Introduction
The purpose of this study was to assess the agreement between Brazilian adolescents and their parents or caregivers regarding the adolescents’ oral health–related quality of life during orthodontic treatment.
Methods
The sample consisted of 102 adolescent-parent/caregiver pairs. Adolescents answered the short form of the child perceptions questionnaire, and the parents or caregivers answered the parental-caregiver perceptions questionnaire. These questionnaires have 14 items in common organized across 4 subscales: oral symptoms, functional limitations, emotional well-being, and social well-being. Agreement on the overall score and the subscales was determined using comparison and correlation analysis. The former was performed through comparison of the mean directional and absolute differences. The latter was analyzed using the intraclass correlation coefficient.
Results
The mean directional difference was significant for the oral symptoms subscale, with parents’ reports lower than adolescents’ reports ( P = 0.012). However, it was not significant for the other 3 subscales and the overall score ( P >0.05). The mean absolute difference for the overall score was 5.15, representing 9.2% of the maximum possible score. The intraclass correlation coefficient was 0.66, indicating substantial agreement.
Conclusions
There was agreement between adolescents and their parents or caregivers in rating adolescents’ oral health–related quality of life during orthodontic treatment.
Highlights
- •
Orthodontic patients and parents showed good agreement in rating quality of life.
- •
This information could be useful for orthodontists in clinical practice.
- •
Quality of life research can be used to guide oral health public policies.
The concept of oral health–related quality of life (OHRQoL) was developed to measure subjective perceptions regarding the impact of oral health and functional status on quality of life. The assessment of OHRQoL has been increasingly recognized as an important health outcome measure in dentistry because oral problems, such as malocclusion, can have a negative impact on the physical and psychological well-being of young people.
Orthodontic treatment can also have an impact on the OHRQoL of children and adolescents. Quality of life is positively affected after the removal of the orthodontic appliances as a result of improvements in emotional and social well-being. However, OHRQoL is often reduced at the onset of treatment because of the worsening of oral symptoms and the impairment of oral functions. Information on the first 3 months of orthodontic therapy has considerable importance to clinicians, since the inconvenience of pain and functional limitations in this period can lead to disappointment on the part of patients and can ultimately result in treatment dropouts. The interval between 6 and 12 months also has relevant significance. There are still deteriorations of oral symptoms and functional limitations. However, improvements of social well-being and overall OHRQoL are also observed.
Research has been conducted to evaluate the agreement between pediatric patients and their parents or caregivers with regard to rating children’s and adolescents’ OHRQoL. The findings of a systematic review indicate that although some studies report good agreement between parents or caregivers and their children or adolescents, other investigations have found a low level of agreement. However, this does not lessen the value of the reports of parents or caregivers in pediatric health outcome research. Parents or caregivers are the main decision makers regarding the oral health of their children or adolescents, and their perceptions exert a major influence on the choices involved in orthodontic treatment. Therefore, even when the children or adolescents are able to self-report, the proxy reports of parents or caregivers regarding the quality of life of their sons and daughters should be considered an additional outcome measure. When the reports of both parents or caregivers and adolescents are used, the former should be interpreted as a complement and not a substitute for the latter. Information provided by parents or caregivers is valuable in guiding orthodontists with regard to clinical decision making. Such information could also be useful for general and pediatric dentists before referring patients for orthodontic treatment.
The impact of ongoing orthodontic treatment on the OHRQoL of adolescents has been well documented. However, the agreement in rating OHRQoL between parents or caregivers and their sons and daughters undergoing orthodontic therapy with a fixed appliance has been underinvestigated, and this question is yet to be answered fully. Studies with different populations on this type of agreement should be highly encouraged. The validity of the reports of parents or caregivers, and therefore whether they can serve as proxies for their sons and daughters, depends on the understanding of this issue in its entirety. Moreover, ethnicity is a marker for oral health outcomes. It is important to take into account that among different populations, there are differences in oral health behaviors and beliefs. Thus, the aim of this study was to assess the agreement between the reports of Brazilian adolescents and of their parents or caregivers regarding the adolescents’ OHRQoL during orthodontic treatment with a fixed appliance. We hypothesized that parents’ or caregivers’ reports on their adolescents’ quality of life during orthodontic treatment would agree with the reports of their adolescents. Our research questions were whether there are no differences in the perceptions of OHRQoL between adolescents and their parents or caregivers in rating the adolescents’ quality of life during orthodontic therapy, in which domains the agreement occurred, and whether adolescents and parents or caregivers are complementary or alternative sources of information.
