enrolled in Delta Dental of Wisconsin (DDWI) in Milwaukee.
Methods: We analyzed DDWI claims data for Milwaukee children aged 0-18 years between 2002 and 2008. We linked the ZIP codes of enrollees to the 2000 US Census information to derive racial/ethnic estimates in the different ZIP codes. We estimated the rates of preventive, restorative, endodontic, and surgical procedures provided to children in different racial/ethnic groups based on the population estimates derived from the US Census data. Descriptive and multivariable analysis was done using Poisson regression modeling on dental procedures per year.
Results: In 7 years, a total of 266,380 enrollees were covered in 46 ZIP codes in the database. Approximately, 64 percent, 44 percent, and 49 percent of White, African American, and Hispanic children had at least one dental visit during the study period, respectively. The rates of preventive procedures increased up to the age of 9 years and decreased thereafter among children in all three racial groups included in the analysis. African American and Hispanic children received half as many preventive procedures as White children.
Conclusion: Our study shows that substantial racial disparities may exist in the types of dental procedures that were received by children.
- Top of page
Dental caries prevalence and incidence among specific groups of children in the United States and other industrialized nations have decreased, but the disease continues to be the single most prevalent condition of childhood in the United States with a large portion of this disease remaining untreated (1). Kaste et al. reported that 25 percent of 5- to 17-year-old children experience 80 percent of all dental caries in the United States (2). Another study based on the 2007 National Survey of Children's Health reported that children from racial and ethnic minority groups and low socioeconomic status (SES) have fewer preventive dental visits (3,4). Regular dental visits combined with the provision of preventive dental procedures are effective in preventing dental caries (5). However, studies on the provision of common dental procedures in different population groups and payer types are limited.
Studies on the provision of dental procedures have reached similar conclusions that African Americans are significantly less likely to have preventive or restorative procedures and more likely to have their teeth extracted compared with Whites (6-8). However, Gilbert et al. study was restricted to adults (6), Okunseri et al. analyzed data from a single dental school (7), and Manski et al. used self-reported data from a nationally representative sample without validating the information with dental records (8). Findings from these studies have profound limitations given that approximately 50 percent of children are enrolled in private dental insurance in the United States (3).Therefore, documenting the rates and patterns of use of different dental procedures in this population is important for program planning and policy development.
Our study is focused on children enrolled in private dental insurance in Milwaukee, Wisconsin, a racially/ethnically diverse population. Milwaukee, the largest city in the state, has a large income disparity between inner-city and suburban neighborhoods and is cited as one of the most racially segregated cities in the United States (9). In addition, the 2000 US Census reported that 45-47 percent of individuals living in inner-city Milwaukee are below the federal poverty level. The median family income ranged from $18,936 in one of the inner-city ZIP codes to $106,681 in a suburban ZIP code selected in this study (10). These differences have persisted through to the 2010 Census (11). We examined rates and patterns for different dental procedures provided to children living in the inner-city and suburban Milwaukee neighborhoods enrolled in Delta Dental of Wisconsin (DDWI).
- Top of page
We analyzed DDWI claims and enrollment data for children aged 0-18 years living in 46 ZIP codes in Milwaukee and its suburbs from 2002 through 2008. The claims dataset contained information on age, ZIP code of residence, date of treatment delivery, and procedure code for the treatment provided. The enrollment dataset had information on the number of insured children for each year broken down by ZIP codes, gender, and age. Children were categorized into one of five age groups; 0-3 years, 4-6 years, 7-9 years, 10-14 years, and 15-18 years. When a child's ZIP code changed during the year, we used the ZIP code of residence from the last dental visit. The age at the last dental visit was used as the age of the child for that year. Per capita annual income for ZIP codes was recorded in units of $10,000 and was used as one of the predictor variables in the models.
We grouped dental treatment procedure codes, based on Current Dental Terminology codes, into four major treatment categories: preventive, restorative, endodontic, and surgical. Preventive procedures included oral prophylaxis (D1120), fluoride varnish (D1206), fluoride gel applications (D1203), and dental sealant placement (D1351). Restorative and endodontic procedures included all the billing codes for those procedures (D2000-D2999 and D3000-D3999), respectively. Surgical procedures included extraction of deciduous teeth (D7111), extraction of erupted teeth (D7140), and extraction of erupted teeth requiring elevation of mucoperiosteal flap (D7210). The claims data were aggregated to obtain the number of procedures of each type performed during a calendar year for each enrollee. The number of enrollees without any dental visit during a year was inferred by subtracting the number of children with at least one claim for each year/ZIP code and age-group category from the total enrollment. These enrollees were counted as having zero procedures of any type during that year.
