Parents usually are the primary decision makers on matters affecting their childrens health and health care.1,2 Therefore, parents, along with clinicians, play a key role in attempts to achieve the best oral health outcomes for their young children. Considering parents central role in ensuring the well-being of young children, it is important to explore their perceptions about their childrens oral health. These perceptions can affect the preventive dental care children receive at home and their use of professional dental services.3 Parents assessments also can provide some indication of childrens need for professional dental care, which, in turn, has implications for the formulation of dental health policies.4 Moreover, understanding factors associated with parents perceptions about their childrens oral health may help the dental community understand some of the reasons why children do not receive the dental care they need. These considerations are especially important for preschool-aged children because their inability to verbalize their emotions and distress increases their dependence on adults. Parental characteristics and beliefs thus may be an important consideration in attempts made to improve preschool-aged childrens oral health.
Understanding parents perceptions of their childrens oral health can help dentistry overcome barriers that parents encounter in accessing dental care for their children.
Research on oral health perceptions has focused on school-aged children5 and adults.6,7 However, no population-based study has reported parents perceptions of the oral health of children younger than 6 years of age. A few studies810 have explored parents assessments of their childrens general health. Results from these studies suggest that parents social demographic characteristics can affect their perceptions about their preschool-aged childrens health. Mothers living in poverty and belonging to minority groups are more likely to report the general health of their children to be worse than are nonminority mothers who are not living in poverty. Importantly, parents global ratings of their childrens health are associated with the presence of acute and chronic disease.9
Studies among adults have used a behavioral model that considers ones social demographic characteristics in combination with beliefs, perceptions of disease and the presence of disease as major factors contributing to perceptions about ones oral health.6,11,12 Gift and colleagues6 found that almost two-thirds of U.S. adults considered their natural teeth to be in excellent, very good or good condition. Those who rated their teeth as being in poor condition had more missing teeth, carious teeth and periodontal disease compared with adults who rated their oral health as excellent. When the authors controlled for clinical status, they found that oral health was rated worse by those who were older, were minorities, were economically disadvantaged, reported the need for more dental care and perceived their general health to be poorer.
Similar associations for adults assessment of their childrens oral health have not been investigated in detail. However, factors identified in studies of adults might affect not only perceptions of their own oral health, but also their perceptions about their childrens oral health. We conducted a study to examine such associations to better understand the nature and correlates of parents perceptions of their preschool-aged childrens oral health.
 |
METHODS
|
|---|
In this study, we explored the relationships between the characteristics of parents of children aged 2 to 5 years and the parents assessments of the childrens oral health, using data from a cross-sectional national survey. We hypothesized that parents characteristics, beliefs and subjective assessments of their childrens need for dental care, along with the childrens social demographic characteristics and actual dental disease status, would be correlated with parents perceptions about their childrens oral health.
Data source and analytical sample.
The data we used in this study were from the Third National Health and Nutrition Examination Survey (NHANES III), which was conducted during the period 1988 through 1994. NHANES III was based on a multistage probability sample of the noninstitutionalized civilian population aged 2 months and older residing in the 50 states. Further details about NHANES III and the oral health component of this survey are reported elsewhere.1315
We obtained the analytic sample of 2- to 5-year-olds by merging the Household Youth and Examination data files. The Household Youth data file contains responses to questions asked of primary caregivers, and the Examination data file contains information pertaining to children and youth (018 years) who underwent a complete physical examination, including a dental examination. In NHANES III, children 23 months of age and younger did not receive a complete examination for dental caries and, thus, we excluded them from our study.16
Preliminary examination of data revealed many missing values for two key variables: the need for dental treatment and preventive services. Of the 4,300 children 2 to 5 years of age who had dental examination results, 3,424 (79.6 percent) had complete information. As children with partial information did not differ from those with complete information with respect to their social demographic characteristics (age, sex, race/ethnicity, education of adult family member, family income, family size, mothers country of birth, urbanization classification or region of country), we included only children with complete data in our analyses.
Variable construction.
We based the dependent variable on the question asked of the primary caregiver: "How would you describe the condition of [childs name]s natural teeth?" Answers were recorded using a Likert scale with response options coded 1 through 5 (excellent = 1, very good = 2, good = 3, fair = 4, poor = 5).
Three variables captured parent-defined beliefs about their childrens dental and medical health status and needs. Parent-defined need for dental care included two variables: perceptions of need for dental treatment and need for cleaning teeth. The need for treatment variable was a dichotomous variable constructed from responses to five questions that asked parents if they felt their children needed to "have teeth filled or replaced, teeth pulled, gum treatment, relief of pain or dental work to improve appearance." A positive response to any of the five questions was coded as a yes; otherwise, answers were coded as a no. Parental perception of need for cleaning was a dichotomous variable (yes, no) that indicated whether the parent felt that his or her childs teeth needed cleaning.
