Objective:

To test the hypotheses that 1) there is no difference between orthodontic patients' and their parents' reports of patients' oral health-related quality of life, and 2) there are no gender differences.

Materials and Methods:

The sample consisted of 182 orthodontic patients (age range, 8–15) and their parents. Respondents were required to complete the Child Oral Health Impact Profile (COHIP). Items were divided into five different subscales, and scores on all subscales were compared between and within groups. Also, scores on six additional items regarding treatment expectations and global health perception were compared. Two hypotheses were tested: first, that no differences between parents and patients would be detected, and second, that no differences between boys and girls would be found.

Results:

The first hypothesis could not be rejected. Only a few minor differences between parents and patients were found. The second hypothesis was rejected. Differences between boys and girls were found on the subscales Emotional Well-Being and Peer Interaction, indicating that girls experience more effects of oral health on their quality of life than do boys.

Conclusions:

Parents' reports on their children's oral health-related qualities of life were in agreement with reports of the orthodontic patients. This suggests that parents are suitable alternatives to their children in surveys measuring oral health-related quality of life.

Poor oral health can have a strong adverse effect on quality of life, as people may experience pain, have trouble eating, or may be ashamed of their teeth.1 Previous studies concerning oral health-related quality of life emphasize the importance of measuring this concept in a valid and reliable way. In the last two decades, various questionnaires have been developed to measure physical, psychological, and social outcomes of oral diseases and disorders.2 

Considering the large number of children and adolescents that receive orthodontic treatment for a wide variety of oro- and craniofacial conditions, measuring oral health-related quality of life in children is an important field of research. Only recently, researchers have started to develop measures suitable for children.3 Measuring self-reported oral health-related quality of life in children and adolescents is complicated, not only because of the different developmental stages children go through, but also because children's oro- and craniofacial features grow and change constantly.3,4 Therefore, different questionnaires have been developed for specific age categories, one of which is the Child Oral Health Impact Profile (COHIP).

The COHIP was designed as part of a large international research project, which was supported by the National Institutes of Health (project 1 R21 DE13721-01). The aim of this project was to develop an instrument to measure oral health-related quality of life in children between 8 and 15 years of age that could be used cross-culturally.5 Items were selected from an existing item pool and divided over different subscales on theoretical grounds. Two versions of the COHIP were developed, one for children and one for parents.

In the Netherlands, a Dutch version of the COHIP was tested on a sample of 35 children with craniofacial anomalies and their parents.6 Subjects were being treated for different conditions, such as craniosynostoses, Treacher-Collins syndrome, Hemifacial microsomia (left- and rightsided), and Goldenhar syndrome. It was found that the COHIP performed adequately, but also that some subscales needed further examination and improvement. Interestingly, no significant differences between patients and parents were found. This raises the question whether parents are able to report adequately their children's oral health-related quality of life. One shortcoming of this study was that the sample was small and heterogeneous. In a follow-up study, a sample of 510 Dutch schoolchildren, without their parents, was used.7 The results of this study showed that the reliability of the Dutch version of the COHIP improved by excluding six items from the original item pool. Although it can be expected that children receiving orthodontic treatment have different experiences with their oral health-related quality of life in comparison with untreated children, no specific account was taken of children in orthodontic treatment in this study.

So far, it is not known how Dutch orthodontic patients perceive their oral health and their oral health-related quality of life. Also, no information is available about perceptions of parents of Dutch orthodontic patients on the oral health-related quality of life of their children. Do parents perceive this quality of their children just like the children themselves? In a recent study of Wilson-Genderson et al,8 only low-to-modest rates of agreement between child and caregiver were found. However, in Geels's6 study, there were no differences between the child's and his or her parent's ratings.

As the dominance of short-term memory and the absence of a fully developed, long-term perspective—as well as possible language problems and a relatively low reading ability—may threaten the reliability and validity of children's responses, it could be worthwhile to use parents' instead of children's reports. Finally, in a study of Broder and Wilson-Genderson,9 no gender differences in the COHIP subscales were found. However, other researchers10 have suggested that women perceive oral health as having a greater impact on their quality of life than do men. Therefore, in our Dutch patient sample, gender differences will be reexamined.

