Now Reading
Relationship of 3 indexes of orthodontic treatment need used by Medicaid and oral health–related quality of life

Relationship of 3 indexes of orthodontic treatment need used by Medicaid and oral health–related quality of life



American Journal of Orthodontics and Dentofacial Orthopedics, 2022-04-01, Volume 161, Issue 4, Pages 574-581, Copyright © 2021 American Association of Orthodontists


Introduction

This study aimed to assess the relationship between 3 indexes of orthodontic treatment need that are used by Medicaid, namely the Salzmann Index (SI), the handicapping labiolingual deviation (HLD) Index, and the HLD California Modification Index, and oral health–related quality of life (OHRQOL).

Methods

The orthodontic records of 100 participants aged 11-14 years were used to calculate occlusal index scores. The condition-specific oral impacts on daily performances (OIDP) index questionnaire was used to quantify OHRQOL and to identify detriments attributable to malocclusion-related conditions (MRCs). The relationship between occlusal index scores and OHRQOL was analyzed using descriptive statistics, Spearman rank-order and biserial correlations, and logistic regression.

Results

The mean index scores were: SI, 15.4; HLD, 13.2; and HLD California Modification, 15.8. Ninety percent of participants did not have normative orthodontic treatment need according to current index criteria. OIDP scores were not normally distributed, and the mean score was 3.1. Of those participants who reported an impact, 83% attributed at least 1 of those impacts to MRCs; however, 90% of these were of mild or moderate intensity. Smiling was the performance most impacted by MRCs. The only statistically significant correlation between an occlusal index and OIDP scores was for the SI, though this association was weak ( r = 0.27). None of the variables used in the logistic regression model (age, sex, 3 index scores) were significant predictors of OHRQOL.

Conclusions

No meaningful association exists between the 3 indexes studied and OHRQOL. These findings challenge the validity of current systems for the allocation of Medicaid-funded orthodontic treatment.

Highlights

  • Medicaid uses various occlusal indexes for orthodontic treatment coverage decisions.

  • Indexes of orthodontic treatment need are not associated with oral health–related quality of life.

  • Medicaid systems of treatment allocation fail to capture a psychosocial component.

  • The impact of malocclusion on oral health–related quality of life is low to moderate in severity.

Title XIX of the Social Security Act, commonly known as Medicaid, was passed in 1965 to afford health benefits to low-income patients. After the recommendations of an American Dental Association task force, the Act was amended in 1967 to establish the Early and Periodic Screening, Diagnosis, and Treatment program, which mandated coverage for treatment of handicapping malocclusions. , However, the breath of this mandate varies by state because each can determine its own definition of a handicapping malocclusion. Although such flexibility allows individual states to modify coverage according to budgetary considerations and public health goals, it has also led to disparities between states in terms of orthodontic treatment coverage granted through Medicaid.

Given the finite resources available for Medicaid expenditures, states have enacted measures to limit the number of patients receiving orthodontic benefits under this program. These include placing age restrictions, requiring prior authorization, and establishing strict definitions of what constitutes a handicapping malocclusion. The most common method for determining the presence of a handicapping malocclusion is the use of various occlusal indexes of orthodontic treatment need (OTN) such as the handicapping labiolingual deviation (HLD) Index, the HLD California Modification (CalMod) index, the IOTN, and the Salzmann Index (SI). These instruments were developed to quantify the severity of malocclusion and objectively identify those patients in most need of treatment. Higher numerical scores correspond to malocclusions of greater severity, and patients typically need to exceed a predetermined cutoff value to be eligible for coverage. While the British National Health Service uses the IOTN, Minick et al reported that 41 states used an index to determine handicapping malocclusions as of 2015, with a majority of these using the HLD, HLD CalMod, and SI.

Oral health–related quality of life (OHRQOL) is a multidimensional concept that includes a subjective evaluation of the perceived physical, psychological, and social aspects of oral health. As the goal of therapeutic interventions in medicine and dentistry has shifted to not only eliminate disease but also improve quality of life, especially for those conditions that are not life-threatening, subjective patient-based measures are increasingly used in dental epidemiologic research to complement normative, clinician-driven measures. , Although clinician-driven measures provide relevant information, patient-based measures provide more substantive information about the impacts of oral conditions on quality of life because patients are considered the best judges of their own OHRQOL. Despite their widespread use for determining orthodontic need, occlusal indexes place relatively little emphasis on patients’ perceptions of the difference orthodontic treatment will make in their daily lives. In 1985, the American Association of Orthodontics denied recognition of any occlusal indexes as “scientifically valid measure(s) of the need for orthodontic treatment” primarily because of their lack of esthetic or psychosocial components. ,

