The European Journal of Orthodontics Advance Access published online on October 30, 2007
The European Journal of Orthodontics, doi:10.1093/ejo/cjm077
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Prediction of compliance and completion of orthodontic treatment: are quality of life measures important?
* School of Dentistry, University of Manchester
** Orthodontic Practice, Lytham St Anne's, UK
Address for correspondence Dr N. A. Mandall, School of Dentistry, University of Manchester, Higher Cambridge Street, Manchester M15 6FH, UK. E-mail: nicola.a.mandall{at}manchester.ac.uk
 |
Summary |
|---|
 Top Summary IntroductionSampleMethodStatistics analysisResultsDiscussionConclusionsFundingReferences |
|---|
The main aim of the present investigation was to evaluate whether there is an association between completion of orthodontic treatment and quality of life measures, i.e. age, gender, socio-economic status, type of appliance and need for orthodontic treatment. The secondary aim was to evaluate whether compliance with orthodontic treatment (missed appointments and appliance breakages) was associated with age, gender, socio-economic status, or type of appliance. This was a multi-centre longitudinal observational study carried out on 144 patients (65 males and 79 females) aged 10–19 years.
Baseline data were collected: patient age, gender, socio-economic status, Index of Orthodontic Treatment Need (IOTN), and type of appliance. Quality of life information, including orthodontic utility values and oral aesthetic subjective impact score (OASIS), were also collected at the start of treatment. The main outcome measure was whether a patient completed treatment. Compliance was assessed by recording the number of failed appointments and appliance breakages. Multiple regression analysis was used to investigate the association between independent and dependent variables.
None of the baseline variables, including quality of life measures, were associated with a patient completing treatment, or their compliance with treatment (P > 0.05). Thus, quality of life measures (utility values or OASIS) do not add to our knowledge of who may complete, or co-operate with, orthodontic treatment. In addition, neither age, gender, socio-economic status nor clinical treatment need (IOTN) were useful in helping a clinician to choose potentially co-operative patients.
|
Introduction |
|---|
|
TopSummaryIntroductionSampleMethodStatistics analysisResultsDiscussionConclusionsFundingReferences |
|---|
Allocation of resources for orthodontics, in the future, may be influenced by quality of life indicators (Cunningham and Hunt, 2001
). Quality of life measures are becoming increasingly important, since clinician-based measures of treatment need not take into account patient perceptions or opinions. Although access to orthodontic services is limited by clinical need, health service use is also driven by demand for treatment which is influenced by the less tangible consumer-based factors that cause a person to seek care. Quality of life measures aim to capture these multiple consumer-based factors and it may be advantageous to combine such information with clinician-based measures of orthodontic need.
Quality of life measures have been developed in dentistry based on a conceptual framework to help explain the multidimensional nature of the impact of oral health on daily life (Locker, 1988). Locker defined various categories in relation to oral health as handicap, disability, discomfort, functional limitation, impairment, disease, and death (Figure 1). Most of the subsequent quality of life dental measures have been developed as questionnaires tested on adults, some of which were accompanied by a clinical examination (Cushing et al., 1986; Locker and Grushka, 1987; Rosenberg et al., 1988; Gooch et al., 1989; Reisine et al., 1989; Atchison and Dolan, 1990; Strauss and Hunt, 1993; Locker and Miller, 1994a,b; Slade and Spencer, 1994; Adulyanon et al., 1996; Kressin et al., 1996; Strauss, 1996; Tickle et al., 1997; Gilbert et al., 1998; Awad et al., 2000; Nuttall et al., 2001). Importantly, few quality of life measures have been developed in orthodontics or for children, but recent work has suggested that utility values (Fox, 1997) and oral aesthetic subjective impact scores (OASIS; Mandall et al., 1999) may be useful.
|
Utility value as a measure of quality of life
A utility value is a number that, in theory, represents a condensation of biological, physical, sociological, and psychological parameters that influence a person's sense of well-being. They represent the values that individuals hold for certain states of health or disease. O'Brien et al. (1998) showed that seekers of orthodontic treatment had lower utility values than children who did not seek orthodontic care, i.e. those seeking treatment perceived their aesthetic dental appearance to be less acceptable.
