This research took place at the Department of Pediatric Dentistry at the Tufts University School of Dental Medicine. The study was approved by the Tufts Health Sciences Institutional Review Board (Study #13222), and all methods were carried out in accordance with relevant guidelines and regulations. Inclusion criteria were English-speaking parent/legal guardian-child dyads presenting to the Tufts Pediatric Dentistry Clinic for the child’s initial dental examination or re-care visit and child’s age between 3–6 years old. Exclusion criteria were medically compromised patients with a physical or intellectual disability that could impact oral health in the opinion of the investigator (ASA III, ASA IV) and patients presenting for a dental emergency such as pain, infection, or trauma.
Informed consent
was obtained from all participants’ legal guardian. One parent of each child was asked to complete a survey using a hard copy format. If a child presented with both parents, the parent who spent more time with the child, based on the parents’ judgment, was asked to fill out the survey. Before any survey item was administered, an information sheet regarding the study was provided by the principal investigator and/or co-investigator to each potential subject. The information sheet met all required elements of consent; the potential subject was given as much time as they desired to read it and consider the study. After the potential subject read the information sheet, any questions were answered. Potential subjects were reminded that their participation was voluntary, that they could choose to stop participating at any time without penalty, and that participation or the refusal to participate would have no effect on the child’s care.
The first part of the survey involved demographic data about the child and the family, including parent’s age, gender, ethnicity, race, marital status, number of children, education level, annual household income, location (urban, rural or suburban), and form of payment, as well as child’s sex, age, birth order, and reason for dental visit. The second part of the survey included questions from the shortened version of the Parenting Style and Dimensions Questionnaire (PSDQ), which has been developed and validated as a reliable instrument to classify parents as authoritative, authoritarian, or permissive.19–21 The original instrument consists of 62 items, and the shortened version consists of 32 items.20–21 Each item relates to one of the three parenting styles; in the shortened version, there are 15 items corresponding to the authoritative style, 12 items corresponding to the authoritarian style, and five items corresponding to the permissive style. Each item asks the respondent how often they (or they and their significant other, if applicable) exhibit a specified parenting behavior, using a Likert scale (1 = “Never”, 2 = “Once in a while,” 3 = “About half the time,” 4 = “Very often,” and 5 = “Always”). The score for each parenting style is defined as the mean of the item-level scores in that parenting style. Once such a mean has been computed for each of the styles, the parent’s style is defined as the style with the highest mean. The shortened version has been found to exhibit adequate Cronbach’s α values.10 Apart from the survey, the child’s decayed, missing, and filled teeth (dmft) index and diet score (which was based on a standardized scale used at our institution and scored with three categories: excellent, good, and needs improvement) from their first dental visit to the Tufts Pediatric Dentistry Clinic had been determined by the treating pediatric dental resident and was obtained from their axiUm record. Data collection occurred from May 2019 to February 2020.
A sample size calculation was conducted using the computer program nQuery Advisor (Version 7.0) (Statistical Solutions Ltd., Cork, Ireland). Based on the findings of Howenstein et al.,7 a sample size of at least n = 20 per parenting style was adequate to obtain power greater than 99% to detect a difference in dmft between parenting styles in conjunction with a Type I error rate of α = 0.05. As the parenting style of a subject was not known until after they had completed the survey, recruitment proceeded until at least 20 subjects of each parenting style were obtained.
Descriptive statistics were calculated including means, standard deviations (SDs), medians, and interquartile ranges (IQRs) for continuous variables and counts and percentages for categorical variables. In bivariate analyses, the difference in dmft between children with parents of different parenting styles was assessed using the Kruskal-Wallis test due to non-normality of the data; post-hoc comparisons were performed using Dunn’s test with the Bonferroni correction. The assumption of normality was assessed using the Shapiro-Wilk test. To adjust for confounding, multivariable negative binomial regression was also used; exponentiated regression coefficients along with 95% confidence intervals (CIs) were calculated. Analogous to the odds ratio (OR) statistic, a value greater than 1 for the exponentiated coefficient indicates a higher dmft for a given group, compared to a reference group, adjusting for the other variables in the regression model; a value less than 1 indicates a lower dmft for the given group, compared to the reference group, adjusting for the other variables in the model. P-values less than 0.05 were considered statistically significant, with the exception of analyses in which the Bonferroni correction was used. SPSS version 25 (IBM Corp., Armonk, NY, USA) was used in the statistical analysis.