Estimation of race/ethnicity from US Census data
Person-level race or income information was not available in the claims dataset, so we estimated race and imputed per capita income based on census data. We used information from the 2000 US Census data to estimate racial/ethnic compositions within ZIP codes for enrollees. The proportions of Whites, African Americans, and Hispanics or Latino of any race living in each of the ZIP codes were obtained from census data. The racial/ethnic composition of a ZIP code based on the US Census information was regarded as the same for all enrollees within that ZIP code in the dataset. We also assumed that the racial proportions in the ZIP codes were constant throughout the study period. Weighted averages were used to estimate the number of preventive, restorative, endodontic, and surgical dental procedures received by children of each race. Regression models were used to assess the affect of the predictors, including race, on the number of procedures of a given type.
We calculated the rate ratios for the dental procedures received by White children in various age categories using the 7- to 9-year category as the reference group. To compare the average number of dental procedures that were provided to children across racial groups, we used White children from a particular age group as the reference group and calculated the rate ratios for procedures that were provided.
Multivariable analysis was performed using Poisson regression modeling on the number of procedures per year. The final multivariate model included income, age, race (as two variables giving the proportion of Black and Hispanic residents within the ZIP code), year when the treatment was delivered, and the interaction between race and age. All analyses were performed using SAS version 9.2 (SAS Institute Inc Cary, NC, USA). A statistical significance level (alpha) of 0.05 was used throughout. This study was approved by the Marquette University's Institutional Review Board.
- Top of page
The total number of enrollees over the 7-year study period was 266,380 and yearly enrollment ranged from 37,824 in 2002 to 41,000 in 2008. Table 1 shows the characteristics of the study population including estimates derived from statistical models for each of the different racial/ethnic groups. Approximately, 64 percent, 44 percent, and 49 percent of White, African American, and Hispanic children, respectively, had at least one dental visit during the study period.
Table 1. Characteristics of Study Populations and Estimates Derived from Statistical Models for Each of the Racial/Ethnic Groups and Overall Utilization
| ||Total number of enrollees from 2002-2008||Estimates derived from statistical models|
|Gender|| || || || |
|Age|| || || || |
|Total number of procedures|| || || || |
|Dental care utilization rates||60.9%||64.6%||44.4%||49.4%|
Figure 1 shows the correlation between the average number of preventive and restorative procedures received by children and the proportion of Whites within the ZIP codes for the year 2008. The average number of preventive procedures per child increased in all age groups with an increasing proportion of White children in a ZIP code. The slope of this relationship, which could be interpreted as a relative degree of racial disparity, was greatest for the age groups 4-6 and 7-9 and smallest for ages 0-3 and 15-18. These last two age groups also had much lower average numbers of preventive procedures per child. The average number of restorative procedures per child appeared similar across all age groups with the exception of the 0-3 age group where there was a much lower average. The slope of restorative procedures to proportion of White children was positive for the three age groups above age 7 and slightly negative for age groups 0-3 and 4-6. The average number of restorative procedures received increased with age while the number of preventive procedures decreased with increasing age across all age groups except the 0-3 group, which had a very low average.
Figure 1. Relationship between average number of preventive and restorative procedures and racial proportions in ZIP codes by age group. The lines are based on a linear fit of the log-transformed averages.
Download figure to PowerPoint
The results from multivariable analyses are presented in a series of tables, each concentrating on a particular factor. Table 2 concentrates on the effect of age: the rate ratios for the different procedures received by White children from different age groups. Children in the 7-9 year age group (reference group) had the highest number of preventive procedures of all age categories. Those in the 0-3 and 15-18 year age groups received 20 percent and 13 percent as many preventive dental procedures compared to children in the reference group. Children in older age groups, with the exception of children in 10-14 year age group, received more restorative procedures than the 7-9 year olds. Children 4-6 years of age had received the highest number of endodontic procedures followed by the 7-9 year old children. The groups with the highest number of extractions were children in 10-14 year age group followed by 7-9 year old children.