The aim of this study was, to investigate the oral health-related quality of life of orthodontic patients. For this purpose, the shortened version of the COHIP was used. The reliability and validity of the questionnaire was analyzed, and two hypotheses were tested. First, it was hypothesized that no differences between patients and parents would be found. Second, it was hypothesized that the oral health of boys and girls would have similar effects on their quality of life; in other words, no gender differences were expected. Finally, differences between patients and parents, and boys and girls, were explored with regard to their treatment expectations and their global health perceptions.

Participants

The participants in this study were all patients who had started orthodontic treatment between March and September 2007 at the Academic Centre of Dentistry Amsterdam in the Netherlands. Given the age range the COHIP was designed for, patients younger than 8 and older than 15 were excluded. Other reasons for exclusion were repeatedly not showing up for appointments and not paying for treatment. This resulted in a sample of 182 patients, with a mean age of 12 years (SD, 1.46), of which 50% were girls. The parents of the subjects also participated (71.8% females).

Data Collection

The COHIP consists of two parallel questionnaires, one for children and one for parents. In September 2007, all participants received an envelope by mail, containing the child's and the parent's versions of the COHIP, along with an accompanying letter and return envelope. Three weeks later, a reminder was sent. Of the 220 questionnaires mailed, 182 were returned (response rate, 83%).

Dutch Version of the COHIP

The Dutch version of the COHIP in its present form comprises 30 items, divided into five conceptually different subscales (Oral Symptoms, Functional Well-Being, Emotional Well-Being, School Interaction, and Peer Interaction). The subscale, Oral Symptoms, contains 10 items concerning specific oral symptoms. Functional Well-Being is composed of six items related to the child's ability to perform everyday tasks or activities. Emotional Well-Being includes eight items related to emotional states. School pertains to four items related to the school environment, and Peer Interaction incorporates two items about interaction with peers.6,7 Items were answered on a five-point scale (1  =  Never to 5  =  Constantly), with the additional response option, 0  =  I don't know (abbreviated as “DK-response”). The DK-response was added in the Dutch version of the COHIP, as it may provide additional information, for instance, about the clarity of items.6,7 Items were formulated negatively, with two exceptions.

In line with earlier studies in which the COHIP was used,69 six items regarding treatment expectations and one global health perception item were added to the questionnaire (Table 6). These items were not included in the COHIP scores but were used for clinical reasons only. They were answered on a five-point scale (1  =  Totally disagree to 5  =  Totally agree) with the additional response option, 0  =  I don't know (DK-response).

Data Analysis

First, the scores on the two positively worded items were reversed. Then, to make our results easier to interpret and more comparable with results from other studies, all scores were recorded, so that a high score on the subscales of the COHIP implies a good oral health-related quality of life. The internal consistency of the entire questionnaire, as well as each of the subscales, was examined using Cronbach's alpha. Subscale scores were determined by summing up the responses of all items of each subscale. The overall OHRQoL score was calculated by summing up the subscale scores. Concordance between patients and parents was determined by comparing their overall and subscale scores, using paired-sample t-tests. Pearson correlations and ICCs between subscales of children and parents were computed. Differences between boys and girls were examined using independent sample t-tests, which were also employed in the exploratory analyses. Finally, stepwise regression analyses were done on the patient-and-parent sample, with the five subscales as predictors and the score on the global health perception items as the dependent variable.

Reliability Analysis

In Table 1, the internal consistency of the subscales and the overall COHIP are given. One item on the subscale, School, had a negative correlation with the other items on the subscale (Missed school for any reason because of your teeth, mouth or face), thereby decreasing Cronbach's alpha. Therefore, this item was deleted in the analysis. On the subscale, Functional Well-Being, one item was excluded for the same reason (Had difficulty keeping your teeth clean because of your teeth, mouth or face). Finally, on the subscale, Emotional Well-Being, the two positively worded items were excluded, thereby increasing Cronbach's alpha on this subscale from .56 to .76 in the patient sample, and from .67 to .88 in the parent sample (Been confident because of your teeth, mouth or face) and (Felt that you were attractive [good looking] because of your teeth, mouth or face).

Table 1

Internal Consistency of the COHIPa and Its Subscales in the Patient and Parent Sample

Internal Consistency of the COHIPa and Its Subscales in the Patient and Parent Sample
Internal Consistency of the COHIPa and Its Subscales in the Patient and Parent Sample

Internal consistency turned out to be sufficient to good, except for the subscale, Peer Interaction. Alpha coefficients in the parent sample were somewhat higher than in the patient sample, except for the subscale, School.