OASIS as a measure of quality of life
OASIS consists of a series of questions to assess the degree of concern/disadvantage that children feel because of the arrangement of their teeth (Mandall et al., 1999). Children were asked to indicate, on a seven-point Likert scale, their concern about tooth appearance, nice comments about teeth, unpleasant comments about teeth, teasing about teeth, avoidance of smiling, and covering the mouth. These scores were then added to the child's self-perceived Aesthetic Component of the Index of Orthodontic Treatment Need (IOTN-AC), (Brook and Shaw, 1989; Figure 2) score to obtain the OASIS. A higher score indicated more concern/negative impact from their tooth arrangement. This work suggested that the higher a child's concern about the appearance of their teeth, the more likely they were to want orthodontic treatment.
|
Therefore, it is suggested that if these orthodontic quality of life measures can predict a wish for and uptake of orthodontic services, maybe such patients would place a higher value on their dental appearance and be more co-operative with treatment.
If it were possible to predict co-operative patients, potentially treatment discontinuation levels may reduce. Discontinuation rates have been reported to be between 12 and 20 per cent (Haynes, 1982; Murray, 1989) increasing to around 40 per cent in 15- to 17-year olds (Haynes, 1991). This has obvious implications in terms of wasted resources, where appliances are placed but have little occlusal benefit before treatment is stopped.
A number of predictors of poor co-operation have been identified including some psychological tests such as the Psychological Test, Achievement and Affiliation Motivation Test and Attribution Motivation Test (El-Mangoury, 1981; Egolf et al., 1990), patient attitude at the beginning of treatment (Sergl et al., 1998), and use of removable appliances (Murray, 1989). In contrast, headgear compliance was seen to increase with the use of headgear calendars (Cureton et al., 1993) and general compliance appears to be influenced by a positive orthodontist–patient relationship with good communication about patient concerns and treatment options (Sinha et al., 1996).
However, the role of quality of life measures on completion and compliance with orthodontic treatment has not been investigated and, thus, formed the focus of this study.
The main aim was to evaluate whether there is an association between completion of orthodontic treatment and quality of life measures, age, gender, socio-economic status, type of appliance, and need for orthodontic treatment (IOTN; Brook and Shaw, 1989). The secondary aim was to evaluate whether compliance with orthodontic treatment (missed appointments and appliance breakages) was associated with age, gender, socio-economic status, or type of appliance.
|
Sample |
|---|
|
TopSummaryIntroductionSampleMethodStatistics analysisResultsDiscussionConclusionsFundingReferences |
|---|
Consecutive subjects at the start of active orthodontic treatment were investigated. The population comprised 10- to 19-year-old subjects from the University Dental Hospital of Manchester, Bolton Royal Hospital, St Anne's Orthodontic Practice, and Hope Hospital, Salford, UK. Three orthodontists carried out the treatment (NAM, DF, and JM).
Sample size calculation
It has been shown that 80 per cent of patients who are accepted for orthodontic treatment are in IOTN grades 4 or 5, the remaining 20 per cent in grade 3 or below (Mandall, 1997). As previously outlined, compliance has been reported to be between 12 and 40 per cent. Therefore, it was assumed that 10 per cent of those with high need and 35 per cent with a low need may fail to complete a course of orthodontic treatment. A two-group continuity-corrected chi-square test, with a 0.050 two-sided significance level, will have an 80 per cent power to detect the difference between a proportion of 0.10 in a high treatment need group and 0.350 in a low treatment need group (odds ratio of 4.85) when the sample sizes are 112 and 28, respectively. Thus, a total sample size of 140 patients was required.
|
Method |
|---|
|
TopSummaryIntroductionSampleMethodStatistics analysisResultsDiscussionConclusionsFundingReferences |
|---|
Ethical approval was obtained from Central Manchester, Bolton, Salford, and Trafford and Blackpool, Wyre, and Fylde ethics committees. Written consent was obtained from the patient and a parent or legal guardian.
