|Home | About | Journals | Submit | Contact Us | Français|
Low-income African American children experience high rates of dental caries compared to the general population. Knowledgeable and efficacious caregivers can play an important role in caries prevention. This study developed and evaluated four brief measures reflecting mothers’ potentially modifiable cognitions associated with African American children’s oral health: (1) knowledge about appropriate bottle use (KBU); (2) children’s oral hygiene (KCOH); (3) oral health-related fatalism (OHF); and (4) oral health-related self-efficacy (OHSE).
Questions were selected based on reviews of the health promotion and oral health literature, with input from low income African American caregivers of young children. Reliability and validity were evaluated using survey and dental examination data from 719 low-income African American mothers and their children ages 1–5 years.
Alpha reliabilities ranged from .76 – .91. KCOH was significantly associated with mothers’ oral health perceptions and children’s caries status. OHSE was significantly positively correlated with children’s brushing frequency and with mothers’ subjective perceptions of children’s oral health, which was in turn significantly associated with children’s caries status.
Results support the reliability and validity of the new measures. If confirmed by further research, these measures can be used to develop tailored educational and cognitive-behavioral interventions to reduce oral health disparities.
Although the prevalence of dental caries has dramatically decreased over the last three decades in the United States, significant levels of disease persist and are concentrated among socioeconomically disadvantaged groups.1 Several studies have found that dental caries remains prevalent among lower-income households and among African-American adults and children.2–5 Early Childhood Caries (ECC) is used to describe extensive tooth decay in children under age six. The effects of ECC can interfere with optimal growth and development, and significantly impair quality of life and functioning in fundamental social activities.6 Due to examination difficulties and variation in ECC clinical diagnostic criteria,7 the prevalence of ECC in the United States is not entirely clear, but the baseline prevalence from the Third National Health and Nutrition Examination Survey 1988–94 (NHANES III) for all children aged 2–4 is 18%, and for African Americans is 24%.8
A recent systematic review identified 106 specific sociodemographic, oral hygiene, dietary, and bacterial risk factors for ECC in children under age six.9 Although most of the sociodemographic risk factors for oral health disparities are not readily amenable to change, a young child’s primary caregiver, usually the mother, has control over many of the behavioral factors that can increase caries risk, and can play a critical role in determining whether or not children develop ECC. Very young children are dependent on their mothers to attend to their oral hygiene and feed them. Inappropriate bottle use patterns, such as the addition of sweeteners to the liquid and prolonging exposure of sugary liquids at bedtime,10,11 and later age at weaning12 have been linked to ECC.
Importantly, mothers can modify their child’s high sugar intake, poor and infrequent brushing, and whether or not the child falls asleep with a bottle of sweet liquid. However, lack of knowledge of ECC risk factors and how to care for newly erupting teeth in very young children may prevent caregivers from meeting young children’s oral health needs. And, although accurate knowledge is necessary, it is not always sufficient, as it is not uncommon for educationally based intervention efforts to only have modest and short-term effects.13,14 Perceptions of self-efficacy, or judgments about ability to perform certain behaviors like brushing a child’s teeth, are an important basis for action and should also be considered in interventions. Those with a strong sense of efficacy tend to set higher goals, persist longer at tasks despite obstacles, and devote more effort and energy to the task compared to those without a strong sense of efficacy.15
Although self-efficacy has been positively associated with favorable health outcomes,16 research investigating caregivers’ self-efficacy in populations of children at high risk for dental diseases is sparse. The research that has been done generally supports a positive relationship between efficacy and oral health behaviors and outcomes,17–22 although there are some conflicting findings.11 Reisine and Litt 11 investigated brushing habits, dietary sugar intake, social class, stressful life events, dental health locus of control, dental self-efficacy, tooth decay, and bacteria in saliva among Connecticut Head Start children. Although not specific to brushing children’s teeth, an 8-item oral health efficacy measure was associated with higher caries rates in children. In a related study, maternal efficacy was an important predictor of sugar intake, which in turn predicted children’s bacterial levels and dental caries.21
Mothers’ dental knowledge, sense of fatalism about children’s oral health, and self-efficacy can influence engagement in oral health promoting behaviors, and in turn, caries risk. These cognitions are potentially modifiable, and can be enhanced to motivate engagement in health promoting and disease preventing behaviors such as regular toothbrushing for young children and appropriate bottle feeding practices. Accurate assessment of caregivers’ knowledge and perceptions about children’s oral health can aid in the planning and implementation of tailored educational and cognitive-behavioral interventions. However, to date, there are no reliable and validated measures available to assess these cognitions.