Differences Between Patients and Parents and Boys and Girls

First, the hypothesis that there were no differences in COHIP scores between patients and parents was tested. In Table 2, the mean subscale and overall scores for both patients and parents are presented. A paired-sample t-test showed only one significant though small difference between parents and patients on the subscale, Oral Symptoms (t  =  3.78; P < .001). In Tables 3 and 4, correlations between subscales and the overall COHIP scores are presented, for both the patient and the parent sample. As expected, many significant correlations were found. Pearson correlations and ICCs between subscales of children and parents are presented in Table 5. All correlations were significant (P < .01). The ICCs suggest high agreement between child and parent reports.

Table 2

Mean Subscale and Overall Scale Scores for Both Patients and Parents

Mean Subscale and Overall Scale Scores for Both Patients and Parents
Mean Subscale and Overall Scale Scores for Both Patients and Parents
Table 3

Correlations Between Overall and Subscale Mean Scores in Patient Sample

Correlations Between Overall and Subscale Mean Scores in Patient Sample
Correlations Between Overall and Subscale Mean Scores in Patient Sample
Table 4

Correlations Between Overall and Subscale Mean Scores in Parent Sample

Correlations Between Overall and Subscale Mean Scores in Parent Sample
Correlations Between Overall and Subscale Mean Scores in Parent Sample
Table 5

Correlations and ICCsa Between Subscales and Overall Scores of Patients and Parents

Correlations and ICCsa Between Subscales and Overall Scores of Patients and Parents
Correlations and ICCsa Between Subscales and Overall Scores of Patients and Parents

Our second hypothesis concerned gender effects on the patient sample, as we expected no differences between boys and girls. In Table 6, mean subscale and overall scores for both girls and boys are shown. Surprisingly, independent sample t-tests revealed two significant differences between boys and girls: girls scored lower on Emotional Well-Being (t  =  −2.03; P < .05) and on Peer Interaction (t  =  −2.23; P < .05).

Table 6

Mean Subscale and Overall Scale Scores for Girls and Boys

Mean Subscale and Overall Scale Scores for Girls and Boys
Mean Subscale and Overall Scale Scores for Girls and Boys

Exploratory Analysis

In Table 7, results of the six items on treatment expectations and global health perception for both patients and parents are presented. Results of boys and girls are shown in Table 8.

Table 7

Results of Treatment Expectations and Global Health Perception Items of Patients and Parents

Results of Treatment Expectations and Global Health Perception Items of Patients and Parents
Results of Treatment Expectations and Global Health Perception Items of Patients and Parents
Table 8

Results of Treatment Expectations and Global Health Perception Items of Boys and Girls

Results of Treatment Expectations and Global Health Perception Items of Boys and Girls
Results of Treatment Expectations and Global Health Perception Items of Boys and Girls

Parents had somewhat higher scores concerning the appearance of patients' teeth and the general health of patients. Boys were more satisfied than girls with the appearance of their teeth, and they believed they would have good health in the future more often than girls. A stepwise regression analysis showed that only the subscale, School, was a significant predictor for the patient's perception of general health, accounting for 5% variance. In the parent sample, the subscale, Oral Symptoms, was the only significant predictor, accounting for 5% of the variance on the global health perception item.

The hypothesis that there were no differences in COHIP scores between patients and parents could not be rejected. In our study, like in others, good agreement between parents and children at the group level was found.11 This suggests that parents are suitable alternatives to their children in surveys measuring oral health-related quality of life. The COHIP and its subscales had sufficient-to-good internal consistency. However, in accordance with the results of Geels et al,6 the parents' questionnaire showed somewhat higher Cronbach's alphas than the children's questionnaire. This suggests that parents may be more reliable in their responses than children.

In our study, only one significant but small difference was found between patients and parents on the subscale, Oral Symptoms. This difference was also found by Wilson-Genderson et al,8 who showed that orthodontic patients reported lower oral health compared with their caregivers. With regard to the exploratory items pertaining to treatment expectations and global health perception, parents were somewhat more positive about the appearance of the patients' teeth and about the general health of the patient. Although it has been stated that parents underestimate the impact of oral conditions on their children's emotional and social quality of life,12 we have shown that there are only minor differences in concordance between patients and parents on ratings of the child's oral health-related quality of life.