To overcome the limitations of normative assessment of need in oral health, a new sociodental approach has been advocated. , This model incorporates the use of OHRQOL measures, along with clinical measures, to establish a more holistic needs assessment. Indeed, research has shown that most patients undertake orthodontic treatment for psychosocial reasons rather than to improve physical factors such as occlusion. By measuring the impact of malocclusion on daily life, information about the potential benefits of treatment can be obtained. This can help an evaluator further prioritize treatment needs beyond normative assessment alone. de Oliveira and Sheiham found that 46% of Brazilian adolescents considered to need orthodontic treatment using the IOTN did not report any sociodental impacts, whereas 27% of adolescents who did report an impact were not considered to have an orthodontic need. Given these findings, O’Brien et al argued that patients falling into the latter cohort have greater need if one considers that most perceived benefits of orthodontic treatment appear to be psychosocial.

Several studies using various instruments for measuring OHRQOL have analyzed the relationship between malocclusion and OHRQOL and have reported equivocal results. Two systematic reviews and meta-analyses by Sun et al , demonstrated that diminished OHRQOL corresponded to increasing severity of the malocclusion. In contrast, Taylor et al concluded that malocclusion did not appear to affect general or OHRQOL. A review by Zhang et al agreed with these findings, but it also called for more rigorous evaluations of this relationship. A major methodological flaw of these prior studies is their use of generic OHRQOL measures because these do not allow researchers to identify the specific oral condition(s) responsible for changes in OHRQOL, thereby introducing possible confounders.

The oral impacts on daily performances (OIDP) index is the only OHRQOL measure that attributes reported oral impacts to specific dental conditions such as malocclusion. This allows evaluators to ensure that proposed orthodontic treatment will lead to an improvement in OHRQOL when impacts are attributed solely to malocclusion-related conditions (MRCs) as opposed to toothache or other non-MRCs. Moreover, the OIDP has shown good validity and reliability in populations of adults as well as children. , Bernabé et al assessed sociodental impacts attributed to malocclusion and normative orthodontic treatment need among Brazilian adolescents using the OIDP and IOTN, respectively. Their findings showed that over 50% of adolescents with a definitive need for orthodontic treatment did not report any impacts attributable to MRCs. The authors stated: “these inconsistencies in findings on normative needs and impacts underline the shortcomings of the normative approach to assessing need.”

Although the relationship of normative treatment need, as determined by the IOTN and other European indexes, and OHRQOL has been studied extensively, the indexes used in the United States for determining Medicaid coverage of orthodontic treatment have received little attention in the OHRQOL literature. Knowledge regarding these indexes’ ability, or lack thereof, to reflect an individual’s OHRQOL is needed to establish whether current Medicaid systems of treatment allocation meet the standards of the sociodental approach. Therefore, the present study aimed to assess the relationship between 3 American indexes of OTN and OHRQOL as measured by the OIDP index.

Material and methods

Participants were recruited from children starting orthodontic treatment at the Seton Hill University Center for Orthodontics residency clinic in Greensburg, PA. Most patients presented because of referral by their general dentist and/or by their parents’ desire to seek an orthodontic consultation. Recruitment of participants was carried out during the visit after the records appointment when a full diagnosis and treatment plan were presented. A convenience consecutive sampling technique was used until 100 participants had been enrolled in the study. A full explanation of the study design and rationale was given to the parent(s) or legal guardian(s) and the patient, and written informed consent and assent were obtained, respectively, to document voluntary participation in the study. Inclusion criteria consisted of participants aged 11-14 years with a complete set of orthodontic records (extraoral and intraoral photographs, panoramic and cephalometric radiographs, and digital study models). Participants with chronic systemic diseases, history of orthodontic treatment, and orthodontic records of insufficient diagnostic quality were excluded. Ethical approval was obtained from the Institutional Review Board of Seton Hill University.