Table 2. Rate Ratios of Dental Procedures among White Children of Different Age Groups as Determined from Multivariate Analysis
| ||0-3 years||4-6 years||7-9 years||10-14 years||15-18 years|
|Preventive||0.208* (0.204-0.211)||0.947 (0.936-0.957)||1.000 reference||0.740* (0.733-0.748)||0.133* (0.131-0.136)|
|Restorative||0.129* (0.123-0.135)||0.969* (0.945-0.993)||1.000 reference||0.779* (0.762-0.797)||1.229* (1.203-1.256)|
|Endodontic||0.176* (0.151-0.207)||1.357* (1.237-1.487)||1.000 reference||0.567† (0.513-0.626)||0.897* (0.817-0.984)|
|Extractions||0.026* (0.022-0.031)||0.359* (0.336-0.385)||1.000 reference||1.145* (1.096-1.196)||0.316* (0.297-0.337)|
Table 3 shows results from the multivariable analyses for determining the patterns of dental procedures received by Whites, compared to African Americans and Hispanics. African American and Hispanic children received significantly fewer preventive procedures than did White children in the same age group. For most age groups, minority children had half as many preventive procedure compared to White children. African American and Hispanic children in younger age groups had more restorative, endodontic, and extraction procedures than White children. This discrepancy decreases with age and in older age groups, minority children receive fewer procedures than Whites. The exact age of crossover varies by procedure type and race, but it is usually around the ages 7-9.
Table 3. Rate Ratios from Multivariate Analysis for Comparing Dental Procedures across Racial/Ethnic Groups
| ||0-3 years||4-6 years||7-9 years||10-14 years||15-18 years|
|African Americans||0.494* (0.458-0.533)||0.471* (0.451-0.493)||0.542† (0.52-0.564)||0.463† (0.446-0.481)||0.343† (0.316-0.373)|
|Hispanics||0.544† (0.459-0.644)||0.478† (0.433-0.527)||0.547† (0.499-0.6)||0.416† (0.38-0.455)||0.317† (0.256-0.392)|
|African Americans||1.248† (1.078-1.444)||0.769† (0.712-0.831)||0.652† (0.605-0.703)||0.635† (0.594-0.68)||0.538† (0.504-0.573)|
|Hispanics||8.058† (6.515-9.965)||1.941† (1.691-2.228)||1.085 (0.939-1.254)||0.831† (0.724-0.954)||0.548† (0.475-0.632)|
|African Americans||3.770† (2.528-5.623)||1.506† (1.178-1.925)||1.074 (0.82-1.407)||0.741† (0.558-0.983)||0.690† (0.531-0.896)|
|Hispanics||14.461† (7.818-26.749)||4.616† (3.09-6.895)||2.981† (1.902-4.674)||1.001 (0.579-1.73)||1.710† (1.058-2.764)|
|African Americans||3.741† (2.223-6.296)||1.942† (1.589-2.374)||0.892 (0.763-1.043)||0.379† (0.328-0.438)||1.119 (0.926-1.352)|
|Hispanics||43.903† (22.725-84.817)||7.609† (5.464-10.595)||1.730† (1.29-2.319)||0.469† (0.348-0.632)||1.431 (0.93-2.201)|
Table 4 examines the effect of calendar year, and socio-economic status: the rate ratios for changes in dental procedure patterns over the study period and across various per-capita income groups. Over the course of the seven year study period, there was an increase in the average number of preventive procedures provided to children from all racial/ethnic groups. The average number of restorative and endodontic procedures increased among African American and Hispanic children, and decreased among White children. There was an increase in the number of teeth extracted among children from all racial/ethnic groups. Table 4 also illustrates that as the per-capita income within a ZIP-code increased, children receive more preventive services and fewer restorative, endodontic, and extraction procedures.
Table 4. Rate Ratios from Multivariable Analysis for Comparing Dental Procedures across Racial/Ethnic Groups
|Odds ratios for change in the mean number of procedures per calendar year for racial groups|
|Whites||1.010* (1.008-1.012)||0.979* (0.975-0.983)||0.972* (0.957-0.988)||1.145* (1.134-1.156)|
|African Americans||1.034* (1.026-1.043)||1.015† (1.002-1.029)||1.075* (1.024-1.128)||1.106* (1.068-1.146)|
|Hispanics||1.11* (1.089-1.133)||1.021 (0.995-1.048)||1.067 (0.982-1.159)||1.119* (1.042-1.201)|
|Impact of per capita income on procedures (increase in annual income in increments of $10,000)|
|All racial groups||1.11* (1.106-1.116)||0.912* (0.903-0.921)||0.803* (0.769-0.838)||0.973† (0.951-0.995)|
- Top of page
This study expands existing information on the receipt of dental procedures and service utilization among children enrolled in a private dental plan. Certain study limitations must be acknowledged. Individual race/ethnicity information was extrapolated from ZIP code level data and not self-reported. This method could lead to under- or overestimation of racial/ethnic proportions in the different ZIP codes. However, geocoded measures that use ZIP codes are useful in inferring characteristics about persons living in those areas such as their race/ethnicity and SES (12). Similarly, using average number of procedures based on the aggregate for all children who received care could potentially lead to under or overestimation of the rates of procedures. Parental education, SES, and certain neighborhood characteristics shown to influence either dental caries experience or ability to access dental services were not included in our study (13-16) due to the complexity of disaggregating this information from ZIP codes. In addition, due to the small sample of Hispanic children there is a potential for high error rates with estimates of dental procedures.