Material and methods
The sample consisted of 104 adolescents, aged 11 and 12 years, undergoing orthodontic treatment with a fixed appliance at the Department of Pediatric Dentistry and Orthodontics of the Federal University of Minas Gerais (Brazil) and their parents or caregivers. For inclusion in the study, the adolescents and their parents or caregivers needed to be literate and fluent in Brazilian Portuguese. The exclusion criteria were adolescents with craniofacial anomalies, cognitive disorders, and a history of dental trauma.
The sample size was determined using the intraclass correlation coefficient (ICC). The null hypothesis for the ICC was set at 0.4 (moderate agreement). The ICC regarded as indicating significant agreement was set at 0.6 (substantial agreement). Considering an α of 0.05 and a β of 0.2, we required a minimum of 87 pairs of adolescents and parents or caregivers. This figure was increased by 20% to compensate for possible losses. Therefore, the sample comprised 104 pairs of adolescents and parents or caregivers.
This study received approval from the human research ethics committee of the Federal University of Minas Gerais (Brazil). All participants received clarifications regarding the objectives of the study and signed a statement of informed consent. Confidentiality was ensured because only the researchers had access to the raw data. All study personnel signed a confidentiality agreement that outlined their responsibilities concerning the privacy of the participants. Hard copies were locked in a filing cabinet. Digital information was kept in a computer protected by a password.
OHRQoL was measured with the Child Oral Health Quality of Life Questionnaire. Adolescents answered the short form of the Child Perceptions Questionnaire (CPQ11-14), and parents or caregivers answered the Parental-Caregiver Perceptions Questionnaire (P-CPQ). Both instruments are valid and reliable and have been cross-culturally adapted for use in the Brazilian population. Adolescents and parents or caregivers answered the questionnaires separately in a quiet area of the orthodontic clinic, with a researcher (L.G.A.) available to clarify any questions. The assessments were done 8 months after the banding and bonding of a fixed appliance.
The CPQ11-14 is composed of 16 items distributed among 4 subscales: oral symptoms, functional limitations, emotional well-being, and social well-being. Each item has 5 response options: 0, never; 1, once or twice; 2, sometimes; 3, often; and 4, every day or almost every day. The P-CPQ consists of 31 items distributed among the same 4 subscales with the same 5 response options. For the P-CPQ, a “don’t know” response is also allowed. The method used to manage this option was to calculate, for each participant, mean scores based on items with responses other than “don’t know.” Thus, those scores were adjusted for the number of items that contributed to the score. The 2 questionnaires have 14 items in common: 4 items on the oral symptoms subscale, 4 on the functional limitations subscale, 3 on the emotional well-being subscale, and 3 on the social well-being subscale. The items on both questionnaires address the frequency of events in the previous 3 months regarding problems with adolescent’s teeth, lips, jaws, or mouth. The overall score is computed by summing all the item scores. Scores for each of the 4 subscales can also be computed. For the 14 items in common, the overall score on both questionnaires ranges from 0 to 56. For the CPQ11-14, a higher score denotes a greater impact on the adolescent’s OHRQoL. For the P-CPQ, a higher score indicates a greater negative perception on the part of parents or caregivers with regard to the OHRQoL of their adolescent sons and daughters.
Parents or caregivers were also asked to answer a question on household income, which was measured in terms of the Brazilian monthly minimum wage (BMMW), which is a standard for this type of assessment and corresponded to approximately $325 (United States) at the time of the data collection. Household income was dichotomized as families with a household income equal to or lower than 3 BMMWs, and families with a household income higher than 3 BMMWs.