Baseline age, gender, and socio-economic status
The date of birth and date of start of treatment was recorded and used to calculate age at the start of treatment. Gender and type of appliance was also noted. Socio-economic status was evaluated using Townsend deprivation scores (Townsend, 1987) whereby postcode data obtained from the child was linked to the place in which they lived. Then, using a postcode directory from data sets information of the United Kingdom (U.K.) Census (1991) a deprivation score was calculated where a higher score indicates increased social deprivation.
Index of Orthodontic Treatment Need
Start of treatment study models were scored for the AC and dental health component (DHC) of the IOTN by SM, a calibrated examiner. Intra-examiner reliability was tested by re-scoring 20 models, at least 1 week later.
Quality of life: utility score
Utility values were measured by referring to the IOTN-AC 8, 5, and 3 (Figure 2). Aesthetic scores were measured relative to the best state (AC = 1) and the worst state (AC = 10) by presenting subjects with standard photographs. For example, the subject was asked to choose whether they would prefer (1) the appearance of AC 8 for the next 20 years or (2) the appearance of AC 10 for 1 year and AC 1 for the remaining 19 years.
The number of years of appearance AC 10 was varied (increasing up to 20) until a point of indifference was reached between the two alternatives. Thus, the subject expressed the number of years they would be prepared to trade-off a poor appearance to ultimately achieve a good appearance. A higher utility score indicated a higher value placed on aesthetic tooth appearance.
Quality of life: OASIS
A series of questions assessed the degree of concern/disadvantage that children felt because of the arrangement of their teeth (Mandall et al., 1999). Children were asked to indicate, on a seven-point Likert scale, their concern about tooth appearance, nice comments about teeth, unpleasant comments about teeth, teasing about teeth, avoidance of smiling, and covering the mouth. These scores were then added to the child's self-perceived IOTN-AC score to obtain the OASIS. A higher score indicated more concern/negative impact from their tooth arrangement.
Measurement of compliance with treatment
Whether the patient completed the course of treatment was recorded together with the reason for non-completion. The reasons for non-completion included poor oral hygiene, failed appointments, or appliance breakages. The number of failed/cancelled appointments and the number of appliance breakages were recorded from the case notes, at the end of treatment, to give a further indicator of compliance with treatment.
|
Statistics analysis |
|---|
|
TopSummaryIntroductionSampleMethodStatistics analysisResultsDiscussionConclusionsFundingReferences |
|---|
Descriptive statistics were generated and the data were checked for normality. Statistical analysis was carried out on the whole sample (n = 144). Multiple regression analysis models were used to investigate the association between the baseline variables (independent variables) and patient completion of and co-operation with treatment (dependent variables).
Multiple regression analysis allows the effect of multiple variables to be assessed at the same time. Therefore, where completion of treatment was considered, the following potential explanatory variables were entered into the model: utility value, OASIS, age, gender, socio-economic status, type of appliance, and IOTN score. For compliance with treatment (missed appointments or breakages) the following potential explanatory variables were entered: age, gender, socio-economic status, and type of appliance.
|
Results |
|---|
|
TopSummaryIntroductionSampleMethodStatistics analysisResultsDiscussionConclusionsFundingReferences |
|---|
The weighted kappa value for examiner calibration was 0.86, standard error (SE) kappa 0.06, 95 per cent confidence interval (CI) 0.74–0.97 for the AC and DHC of IOTN. Intra-examiner reliability for the use of this index was as follows—AC: weighted kappa 0.81, SE kappa 0.06, 95 per cent CI 0.69–0.93, and DHC: weighted kappa 0.82, SE kappa 0.12, 95 per cent CI 0.61–1.00.