In this study approximately 60 percent of the children had at least one dental visit in a given year. This result is slightly lower (67.2 percent) than what was reported based on data from the National Health and Nutrition Examination Survey (NHANES) 1999-2004 (17), but higher (45 percent) than what was reported based on data from the 2004 Medical Expenditure Panel Survey (MEPS) (18). However, in the MEPS, children with private insurance had a utilization rate of 57.5 percent. In addition, utilization rates for children from racial and ethnic minority populations in this study were 44 percent for African American and 49 percent for Hispanic children. The results based on NHANES (1999-2004) show rates of approximately 64 percent and 56 percent for African American and Mexican American children, and the 2004 MEPS show 34 percent and 33 percent for these same groups (17,18). The discrepancy in utilization rates between all studies for all groups could be attributed to the differences in sampling methodology.
Furthermore, we found that African American children above 7 years were less likely to receive most procedures. Lewis et al. (4) reported that, even when they had private insurance, Black children were less likely to have a dental visit than White children. This pattern was seen even after adjusting for household income indicating that perceptions of dental care may be an important factor governing utilization patterns (4). Also, some of the racial differences identified by us in this study may have been due to complex interactions and associations between race and other variables. Such variables include parental perception of oral health and need (19), cultural beliefs on efficacy of prevention (20), as well as education and health literacy. These are some of the independent predictors of dental care utilization that tend to be associated with race. The ZIP codes used in our study included a high proportion of minorities living in regions with lower SES, lower education levels, and lower dentist to population ratios. Hence, we suspect that both person/family level and community level factors may have had a significant impact on the patterns of utilization and disparities identified by us. Our findings do show a need for educational and policy initiatives that can help increase dental utilization in this population to help improve oral health of the children.
Trends and disparities in receipt of dental procedures
Eklund analyzed trends in dental procedures provided to enrollees of Delta Dental insurance and reported a decrease in restorative, endodontic, and extraction procedures from 1992-2007 (21). These findings are consistent with our results for White children with decreased restorative and endodontic procedures, but different for racial and ethnic minorities with increased rates of receipt of these same procedures. In addition, there was an increase in the number of preventive and extraction procedures provided to children from all racial groups.
In our analysis of preventive dental procedures, we found that African American and Hispanic children of all ages received significantly fewer preventive procedures when compared with White children of the same age, despite being enrolled in the same private dental insurance plan. Our findings are consistent with previous studies on dental and medical procedures which confirmed the existence of racial disparities in preventive procedure utilization among insured populations (4,22-24). We found that for most age groups, children from minority groups had fewer restorative procedures compared with White children of the same age group similar to what was reported by studies on dental procedures among adult populations (6-8).
Few studies have examined racial disparities in endodontic procedures and extractions among children. Studies on adults reported that people from minority groups were more likely to have their teeth extracted than to follow through with a root canal treatment (25,26). In our study, we found that Hispanic children had more extractions or endodontic procedures when compared with White children. However, African American children in younger age groups (0-6 years) had more and those in older age groups had fewer extractions and endodontic procedures to have endodontic procedures and extractions when compared with White children. We suspect that fewer dental visits among African American children in this study could be the primary reason for fewer treatment procedures in this group. The differences in number of visits may partially explain the apparent higher odds for receiving restorative, endodontic, or extraction procedures among White children. Previous studies have shown that people from minority groups seek dental services when they experience symptoms such as pain (6,19). Endodontic treatment, as opposed to simple restorations, may be the indicated choice of treatment by the time services are sought. We think this could be a reason for fewer restorative procedures in this group.
We found that White children in the 10- to 14-year-old age group had significantly more tooth extractions when comparisons were made across racial groups, and also across various age groups of White children. In order to ascertain the reason for higher odds of extraction in this group, we performed additional analyses to examine if the extractions were related to orthodontic treatment, but found no such relation (data not shown). We also examined the frequencies of each of the surgical treatment codes that we used and found that the odds of extraction of deciduous teeth (D7111) and erupted teeth (D7140) was higher for White children in the 10- to 14-year age group (data not shown).