Introduction
The aim of this study was to compare the assessment of oral health-related quality of life (OH-QoL) between children with malocclusion and their mothers, by using responses to the child perceptions questionnaire and the parental-caregivers perceptions questionnaire.
Methods
The study was conducted in 90 children, aged 11 to 14 years, with a malocclusion grade of 4 or 5 according to the index of orthodontic treatement need dental health component. The children and their mothers completed the questionnaires independently.
Results
The mean ratings were similar for total scores (children, 20.4; mothers, 20.1), oral symptoms (children, 5.2; mothers, 4.7), and social well-being (children, 4.3; mothers, 4.8). However, the mothers group had a lower mean score for functional limitations (children, 5.3; mothers, 3.6) and a higher mean score for emotional well-being (children, 5.6; mothers, 7.1). The correlations between children’s and mothers’ responses ranged from rs = 0.545 for total score and emotional well-being to rs = 0.357 for functional limitations. There were good correlations between their responses to global (rs = 0.466) and life overall (rs = 0.427) questions, but poor correlations between the 2 questions, suggesting that these concepts were considered differently.
Conclusions
Maternal opinions were similar to those of their children for the overall impact on OH-QoL of malocclusion, but mothers were more dissatisfied with the appearance of their children’s teeth and overestimated the emotional impact of malocclusion. It would be useful to develop a specific measure to assess OH-QoL in children with malocclusion.
Quality-of-life measures are increasingly being developed and used in dentistry as the importance of gaining the perspectives of patients and the public is acknowledged. These measures are patient-centered because they capture how oral conditions impact people’s lives, rather than a narrow focus on disease and mouth-centered approaches to the assessment of oral health. Potential uses include political applications to influence policy makers, the development of theory by exploring models of health, and practical purposes to evaluate the effectiveness of interventions or to aid discussion of problems with patients. Recently, oral health-related quality of life (OH-QoL) has been defined as “the impact of oral disorders on aspects of everyday life that are important to patients and persons, with those impacts being of sufficient magnitude, whether in terms of severity, frequency or duration, to affect an individual’s perception of the life overall.”
Several measures have been designed to assess OH-QoL, although the relationship between their outcomes and OH-QoL has recently been questioned. Many of the measures were developed for adults and might not address issues relevant to children.
A further problem with the assessment of quality of life is that it is a dynamic rather than a static phenomenon. People alter the standards by which they rate their OH-QoL over time, because of changes in circumstances or their physical and emotional development. To overcome this limitation, it was suggested that a range of measures should be used to evaluate OH-QoL, one of which could include information from parents or caregivers when children’s OH-QoL is investigated.
Jokovic et al developed the child oral health quality of life questionnaire, which includes age-specific measures for children 6 to 14 years of age (child perceptions questionnaire [CPQ]) and the parental-caregiver perceptions questionnaire (P-CPQ). The CPQ for ages 11 to 14 years and the P-CPQ are comparable questionnaires with 31 items in common organized into 4 domains: oral symptoms, functional limitations, emotional well-being, and social well-being. In a study of 42 mothers and children, Jokovic et al found generally good agreement in the groups. However, there were significant discrepancies between pairs, particularly in the emotional and social well-being domains. This suggests that it is not appropriate to use mothers as proxies for their children at the individual level.
Zhang et al examined agreement between P-CPQ and CPQ in mothers, fathers, and children with orthodontic treatment need in Hong Kong. They found that the parents rated the OH-QoL poorer than did the children in all domains. They showed that, although there was generally good agreement between mothers and fathers at the group level, agreement among mothers, fathers, and their children at the individual level was poor. They concluded that not only did the children and parents have differing views about the impact of malocclusion, but also the 2 parents sometimes disagreed. This underlined the importance of consulting the whole family when discussing orthodontic need and treatment.
The aim of this study was to examine the relationship between reports of OH-QoL from children with malocclusion and their mothers in a sample in the United Kingdom. More specifically, we examined in which of the 4 domains of the child oral health quality of life questionnaire the agreement or disagreement occurred and to what extent it was apparent.
Material and methods
Ethical approval was granted by the South Sheffield Research Ethics Committee (reference number 03/262), and site-specific issues were reviewed by the Research and Development Department at Chesterfield Royal Hospital.
The sample consisted of consecutive orthodontic patients between the ages of 11 and 14 years who were removed from a treatment waiting list, had pretreatment records taken, and were considered ready to start treatment at the Charles Clifford Dental Hospital in Sheffield or the Chesterfield and North Derbyshire Royal Hospital in Chesterfield. The subjects were recruited to a study examining the effect of malocclusion on OH-QoL. They were assessed to be in grades 4 or 5 of the index of orthodontic treatment dental health component by a trained and calibrated examiner C.O.’B., and all agreed to take part before treatment. Patients with active dental disease, cleft lip or palate, complicating medical history, or severe dental mottling were excluded. A sample size calculation suggested that 90 patients should be recruited to detect a difference of 30% in the total CPQ (α = 0.05; β = 0.90) between subjects with and without malocclusion.