One hundred and fifty-five patients were registered, of which 11 declined to proceed with treatment. Therefore, the analysis was carried out on 144 subjects. The mean age at the start of treatment was 13.7 years [standard deviation (SD) 2.1 years] with a range of 9.6–19.0 years. There were 65 males (45 per cent) and 79 females (55 per cent) and the mean treatment duration was 16.6 months (SD = 9.3 months). Tables 1, 2 and 3 show descriptive statistics for the baseline variables and indicators of co-operation.
|
|
|
Of these 144 subjects, 82 completed treatment (57 per cent) and 62 (43 per cent) did not. For the 62 patients not completing treatment, the primary reasons for this were as follows: poor oral hygiene (n =19, 31 per cent); multiple failed appointments (n = 27, 43 per cent); appliance breakages (n = 10, 16 per cent); and reason not recorded (n = 6, 10 per cent).
There was no statistically significant difference between operators for patients who completed/did not complete treatment (chi-square value 4.89, 2 degrees of freedom, P = 0.09) or for the number of failed or cancelled appointments [analysis of variance (ANOVA) F ratio 1.7, P value = 0.18, 95 per cent CI –1.32 to 0.23]. However, there was a statistically significant difference between operators NAM and DF only, for number of appliance breakages (ANOVA F ratio 6.0, P value 0.003, 95 per cent CI 0.56 to 3.18).
Multiple regression analysis to investigate whether need for treatment or quality of life measures influenced completion of, or compliance with, orthodontic treatment
Importantly, none of the baseline variables, including quality of life measures, age, gender, socio-economic status, type of appliance, or IOTN score showed an association with completion of orthodontic treatment (logistic regression analysis; P > 0.05). In addition, multiple linear regression analysis revealed neither age, gender, and socio-economic status nor type of appliance to be significantly associated with the number of appliance breakages or number of failed/cancelled appointments (P > 0.05).
|
Discussion |
|---|
|
TopSummaryIntroductionSampleMethodStatistics analysisResultsDiscussionConclusionsFundingReferences |
|---|
This study has found that quality of life utility scores and OASIS questionnaire, clinical treatment need, age, gender, socio-economic status, and type of appliance were not useful in identifying whether a child might complete orthodontic treatment.
Effect of utility values and OASIS on co-operation with orthodontic treatment
It was surprising that quality of life measures did not influence completion of orthodontic treatment. At the time of starting this study, previous work had suggested that utility score and OASIS may influence the wish for treatment and the use of orthodontic services (Fox, 1997; Mandall et al., 1999). It was hypothesized that if these quality of life indicators could predict the uptake of orthodontic services, maybe patients with higher utility values or OASIS would place a higher value on their dental appearance and be more co-operative in completing treatment. However, a more recent prospective study showed no effect of these quality of life measures on uptake of orthodontic treatment (Mandall et al., 2005). It is probably, therefore, unnecessary to calculate utility values and OASIS for patients, in addition to clinical measures of need. However, another interpretation of the lack of effect of utility score on co-operation could be that the younger patients in the sample did not yet have the cognitive ability to imagine time periods of 10–20 years.
Effect of age, gender, socio-economic status, and type of appliance on co-operation
The lack of effect of age or gender on compliance agrees with Murray (1989) and Richter et al. (1998). In contrast, Haynes (1991) showed that 15- to 17-year-old patients had the highest discontinuation rates (39.8 per cent) compared with 10- to 14-year old patients at 21.3 per cent. Additionally, Tung and Kiyak (1998), investigating psychosocial influences on the timing of orthodontic treatment, found that 9- to 12-year-old children made more suitable candidates for orthodontic treatment. It may be that compliance with treatment may be more predictable using psychological outcomes rather than age, gender, or socio-economic status.
It was perhaps surprising that the type of appliance did not influence completion or co-operation with treatment, since Murray (1989) showed that use of removable appliances was associated with failure to complete treatment. It may be that the present data did not show this because all removable appliances were used for a short time at the beginning of a fixed phase of treatment. Therefore, the effect of a removable appliance only on completion and co-operation could not be evaluated.