Our second hypothesis that there were no differences between boys and girls was rejected. Girls reported more problems with Emotional Well-Being and Peer Interaction, suggesting that they experience more effects of their oral health on their quality of life than do boys. This finding is in agreement with the findings of McGrath and Bedi,10 who found that women perceive oral health as having a greater effect on their quality of life, in a greater number of ways, than men. Women in their study reported to be more embarrassed and self-conscious, and to experience more pain and unhappiness because of their oral health than men. However, in that same study, women also reported more than men that their oral health contributed positively to their appearance, their general well-being, and general health. Since almost all items of the COHIP were negatively formulated, it remains unknown whether there are any differences between girls and boys in their perceptions of oral health as enhancing their quality of life.

In this study, we have demonstrated that children aged 8–15 years are able to give reliable and valid reports about their oral health-related quality of life. In line with Jokovic et al,11,13 we believe that data should be obtained directly from children whenever possible, but if not, parents may be a suitable proxy. For a future study, we recommend an elaboration of the COHIP. That is, new items need to be included in the Dutch version. More specifically, items related to school and peer interaction need to be incorporated to make subscale comparisons more reliable. Also, for further investigation of gender differences, the inclusion of more positively stated items may be worthwhile.

  • Parents' reports on their childrens' oral health-related qualities of life were in agreement with reports of orthodontic patients.

  • Girls experience more effects of oral health on their quality of life than do boys.

The authors thank Lot Geels for her assistance in collecting the data.

1
Petersen
,
P. E.
The World Oral Health Report 2003. Continuous Improvement of Oral Health in the 21st Century—The Approach of the WHO Global Oral Health Programme
Geneva, Switzerland: World Health Organization
.
2003
.
1
38
.
.
2
Slade
,
G. D.
and
S. T.
Reisine
.
The child oral health impact profile: current status and future directions.
Community Dent Oral Epidemiol
2007
.
35
:
50
53
.
3
Jokovic
,
A.
,
D.
Locker
,
M.
Stephens
,
D.
Kenny
,
B.
Tompson
, and
G.
Guyatt
.
Validity and reliability of a questionnaire for measuring child oral-health-related quality of life.
J Dent Res
2002
.
81
:
459
463
.
4
McGrath
,
C.
,
H.
Broder
, and
M.
Wilson-Genderson
.
Assessing the impact of oral health on the life quality of children: implications for research and practice.
Community Dent Oral Epidemiol
2004
.
32
:
81
85
.
5
Broder
,
H. L.
Children's oral health-related quality of life.
Community Dent Oral Epidemiol
2007
.
35
:
5
7
.
6
Geels
,
L. M.
,
J. M.
Kieffer
,
J.
Hoogstraten
, and
B.
Prahl-Andersen
.
Oral health-related quality of life of children with craniofacial conditions.
Cleft Palate Craniofac J
2008
.
45
:
461
467
.
7
Geels
,
L. M.
,
J.
Hoogstraten
, and
B.
Prahl-Andersen
.
Confirmative factor analysis of the dimensions of the Child Oral Health Impact Profile (Dutch version).
Eur J Oral Sci
2008
.
116
:
148
152
.
8
Wilson-Genderson
,
M.
,
H. L.
Broder
, and
C.
Philips
.
Concordance between caregiver and child reports of child's oral health-related quality of life.
Community Dent Oral Epidemiol
2007
.
35
:
32
40
.
9
Broder
,
H. L.
and
M.
Wilson-Genderson
.
Reliability and convergent and discriminant validity of the Child Oral Health Impact Profile (COHIP Child's Version).
Community Dent Oral Epidemiol
2007
.
35
:
20
31
.
10
McGrath
,
C.
and
R.
Bedi
.
Gender variations in the social impact of oral health.
JIDA
2000
.
46
:
87
91
.
11
Jokovic
,
A.
,
D.
Locker
,
M.
Stephens
, and
G.
Guyatt
.
Agreement between mothers and children aged 11–14 years in rating child oral health-related quality of life.
Comm Dent Oral Epidemiol
2003
.
31
:
335
343
.
12
Kiyak
,
H. A.
Does orthodontic treatment affect patients' quality of life?
J Dent Educ
2008
.
72
:
886
894
.
13
Jokovic
,
A.
,
D.
Locker
, and
G.
Guyatt
.
How well do parents know their children? Implications for proxy reporting of child health-related quality of life.
Qual Life Res
2004
.
13
:
1297
1307
.