The orthodontic records of participants were used to calculate scores for the SI, HLD, and HLD CalMod indexes. The index criteria used were obtained from the Medicaid Web sites of Pennsylvania (SI), Texas (HLD), and California (HLD CalMod). According to these criteria, a patient must have a SI score of ≥25 to be eligible for orthodontic coverage; for the HLD and HLD CalMod, this threshold is ≥26 (unless an automatic qualifying condition is present, see Discussion). Measurements from digital diagnostic casts were taken using the MyCadent software (Align Technology; San Jose, Calif). Two examiners were trained and calibrated on the use of these indexes using records from persons who were not part of the present study. On a separate occasion, the cohort was evenly divided and scored independently by the examiners (D.C. and M.R.). Two weeks later, each examiner re-scored 20% of their original patients, and 20% of the patients scored by the other examiner during the first session to assess intra- and interrater reliability.

OHRQOL was measured using the OIDP index, which has previously been validated for this study population. The OIDP quantifies oral impacts on 8 daily performances: eating, speaking, cleaning the mouth, relaxing, smiling, studying, emotion, and social contact. If a participant reported an impact on any performance, both the frequency (measured from 1 to 3) and severity (also measured from 1 to 3) of the impact were scored. The impact score per performance was then calculated by multiplying the frequency and severity scores. The intensity of the impact score was classified into 3 levels: mild (1-3), moderate (4-6), and severe (7-9). If no impact was reported, a zero was scored for that performance. The overall OIDP score was the sum of the 8 performance scores (possible scores ranged from 0 to 72) multiplied by 100 and divided by 72. In addition, participants were asked to choose from a list of oral problems the condition(s) that, in their opinion, caused the impact(s). Only the conditions of bad position of teeth, space between teeth, and deformity of mouth or face were considered MRCs.

Statistical analysis

Descriptive statistics were generated, and the data were checked for normality. OIDP scores were found not to be normally distributed using the Shapiro-Wilk test; therefore, nonparametric statistics were used. The relationship between the occlusal index and OIDP scores was assessed with Spearman rank-order correlations. The widely accepted thresholds of 0.2, 0.5, and 0.8 were used to define a weak, moderate, and strong correlation. A lack of association between the variables was considered the null hypothesis ( P >0.05). The predictive value of age, sex, and the occlusal index scores on the OIDP score was quantified with logistic regression. Using a dichotomized OIDP score, a binary logistic regression was carried out with OIDP score as the dependent variable and age, sex, and the occlusal index scores as predictors. Finally, Pearson product-moment correlation coefficients were used to assess intraexaminer and interexaminer reliability. Statistical analyses were performed using the SPSS for Windows (version 16.0; SPSS Inc, Chicago, Ill).

Results

Interexaminer reliability was 0.93 for the SI, 0.87 for the Handicapping Labiolingual Deviation Index, and 0.89 for the HLD CalMod (California modification) Index. Values for intraexaminer reliability were 0.96 and 0.99 (SI), 0.95 and 0.98 (HLD), and 0.96 and 0.98 (HLD CalMod). Overall examiner reliability was thus considered satisfactory.

A total of 100 participants were recruited; however, 20 were excluded from analysis because of missing values in their OIDP index questionnaires. The mean age was 12.3 years (SD, 1.1). The sample consisted of 42 female (52%) and 38 male (48%). The mean index scores were: SI, 15.4 (SD, 6.2; range, 3-37); HLD, 13.2 (SD, 7.5; range, 3-48); and HLD CalMod, 15.8 (SD, 7.1; range, 5-50). Normative treatment needs, according to the HLD CalMod was present in 8 (10%) participants, followed by the SI with 7 (9%), and the HLD with 5 (6%). Most participants (90%) did not have a normative need for orthodontic treatment according to current index criteria.

Fig 1 , Table I shows the frequency distribution of OIDP index scores; the data did not follow a normal distribution. The mean OIDP score was 3.1 (SD, 5.0; range, 0-27), and 34 (42.5%) participants had a score of zero. Of the remaining 46 participants who reported some impact on daily performances, 38 (82.6%) attributed at least 1 of those impacts, either entirely or in part, to a malocclusion-related condition ( Fig 2 , Table II ). The most impacted performance was smiling, with 32 participants reporting an impact on this performance because of a malocclusion-related condition. There was no reported impact on smiling that did not include at least 1 of the 3 malocclusion-related conditions as a cause. Other performances were impacted by malocclusion-related conditions to a considerably lesser extent (<10%). When the intensity of all malocclusion-attributed impacts was analyzed, 50.1% were considered of low intensity, 39.0% of moderate intensity, and 10.2% of high intensity.

Distribution of OIDP Index Scores.
Fig 1
Distribution of OIDP Index Scores.