This paper presents four new measures developed to assess African American mothers’ knowledge about children’s oral hygiene needs and appropriate bottle use, sense of fatalism, and level of self-efficacy related to brushing her young child’s teeth regularly. The measures were developed for a population-based sample of low-income African American mother-child dyads living in Detroit, a group that experiences a significant amount of dental disease. This paper describes the reliability and validity of these newly developed measures and how they relate to the ECC status and brushing practices of children in this population.
Data for this study are from the first phase (2002–3) of a seven-year longitudinal study being conducted by the Detroit Dental Health Project (DDHP). The DDHP is one of five centers funded by the National Institute of Dental and Craniofacial Research to reduce oral health disparities across the United States (NIDCR U-54 DE 14261).23 The DDHP research program focuses on understanding the social, familial, biological, and neighborhood context of oral health among low-income African American families in Detroit. A multistage area probability sample design was used to select the population-based study sample of African American families. We selected the 39 poorest Census tracts in the city of Detroit based on 2000 Census data. Families were eligible if they had at least 1 child less than 6 years of age at baseline and were below 250% of federal poverty line. The final population based study cohort included 1,021 children and their primary caregivers.
The procedures, possible discomforts or risks, as well as possible benefits were explained fully to the subjects involved and their informed consent was obtained prior to the investigation. The research was reviewed and approved by the University of Michigan Health Sciences Institutional Review Board. Trained staff conducted face-to-face interviews with caregivers at the Dental Assessment Center in Detroit. Caregivers were surveyed about their beliefs, behaviors, diet, background, living conditions, and a wide array of psychosocial factors using a series of structured questionnaires. All children and primary caregivers received a dental exam. The present study analyzed data from biological mothers of children aged one through five (N=749).
The investigators constructed a questionnaire to measure a broad array of hypothesized determinants of children’s oral health, including caregivers’ knowledge, attitudes, and beliefs about children’s oral health; feelings of self-efficacy related to brushing at bedtime in the face of a variety of potential barriers; the child’s brushing habits; and caregiver’s subjective perceptions of the child’s oral health. A total of 27 statements reflecting knowledge, attitudes, and beliefs about oral health were generated based on reviews of the pediatric and general dental literature. Item statements ranged from expectations about future oral health status and feelings of fatalism about dental caries to knowledge about how to care for children’s teeth, baby bottle use, and the process of developing caries. Caregivers were asked to express their level of agreement with each statement on a Likert scale (1=strongly disagree, 5=strongly agree). Several items were similar in content but framed differently, and not all items were included in the final scales.
A nine-item measure of oral health-related self-efficacy (OHSE) was adapted from a health promotion measure for exercise based on the Transtheoretical Model, a widely used health behavior model that conceptualizes health behavior along a continuum of stages ranging from pre-contemplation through maintenance.24–26 Only one other recently published efficacy instrument specifically measures caregivers’ ability to care for several aspects of children’s oral health needs, including toothbrushing, however this instrument was not available at the time the present study was undertaken.17
Children’s one-week brushing frequency was measured by their mothers’ report of the total number of times the child’s teeth were brushed in the last week by the child, caregiver, or someone else. The mother’s perception of her child’s oral health status was based on her answer to the following survey question: “How would you describe the condition of his/her (the child’s) mouth and teeth? Would you say… excellent, very good, good, fair, or poor.” The few “poor” responses were collapsed with the “fair” category.
Detailed examinations of each tooth surface were performed by a team of four dentists, although two performed the majority of the core examinations. The International Caries Detection and Assessment System (ICDAS) criteria for caries detection developed by an expert panel during a series of workshops were used in the DDHP study. The dental team established reliable and consistent ratings based on this criteria system while following a specific examination protocol. All assessments were made visually on clean and dry teeth. The inter-rater reliability Kappa coefficient was 0.83 overall, with 94–98% rater agreement. The intra-rater reliability Kappa coefficient was 0.74.27 Each child was classified as either caries-free, having ECC, or having severe ECC (S-ECC) based on the case definition proposed by an expert panel for research with preschool-aged children.28 This definition has been used in prior research,29 and was adopted by the American Academy of Pediatric Dentistry Council on Clinical Affairs in May 2000.30
Some survey questions had a small number of missing values, usually less than 4% of all responses. Most missing values were imputed with Imputation and Variance Estimation software (IVEware), except for brushing frequency variables. The brushing variables had inconsistencies in the responses, a higher frequency of missing values (about 10%), and complicated skip patterns, so were not imputed. Imputation was done for individual items before calculating scores for scales. This process allowed for a more consistent sample size to be used in analyses. The final analyses were also adjusted with sample weights created to account for the design features of the study. Specifically, the weight adjusts for the unequal probability of selection, participant non-response, and a post-stratification control to make the sample representative of the population of children in Detroit in terms of race, gender, and age.