Clinical need for treatment IOTN and co-operation
The lack of effect of IOTN on completion of treatment is surprising since El-Mangoury (1981) found that those with high need co-operated better, as assessed by psychometric tests. The difference in results could be explained by the measure of need in this study being clinical and the former being psychological. A study of a sample of school children in Greater Manchester showed that IOTN could be used to predict uptake of orthodontic treatment (Mandall et al., 2005). Therefore, it seems that high clinical need cannot be used to predict completion of treatment, although these children are more than three times more likely to access care.
|
Conclusions |
|---|
|
TopSummaryIntroductionSampleMethodStatistics analysisResultsDiscussionConclusionsFundingReferences |
|---|
Age, gender, socio-economic status, type of appliance, and clinical treatment need (IOTN) are not useful in helping a clinician to choose potentially co-operative patients. Similarly, quality of life measures (utility values or OASIS) do not add to our knowledge of who may complete orthodontic treatment.
|
Funding |
|---|
|
TopSummaryIntroductionSampleMethodStatistics analysisResultsDiscussionConclusionsFundingReferences |
|---|
NHS R&D Programme on Primary Dental Care, UK.
|
Acknowledgement |
|---|
The authors thank Jonathan Miller for registering patients and completing data forms and Barbara Doherty for assistance with data entry and data cleaning.
|
References |
|---|
|
TopSummaryIntroductionSampleMethodStatistics analysisResultsDiscussionConclusionsFundingReferences |
|---|
-
Adulyanon S, Vourapukjaru J, Sheiham A. Oral impacts affecting daily performance in a low dental disease Thai population. Community Dentistry and Oral Epidemiology (1996) 24:385–389.[CrossRef][Web of Science][Medline]
Atchison KA, Dolan TA. Development of the geriatric oral health assessment index. Journal of Dental Education (1990) 54:680–687.[Abstract]
Awad MA, Locker D, Korner-Bitensky N, Feine JS. Measuring the effect of intra-oral rehabilitation on health related quality of life in a randomised controlled clinical trial. Journal of Dental Research (2000) 79:1659–1663.
Brook PH, Shaw WC. The development of an index of orthodontic treatment priority. European Journal of Orthodontics (1989) 11:309–320.
Cunningham SJ, Hunt NP. Quality of life and its importance in orthodontics. Journal of Orthodontics (2001) 28:152–158.
Cureton SL, Regenitter FJ, Yancey JM. The role of the headgear calendar in headgear compliance. American Journal of Orthodontics and Dentofacial Orthopedics (1993) 104:387–394.[Web of Science][Medline]
Cushing AM, Sheiham A, Maizels J. Developing socio-dental indicators—the social impact of dental disease. Community Dental Health (1986) 3:3–17.[Medline]
Egolf RJ, BeGole EA, Upshaw HS. Factors associated with orthodontic patient compliance with intraoral elastic and headgear wear. American Journal of Orthodontics and Dentofacial Orthopedics (1990) 97:336–348.[Web of Science][Medline]
El-Mangoury NH. Orthodontic cooperation. American Journal of Orthodontics (1981) 80:604–622.[CrossRef][Web of Science][Medline]
Fox D. The utility of anterior tooth appearance (1997) Thesis, University of Manchester, UK.
Gilbert GH, Duncan RP, Heft MW, Dolan TA, Vogel WB. Multi-dimensionality of oral health in dentate adults. Medical Care (1998) 36:988–1001.[CrossRef][Web of Science][Medline]
Gooch BF, Dolan TA, Bourque LB. Correlates of self-reported dental health status upon enrolment in the Rand Health Insurance experiment. Journal of Dental Education (1989) 53:629–637.[Abstract]
Haynes S. Discontinuation of orthodontic treatment in the General Dental Services in England and Wales 1972–1979. British Dental Journal (1982) 152:127–129.[CrossRef][Web of Science][Medline]
Haynes S. Trends in the numbers of active and discontinued orthodontic treatments in the General Dental Services 1964–1986/87. British Journal of Orthodontics (1991) 18:9–14.[Medline]
Kressin N, Spiro A, Bosse R, Garcia R, Lazis L. Assessing oral health-related quality of life. Medical Care (1996) 34:416–427.[CrossRef][Web of Science][Medline]
Locker D. Measuring oral health: a conceptual framework. Community Dental Health (1988) 5:3–18.[Medline]
Locker D, Grushka M. The impact of dental and facial pain. Journal of Dental Research (1987) 66:1414–1417.