The child and the parent who was present at the consultation were asked to independently complete the CPQ for children aged 11-14 years and the P-CPQ, respectively. The child and the parent completed the questionnaires separately in a quiet area of the orthodontic clinic with a researcher available to answer questions. Both the CPQ and P-CPQ have been evaluated for use in the United Kingdom and were found to have acceptable psychometric properties. These questionnaires also include 2 global questions: (1) rating the health of their, or their child’s, teeth, lips, jaws, and mouth; and (2) how much their, or their child’s, teeth, lips, jaws, or mouth affect life overall. These global questions were previously evaluated. They were worded, “would you say that the health of your teeth, lips, jaws, and mouth is __?” with a 5-point response format from excellent to poor; and “how much does the condition of your teeth, lips, jaws, or mouth affect your life overall?” with a response range from not at all to very much. A third question asked about satisfaction with their, or their child’s, teeth on a 5-point scale from very satisfied to very dissatisfied.
Statistical analysis
The response option codes used for both the CPQ and P-CPQ were 0, never; 1, once or twice; 2, sometimes; 3, often; and 4, every day or almost every day. The P-CPQ questions also had a “don’t know” response option that was given a score of 0 in the analysis. The “don’t know” response was included in the parents’ questionnaire to acknowledge the limitations of the parent’s knowledge of the child’s oral health or everyday activities. Total and domain scores were obtained by summing the response option codes for each question. The response format for the global, life overall, and satisfaction with teeth questions was also a 5-point Likert scale.
The relationship between the CPQ and the P-CPQ was assessed in a number of ways. Comparisons were made by examining the mean scores and the differences between the scores. Mean directional differences were tested by using a paired t test. The magnitude of any systematic differences was examined by dividing the mean by the standard deviation to obtain a standardized difference. A standardized difference of 0.2 was considered small, 0.5 was moderate, and 0.8 was great. An absolute mean difference was calculated by ignoring the positive and negative sign of the individual differences. This was then expressed as a percentage of the maximum score to assess the size of the absolute differences.
Discrepancies between parent and child were also assessed by subtracting the child’s score from the parent’s score and using the standard deviation of the differences between the 2 scores to rate concordance. The scores were rated similar when the difference between the scores was within .5 SD above or below a difference of zero. If scores had a difference greater than .5 SD below an overall difference of zero, the child was rated as giving a higher score than the parent. If the scores had a difference greater than .5 SD above an overall difference of zero, the parent was rated as giving a higher score than the child.
The internal consistency of the measure used in this sample was assessed with Cronbach’s alpha. The association between the parents’ and children’s responses to the global question, the life overall, the satisfaction with teeth, and the CPQ and P-CPQ total and domain scores were examined by using the Spearman correlation. Agreement was assessed with intraclass correlation coefficients.
Results
A total of 116 pairs of children and parents or caregivers completed both questionnaires. The parents and caregivers included 90 mothers, 20 fathers, and 6 others. Because there were few fathers and others, only the responses between the children and the mothers were compared. The sex and age of these children are shown in Table I .
| n | % | ||
|---|---|---|---|
| Subjects | Boys | 34 | 37.8 |
| Girls | 56 | 62.2 | |
| Age (y) | 11 | 14 | 15.6 |
| 12 | 23 | 25.6 | |
| 13 | 22 | 24.4 | |
| 14 | 31 | 34.4 | |
| Mean age (SD) | 13.3 (1.1) |
The number of “don’t know” responses per mother ranged from 0 to 12. Almost half (47.8%) of the mothers responded “don’t know” at least once, 28.9% had 3 or more “don’t know” responses, and 15.6% had 6 or more.
The Cronbach’s alpha values were 0.91 for the total CPQ score and 0.90 for the total P-CPQ score. The Cronbach’s alpha values for the respective CPQ and P-CPQ domain scores were 0.70 and 0.58 for oral symptoms, 0.58 and 0.67 for functional limitations, 0.89 and 0.90 emotional well-being, and 0.74 and 0.77 for social well-being. These figures are slightly lower than those obtained by Jokovic et al but still represent good internal consistency.