Because of extensive evidence of strong correlations between perceptions of general health and oral health among adults,6,7,1719 we explored whether parents made similar associations when reporting their childrens health. We derived parental perceptions of childrens general health from a question that asked the parent to rate his or her childs general health on a five-point scale (1 = excellent, 2 = very good, 3 = good, 4 = fair and 5 = poor). We combined the fair and poor categories, as there were a small number of responses in the poor category.
We determined a childs dental health status by the presence or absence of carious tooth surfaces. We calculated and converted a count of carious tooth surfaces into a dichotomous variable (0,
1) that indicated whether the child had any untreated dental caries.
The social demographic characteristics in our study included parent, child and family variables. Parental characteristics included parents education (less than high school, high school or more than high school) and mothers country of birth (United States versus Mexico/other). Child demographic characteristics included the childs age in individual years (ages 2 to 5 years), sex and race/ethnicity (non-Hispanic black, non-Hispanic white, Mexican-American, other). Family characteristics included family income as a percentage of the federal poverty line (FPL) (0100 percent, 101200 percent, 201300 percent and more than 300 percent of FPL), family size (up to four people, five or more people), urbanization classification (metropolitan area, non-metropolitan area/other) and region of the country (Northeast, Midwest, South, West).
We also considered the childs insurance status. Having insurance increases the likelihood that a child will receive professional dental care,20 and the resulting dental visit might affect dental health perceptions.6 However, because of limitations imposed by the data set, we were unable to include a dental insurance variable. We were able to include variables that measured participation (yes, no) in two public assistance programs: the Food Stamp Program and the Special Supplemental Nutrition Program for Women, Infants and Children. These two variables likely identified a large percentage of study subjects who lacked private insurance and were covered by Medicaid.
Parents of children who have visited a dentist differ from parents of those who have not, according to important social demographic characteristics such as being better educated and having higher income.21,22 Furthermore, a visit to the dentist can result in the provision of care and the alleviation of symptoms, with the potential for an impact on parents perceptions.6 For this reason, we examined parental perceptions among those who had used dental care for their children and those who did not use such care. The question about dental visits asked the parent how long ago the child had visited a dentist or dental hygienist preceding the interview. We created a dichotomous variable indicating whether the child had ever visited a dentist or dental hygienist.
Analysis strategy.
We used preliminary descriptive analyses to assess the distribution of responses for all study variables. We set a minimum cell sample size at 30 in deciding whether to collapse categories of variables.23 We evaluated overall associations between the dependent and explanatory variables using
2 tests when the dependent variable was categorical, and one-way analysis of variance when the dependent variable was linear.
We estimated linear regression models to examine the correlates of parents perceptions of their preschool-aged childrens oral health. The dependent variable was the five-point, ordinal rating of childrens dental health. We used linear regression models in preference to other models, such as ordinal logistic regression, to maintain consistency with the methodology used by Gift and colleagues.6 Furthermore, global ratings of health made on Likert-type scales are known to yield a continuum from excellent to poor.24 We created dummy variables for all explanatory variables with more than two response categories to allow for interpretation of the marginal effects of variable categories. We used correlation coefficients and tolerance to examine for multi-collinearity.25,26 We entered all explanatory variables into the linear regression model at the same time.
Because social demographic characteristics acting in combination can shape peoples life experiences and may affect their health perceptions,27 we examined the joint effect of these variables, in addition to each of their effects. For this purpose, we used the F statistic to test for multiple exclusion restrictions. We also conducted regression analysis that was conditional on the childs past dental visits to determine if parental perceptions were modified by dental use. We conducted all multivariate analyses using statistical software (SUDAAN, Release 8.0, Research Triangle Institute, Research Triangle Park, N.C.) that was able to accommodate the complex sampling design used in NHANES III.26
 |
RESULTS
|
|---|
The average age of children in the sample was 2.5 years. A total of 48.4 percent of the children were girls, 64.0 percent were white, 16.0 percent were non-Hispanic black, 9.1 percent were Mexican-American, and 10.9 percent were from other racial groups; the sample was distributed almost equally across all family income categories. These and other social demographic characteristics of the sample are presented in Table 1
. In addition, 50.6 percent of the children had never visited a dentist, and 19.0 percent had one or more carious tooth surfaces.