Statistical analysis
Statistical analysis was carried out using the Statistical Package for the Social Sciences software (SPSS for Windows, version 17.0; SPSS, Chicago, Ill). Descriptive statistics were performed. Directional and absolute differences between adolescents’ and parents’ or caregivers’ scores were determined. The directional differences were calculated by subtracting the CPQ11-14 score from the P-CPQ score. The overall and subscale directional differences were then compared with zero using paired t tests to evaluate statistical significance. To assess the magnitude of systematic bias, mean directional differences were divided by their respective standard deviations (SDs). For interpretation of the magnitude of the differences, a standardized difference of 0.2 was considered small, 0.5 was considered moderate, and 0.8 was considered large. Mean absolute differences were calculated by ignoring the positive and the negative signs of the directional differences, which provided an indicator of agreement. This was then expressed as a percentage of the maximum score to assess the size of the absolute differences. Directional and absolute differences regarding the adolescents’ sex and the household income were also determined. ICC values were calculated for the overall and subscale scores, and the level of agreement was categorized as follows: poor (<0.2), fair (0.2-0.4), moderate (0.41-0.60), substantial (0.61-0.80), and excellent (0.81-1.0).
Results
A total of 102 pairs of adolescents and parents or caregivers agreed to answer the questionnaires (response rate, 98.1%). Among the 102 adolescents, 47 were male (46.1%) and 55 were female (53.9%). The mean age of the adolescents was 11.37 ± 0.67 years. Among the 102 families that participated in the study, 74 earned less than 3 times the BMMW. Information about the 14 similar items of the CPQ11-14 and the P-CPQ is presented in Table I .
| CPQ11-14 and P-CPQ subscales | CPQ11-14 items | CPQ11-14 score range | P-CPQ items | P-CPQ score range |
|---|---|---|---|---|
| Oral symptoms |
|
0-16 |
|
0-16 |
| Functional limitations | For the next questions, has this happened because of your teeth, lips, jaws, or mouth?
|
0-16 |
|
0-16 |
| Emotional well-being |
|
0-12 |
|
0-12 |
| Social well-being | Have you had these experiences because of your teeth, lips, jaws, or mouth? If it was for another reason, answer “Never.”
|
0-12 |
|
0-12 |
The directional differences regarding the adolescents’ sex and the household income were not statistically significant ( P >0.05). There was no directional difference between boys and girls and those whose families had a household income equal to or lower than 3 BMMWs compared with those whose families had a household income greater than 3 BMMWs.
Adolescents had higher OHRQoL scores in the oral symptoms and social well-being subscales as well as in the overall scale than did their corresponding parents or caregivers ( Table II ). However, the mean directional difference of 0.79 for the overall score was not significant ( P = 0.248). The mean directional differences for the subscales ranged from 0.08 to 0.64. The mean directional difference was statistically significant for the oral symptoms subscale, with adolescents’ scores higher than parents’ or caregivers’ scores. When the mean directional differences were standardized, the magnitude of the directional difference for the overall OHRQoL score was 0.11. The mean absolute difference between the overall CPQ11-14 and P-CPQ scores was 5.15 ± 4.52, representing 9.2% of the maximum possible score of 56. Among the subscales, this percentage ranged from 12% to 13.8%. The ICC for the overall OHRQoL score was 0.66, demonstrating substantial agreement between adolescents and parents or caregivers in rating the adolescents’ OHRQoL during orthodontic therapy with a fixed appliance. Among the different subscales, the ICC ranged from 0.52 to 0.59, demonstrating moderate agreement ( Table III ). The lower limits of the confidence intervals (CIs) for the ICC values of the oral symptoms, functional limitations, emotional well-being, and social well-being subscales varied from 0.28 to 0.39. Those values were in the “fair agreement” category, which ranges from 0.2 to 0.4.