Table I
Frequency distribution of OIDP index scores
Valid Frequency % Valid % Cumulative %
0 34 42.5 42.5 42.5
1 7 8.8 8.8 51.2
2 13 16.3 16.3 67.5
3 1 1.3 1.3 68.8
4 6 7.5 7.5 76.3
5 3 3.8 3.8 80.0
6 4 5.0 5.0 85.0
7 2 2.5 2.5 87.5
8 2 2.5 2.5 90.0
9 2 2.5 2.5 92.5
10 1 1.3 1.3 93.8
11 1 1.3 1.3 95.0
12 1 1.3 1.3 96.3
13 1 1.3 1.3 97.5
26 1 1.3 1.3 98.8
27 1 1.3 1.3 42.5
Total 80 100.0 100.0

Prevalence of Malocclusion-Related Conditions.
Fig 2
Prevalence of Malocclusion-Related Conditions.

Table II
Prevalence and intensity of MRCs
Variables Daily performances impacted
Eating Speaking Cleaning mouth Relaxing Emotion Smiling Studying Social contact Overall impact
Prevalence (n = 80) 7 7 6 2 2 32 0 3 38
Percentage 8.8 8.8 7.5 2.5 2.5 40.0 0 3.8 47.5
Intensity of impacts in 38 participants who reported impacts
Mild (1-3) 3 5 2 1 2 16 0 1 50.8
Moderate (4-6) 4 2 3 0 0 12 0 2 39.0
Severe (7-9) 0 0 1 1 0 4 0 0 10.2

Spearman rank-order correlation coefficients were calculated ( Fig 3 , Table III ). Among the 3 indexes, only the correlation between the SI and OIDP scores reached statistical significance at 0.27 ( P <0.05); r values for the HLD and HLD CalMod were 0.21 and 0.16, respectively. Next, because most participants reported OIDP index scores of zero, the data were dichotomized to form 2 groups—those participants who reported no impact (OIDP score = 0) and those participants who reported some impact (IODP score >0)—to determine if this would yield different results. The dichotomized data were used to calculate a biserial correlation coefficient, in which 1 variable is continuous (ie, occlusal index scores) and the other variable is a dichotomy. The correlation of the SI and OIDP scores was essentially unchanged (0.274), but the r values for the HLD and HLD CalMod were even lower (0.11 and 0.0001, respectively). In essence, very little relationship between the 3 indexes and OHRQOL was observed.

Intensity of Malocclusion-Related Conditions.
Fig 3
Intensity of Malocclusion-Related Conditions.

Table III
Spearman rank-order correlation coefficients
Printers Variables SI HLD HLD (CalMod) OIDP
Spearman’s rho SI Correlation coefficient 1.000 0.481 ∗∗ 0.232 0.273
P (2-tailed) <0.001 0.039 0.014
n 80 80 80 80
HLD Correlation coefficient 0.481 ∗∗ 1.000 0.839 ∗∗ 0.213
P (2-tailed) <0.001 <0.001 0.057
n 80 80 80 80
HLD (CalMod) Correlation coefficient 0.232 0.839 ∗∗ 1.000 0.161
P (2-tailed) 0.039 <0.001 0.154
n 80 80 80 80
OIDP Correlation Coefficient 0.273 0.213 0.161 1.000
P (2-tailed) 0.014 0.057 0.154
n 80 80 80 80

Correlation is significant at the 0.05 level (2-tailed).

∗∗ Correlation is significant at the 0.01 level (2-tailed).

A binary logistic regression with OIDP scores as the dependent variable and age, sex, and occlusal index scores as predictors yielded a Nagelkerke r 2 value of 0.09, indicating that only 9% of the variability in OHRQOL is accounted for by the 5 predictor variables. This is supported by the finding that none of the 5 variables achieved statistical significance as a predictor in the regression equation ( Table IV ). The model equation correctly predicted no impact on OHRQOL only 41.2% of the time.

Table IV
Variables in the logistic regression equation
Variables B Standard error Wald df P Exp(B)
Step 1 Sex (1) −0.376 0.497 0.572 1 0.449 .687
Age 0.011 0.245 0.002 1 0.964 1.011
SI 0.078 0.055 2.020 1 0.155 1.081
HLD 0.054 0.098 0.310 1 0.578 1.056
HLD (CalMod) −0.069 0.091 0.578 1 0.447 .933
Constant −0.436 3.179 0.019 1 0.891 .647

Variable(s) entered on step 1: Sex, Age, SI, HLD, and HLD (CalMod).

You're Reading a Preview

Become a DentistryKey membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here