Exploratory factor analysis (EFA) using varimax rotation and item-total analyses were used to inform the construction of scales from the 27 knowledge, belief, and attitude items and nine efficacy items.31 Validity was assessed by calculating the reliability of each scale. The relationships between the scales and the dental outcomes (children’s brushing frequency and oral health ratings) were examined. Pearson’s correlation coefficients were computed to observe the association between two continuous variables, chi-square tests of association were used for two categorical variables, and ANOVAs (including post-hoc analyses) were used to assess the association between continuous and categorical variables. All analyses were conducted in SAS-callable SUDAAN version 832 statistical software to account for the complex sample design and produce robust variance estimations.
The final sample included 719 cases with no missing data on the study variables. Nearly half (46%) of the mothers reported an annual household income of less than $10,000, with an average household size of four, and 49% had not completed high school. Children in the study averaged 3.05 years, and 52% were females. Disease was highly prevalent: one third of the children had at least one cavitated carious lesion and 54.6% were classified as having ECC or S-ECC. The age-adjusted prevalence of disease was even higher for the older children in the group, described in detail elsewhere.33
Children’s teeth were brushed about nine times per week either by the child or someone else (weighted mean= 8.96, standard error (se) = 0.22), which is slightly more often than once every day. Although there was a wide range of responses (0–40), expected peaks at seven times per week (once a day) and fourteen times per week (twice a day) were observed. Brushing was not related to oral health ratings or disease severity, but the association between mothers’ perceptions of her child’s oral health and the disease severity measure was significant (χ2 = 72.2, p<.0000), and most of the 20% of children with fair/poor ratings had cavities. Among children classified as having ECC or S-ECC, 72% had fair/poor oral health ratings, and 70% of children who did not have any cavities had excellent ratings.
The results of the EFA yielded eight factors initially, of which only four suggested scales had acceptable reliabilities (Cronbach’s alphas above 0.55). These four suggested scales reflected OHSE, knowledge about appropriate bottle use (KBU), knowledge of children’s oral hygiene needs (KCOH) and oral health fatalism (OHF). Item-total analyses revealed the combination of items within each suggested scale that yielded the highest internal consistency of each scale. The composition of the final scales was determined by the combination of items suggested by the EFA and item-total analyses that yielded the highest Cronbach’s alpha above 0.70, a commonly accepted standard of high internal consistency. However, combinations of the four items identified by the EFA to constitute OHF consistently yielded relatively low alphas. The associations between the individual OHF items and dental outcomes were examined, and it was found that a single fatalism statement, “most children eventually develop dental cavities,” seemed to best reflect this construct. The final KBU, KCOH, and OHSE scales and the OHF measure are summarized in Table 1. Items were averaged to compute a single scale score for KBU, KCOH, and OHSE.
The KBU, KCOH, and OHSE scales had high Cronbach’s alphas (0.76, 0.77, and 0.91 respectively), indicative of good internal consistency reliability. Average scores on both KBU (4.00) and KCOH (4.51) were relatively high on the 1–5 scale and revealed that mothers were quite knowledgeable about appropriate bottle use practices and children’s oral hygiene needs. Mothers’ average OHSE was relatively high as well (2.99 on a 1–4 scale), indicating that most felt moderately confident about their ability to ensure their children’s teeth were brushed at bedtime. Responses to the individual efficacy items were also positively skewed, and the most typical response was “moderately confident.” Relatively few mothers (less than 12%) reported “not being confident” about brushing their child’s teeth in each situation. This response was most common in the situations where the mother was tired (11.7%), too busy (11%), or depressed (10%). Yet, despite being aware of preventive oral health behaviors and generally feeling efficacious about their children’s brushing, the majority (568 out of 719, or 78.8%) of mothers endorsed a fatalistic oral health belief.