Locker D, Miller Y. Evaluation of subjective oral health status in an adult population. Journal of Public Health Dentistry (1994a) 54:167–176.[Web of Science][Medline]
Locker D, Miller Y. Subjectively reported oral health status in an adult population. Community Dentistry and Oral Epidemiology (1994b) 22:425–430.[CrossRef][Web of Science][Medline]
Mandall NA. The oral self-perception of ethnic groups (1997) Thesis, University of Manchester, UK.
Mandall NA, McCord JF, Blinkhorn AS, Worthington HV, O'Brien K D. Perceived aesthetic impact of malocclusion and oral self-perceptions in 14-15 year old Asian and Caucasian children in Greater Manchester. European Journal of Orthodontics (1999) 21:175–183.
Mandall NA, Wright J, Conboy FM, Kay E, Harvey L, O'Brien KD. Index of orthodontic treatment need predicts orthodontic treatment uptake. American Journal of Orthodontics and Dentofacial Orthopedics (2005) 128:703–707.[CrossRef][Web of Science][Medline]
Murray AM. Discontinuation of orthodontic treatment: a study of the contributing factors. British Journal of Orthodontics (1989) 16:1–7.[Abstract]
Nuttall NM, Steele JG, Pine CM, White D, Pitts NB. The impact of oral health on people in the U.K. in 1998. British Dental Journal (2001) 23:331–334.
O'Brien KD, Kay E, Fox D, Mandall NA. Assessing oral health outcomes for orthodontics—measuring health status and quality of life. Community Dental Health (1998) 15:22–26.[Medline]
Reisine ST, Fertig J, Weber J, Leder S. Impact of dental conditions on patient's quality of life. Community Dentistry and Oral Epidemiology (1989) 17:7–10.[CrossRef][Web of Science][Medline]
Richter DD, Nanda RS, Pramod KS, Sinha PK, Smith DW, Currier GF. Effect of behavior modification on patient compliance in orthodontics. Angle Orthodontist (1998) 68:123–132.[Web of Science][Medline]
Rosenberg D, Kaplan S, Senie R, Badner V. Relationships among dental functional status, clinical dental measures and generic health measures. Journal of Dental Education (1988) 52:653–657.[Abstract]
Sergl HG, Klages U, Zetner A. Pain and discomfort during orthodontic treatment: causative factors and effects on compliance. American Journal of Orthodontics and Dentofacial Orthopedics (1998) 114:684–691.[CrossRef][Web of Science][Medline]
Sinha PK, Nanda RS, McNeil DW. Perceived orthodontic behaviors that predict patient satisfaction, orthodontist-patient relationship, and patient adherence in orthodontic treatment. American Journal of Orthodontics and Dentofacial Orthopedics (1996) 110:370–377.[CrossRef][Web of Science][Medline]
Slade GD, Spencer AJ. Development and evaluation of the oral health impact profile. Community Dental Health (1994) 11:3–11.[Medline]
Strauss RP. Culture, dental professionals and oral health values in multi-cultural societies. Gerodontology (1996) 13:82–89.[CrossRef][Medline]
Strauss RP, Hunt RJ. Understanding the value of teeth to older adults: influences on the quality of life. Journal of the American Dental Association (1993) 124:105–110.[Abstract]
Tickle M, Craven R, Blinkhorn AS. An evaluation of a measure of subjective oral health status in the U.K. Community Dental Health (1997) 14:175–180.[Medline]
Townsend P. Deprivation. Journal of Social Policy (1987) 16:125–146.[CrossRef][Web of Science]
Tung AW, Kiyak HA. Psychological influences on the timing of orthodontic treatment. American Journal of Orthodontics and Dentofacial Orthopedics (1998) 113:29–39.[CrossRef][Web of Science][Medline]
United Kingdom Census. (1991) Scotland: Office of Population Census & Surveys, General Register Office.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||