Comparisons between the mean total and 4 domain scores from the CPQ and P-CPQ responses are shown in Table II . The mean ratings were similar for the total score (children, 20.4; mothers, 20.1), oral symptoms (children, 5.2; mothers, 4.7), and social well-being (children, 4.3; mothers, 4.8) domains. Differences in the mean scores suggest that mothers underestimated functional impacts (children, 5.3; mothers, 3.6) and overestimated emotional impacts (children, 5.6; mothers, 7.1) compared with the children’s scores.
| Children | Mothers | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Items (n) | Mean | SD | Minimum | Maximum | Mean | SD | Minimum | Maximum | |
| Total score (0-124) | 31 | 20.4 | 12.7 | 0 | 50 | 20.1 | 14.8 | 1 | 73 |
| Domains | |||||||||
| Oral symptoms (0-24) | 6 | 5.2 | 3.4 | 0 | 13 | 4.7 | 3.2 | 0 | 13 |
| Functional limitations (0-28) | 7 | 5.3 | 3.6 | 0 | 19 | 3.6 | 3.9 | 0 | 18 |
| Emotional well-being (0-32) | 8 | 5.6 | 5.4 | 0 | 21 | 7.1 | 6.6 | 0 | 25 |
| Social well-being (0-40) | 10 | 4.3 | 4.1 | 0 | 20 | 4.8 | 4.8 | 0 | 20 |
The difference between the answers of the children and their mothers in the functional impacts domain was mainly due to 2 questions. The biggest discrepancy in responses was to the question “in the past 3 months, because of your or your child’s teeth, mouth, lips, and jaws, how often have you, or has your child, breathed through the mouth?’ The proportion of children responding “never” to this question was 26%; the proportion of mothers responding “never” was 43%, and 20% responded “don’t know.” The second question causing the discrepancy in the functional domain was: “in the past 3 months, because of your or your child’s teeth, mouth, lips, and jaws how often have you, or has your child, taken longer than others to eat a meal?” The proportion of children responding “never” to this question was 50%; the proportion of mothers responding “never” was 75%, with no mothers responding “don’t know.”
The difference in the answers between children and mothers in the emotional well-being domain was also mainly due to the responses to 2 questions. The question, “in the past 3 months, because of your or your child’s teeth, mouth, lips, and jaws, how often have you, or has your your child, been upset?’ produced a “never” response in 64% of children compared with 44% of mothers, with just 3 mothers choosing “don’t know.” The question, “in the past 3 months, because of your or your child’s teeth, mouth, lips, and jaws, how often have you, or has your child, been nervous, anxious, or fearful?’ produced a “never” response in 72% of children compared with 50% of mothers, with 7 mothers choosing “don’t know.”
Table III shows the mean directional differences and confirms that the responses for the functional limitations ( P <0.001) and emotional domains ( P = 0.026) were significantly different. The standardized differences are shown in Table IV . They were generally small, ranging from –0.02 for the total score to –0.46 for the functional limitations domain, which approaches moderate disagreement. The absolute differences for the total and domain scores are also shown in Table IV . They ranged from 0 to 31, with 59% of the child-mother pairs showing a difference of 10 or less. The mean absolute differences ranged from 7% of the maximum score for social well-being to 15% of the maximum score for emotional well-being.
| 95% CI of difference | |||||
|---|---|---|---|---|---|
| CPQ–P-CPQ | Mean difference | SD | Lower | Upper | P |
| Total score | 0.3 | 12.4 | −2.3 | 2.9 | 0.826 |
| Domains | |||||
| Oral symptoms | 0.5 | 3.6 | −0.2 | 1.2 | 0.185 |
| Functional limitations | 1.7 | 3.7 | 0.9 | 2.5 | <0.001 |
| Emotional well-being | −1.4 | 6.0 | −2.7 | −0.2 | 0.026 |
| Social well-being | −0.5 | 3.9 | −1.3 | 0.3 | 0.224 |
| Absolute differences | ||||
|---|---|---|---|---|
| Standardized differences (mean [SD]) | Mean | SD | % of maximum score | |
| Total score | −0.02 | 9.6 | 7.8 | 8 |
| Domains | ||||
| Oral symptoms | −0.14 | 2.8 | 2.2 | 12 |
| Functional limitations | −0.46 | 3.2 | 2.5 | 12 |
| Emotional well-being | 0.24 | 4.7 | 4.0 | 15 |
| Social well-being | 0.13 | 2.7 | 2.8 | 7 |
Nearly half of the mother and child total scores (46%) were within .5 SD above and below a difference of zero, although this was true for only 25% of the functional limitations domain, with 52% of the children scoring higher than their mothers. Conversely, 41% of the mothers scored higher than their children in the emotional well-being domain. The concordance between mother and child was similar for both oral symptoms and social well-being.
Table V shows the mean scores from the global, life overall, and satisfaction with teeth questions, and Table VI shows the correlations between the mother and child responses to the global question, the life overall, the satisfaction with teeth, and the CPQ and P-CPQ total and domain scores. The Spearman correlations (rs) ranged from 0.545 for total CPQ and emotional well-being to 0.097 for satisfaction with teeth. The correlations were all significant, except for satisfaction with teeth. The intraclass correlation coefficients ranged from substantial for social well-being (0.62) to moderate for oral symptoms 0.42. The association between the global and life overall responses were weak for both children (rs = 0.390) and mothers (rs = 0.265).