The two knowledge scales were correlated with one another (r = 0.33, p<.001). KCOH was associated with OHF, such that mothers who endorsed a fatalistic belief had significantly lower scores on this scale, reflecting less knowledge of children’s oral hygiene needs (4.45 ± 0.04 vs. 4.74 ± 0.04, p<.0001). KBU was also related to OHF; fatalistic mothers had lower scores than non-fatalistic mothers (3.92 ± 0.06 vs. 4.28 ± 0.08, p<.01). OHSE was not significantly correlated with either knowledge scale or OHF.
Mothers’ self-efficacy was positively correlated with children’s brushing (r = .18, p<.0001). Mothers who were more confident in their ability to make sure their child’s teeth were brushed at bedtime were more likely to have children who brushed more frequently. Both knowledge scales were also significantly positively correlated with brushing frequency, but the magnitude of correlation was smaller. Children of fatalistic mothers brushed less often on average than children of non-fatalistic mothers (8.52 ± 0.21 vs. 10.59 ± 0.62, p<.01).
OHSE and KCOH were each significantly positively associated with better oral health status perceptions in the bivariate analyses (Table 2). The relationship between KBU and the child’s oral health rating followed this positive trend, and approached standard statistical significance levels. When this relationship was examined separately for children aged 1–3, who are more likely than older children to be drinking from bottles regularly, the relationship became stronger (p<.01). Fatalism was not associated with oral health status perceptions.
Although not associated with mothers’ perceptions of children’s oral health status, OHF was significantly associated with disease severity (χ2 = 10.69, p<.05). Table 2 also shows the relationship between the other scales and disease severity. KCOH was significantly associated with the child’s caries status; mothers of children with S-ECC had the lowest scores, reflecting less knowledge of this topic compared to mothers of children with ECC or who were caries-free. The ECC group had a slightly higher average knowledge score than the caries-free group, but the confidence intervals overlapped. Although KBU was not significantly associated with disease severity among all children, when 1–3 year olds were examined separately, the association became significant (p<.05).
This study supports the reliability and validity of four newly developed measures designed to quickly assess dental-specific self-efficacy, fatalism, and knowledge in mothers of young African American children. The KBU, KCOH, OHF and OHSE scales demonstrated feasibility and acceptability given the low rate of missing data in this large study of lower income African American families. This suggests that they are can be readily administered and understood. The associations between these measures and children’s dental practices and outcomes in this high risk population are also noteworthy and indicative of the sensitivity of these instruments. While the majority of scores on the scales were positively skewed, the measures were able to detect significant variation.
Young children develop rapidly, and it is not surprising that KBU was more sensitive for mothers of infants and toddlers, but not preschoolers. This age group also has unique dental hygiene needs. KCOH was positively associated with all the dental practices and outcomes, reaffirming the utility of assessing knowledge in this area. It is unrealistic to expect mothers to conscientiously practice oral health promoting behaviors without understanding that baby teeth are important, require care and cleaning, and can develop caries. Although the majority of mothers endorsed a fatalistic belief, given the burden of disease among children in the community, this perception could be the observed reality and hence relatively stable. This could undermine oral health promoting behaviors if not specifically addressed, as some research has found that lower income populations often endorse fatalistic health beliefs, which may lower the effort they put forth in their self-care routines.34
Although the limitations of cross-sectional data must be kept in mind, if confirmed by further research, the brief measures presented here hold promise for the rapid and accurate assessment of maternal cognitions that influence oral health promoting behaviors. Such cognitions are potentially modifiable, and can be enhanced through tailored educational and cognitive-behavioral interventions to lower caries risk.
Future research should also examine whether the associations reported here persist among a broader socioeconomic sample of African American families, as well as among other racial/ethnic groups. Such replication can advance understanding of the broader determinants of children’s oral health, as well as inform interventions to promote it.
The authors gratefully acknowledge the support of the National Institute for Dental and Craniofacial Research, National Institutes of Health (NIDCR grant # U-54 DE 14261-01), a Minority Research Supplement under the NIDCR parent grant to the first author, the Delta Dental Fund of Michigan, the University of Michigan’s Office of the Vice President for Research, and a pre-doctoral traineeship from the National Institute of Mental Health, National Institutes of Health to the first author (NIMH grant # 5 T32 MH16806).