Evaluation of psychometric properties and differential item functioning of 8-item Child Perceptions Questionnaires using item response theory

Impact of primary dentition malocclusion on the oral health-related quality of life in preschoolers

BACKGROUND

Malocclusion is a condition frequently seen in primary dentition due to the interaction of environmental, genetic and behavioural factors. The occurrence of some types of malocclusions can have an impact on oral health-related quality of life in children. Hence, the present study aimed to verify the impact of primary dentition malocclusion on oral health-related quality of life in preschool children.

METHODS

A population-based cross-sectional study was conducted in Florianopolis, Brazil, with a representative sample of 1050 preschoolers aged between 2 and 5 years, randomly selected. Parents answered the Brazilian version of the Early Childhood Oral Health Impact Scale and also to a questionnaire on socio-economic indicators. Data obtained from the questionnaire were obtained by item response theory based on model of gradual response. The malocclusion assessed was: anterior open bite, increased overjet and posterior crossbite. Poisson regression model was employed for multivariate analysis (P < 0.05).

RESULTS

Malocclusion was observed in 36.7% of the children. Of these, 11.4% were anterior open bite, 67.2% were increased overjet, and 21.4% were posterior crossbite. Malocclusion's impact on oral health-related quality of life was 28.6%. In children aged 4-5 years, the prevalence of malocclusion's impact on quality of life was 49.5% higher than in children aged 2-3 years. Statistical analysis showed that preschool children with malocclusion showed no significant impact on quality of life.

CONCLUSIONS

The findings of the present study indicate that the occurrence of primary dentition malocclusion has no impact on the quality of life of children aged 2-5 years.

Validity and reliability of the Child Perceptions Questionnaire 11-14 for Colombian school children

PURPOSE

To evaluate the validity and reliability of the Child Perceptions Questionnaire 11-14 (CPQ11-14) for children attending public educational institutions in Floridablanca (Colombia).

METHODS

An evaluation study of diagnostic technology was conducted. The questionnaire was completed by 500 students aged 11-14 years who were selected by simple random sampling. The test-retest reproducibility was evaluated in 36 students chosen at random from one of the schools; the questionnaire was completed a second time by these students, with an interval of two weeks between administrations. Internal consistency was estimated using the Cronbach's alpha coefficient, and test-retest reproducibility was determined with the intraclass correlation coefficient (ICC). The Rasch Rating Scale Model was used to perform the validation; the structure of the categories, the quality of the items, the invariance and the one-dimensionality were analyzed; invariance was confirmed using differential item functioning (DIF). Informed consent was obtained from the parents or guardians, and assent was obtained from the minors.

RESULTS

The mean age was 12.7 ± 1.1 years, and 210 (42.0%) participants were female. The Cronbach's alpha and ICC values were 0.91 and 0.63, respectively. The 37 items fit the Rasch model, but a uniform DIF was observed for an item that was eliminated. The questionnaire was not one-dimensional, and the proportion of variance explained by the model was poor (26.3%).

CONCLUSION

The CPQ11-14 showed unsatisfactory test-retest reproducibility and excellent internal consistency; no unidimensionality was observed, but the items fit the Rasch model.

Validating the Foot and Ankle Outcome score for measuring foot dysfunction among hallux valgus surgery patients using item response theory

BACKGROUND

There is an absence of high quality research validating instruments that measure foot and ankle related quality of life among hallux valgus (bunion) patients' perspectives. The Foot and Ankle Outcome Scale is a patient-reported outcome instrument, that when administered to patients with symptomatic hallux valgus, provides a patient-centric perspective of their foot function. The aim of this study is to assess the psychometric properties of the instrument's five subscales among preoperative bunion surgery patients.

METHODS

The Foot and Ankle Outcome Scale instrument measures Pain, Symptoms, Activities of Daily Living, Sport and Recreational Activities and Foot/Ankle Related Quality of Life. Preoperative data is collected from a sample of patients scheduled for surgical treatment of their condition in Vancouver, Canada. Classical and item response theory methods are used to report on reliability, validity and differential item functioning among subgroups.

RESULTS

This study included 249 surveys, representing an overall response rate of 44.1% among 564 eligible patients. The instrument demonstrated high reliability for all subscales, though 18 items across subscales, exhibited poor discrimination between item levels. Four items score differently according to patients' sex and one item scored differently by age.

CONCLUSIONS

The instrument measures five domains of health important to bunion patients. These findings suggest that the current instrument can be used with an understanding of its limitations, including redundant questions and sex-based differences. Future research should revise a number of items. The results highlight the importance of the psychometric analyses of instruments in specific patient populations.

Development and evaluation of CARIES-QC: a caries-specific measure of quality of life for children

BACKGROUND

Existing paediatric oral health-related quality of life (OHRQoL) measures are generic instruments designed to evaluate a range of oral conditions. It has been found that disease-specific measures may be more adept at detecting subtle changes which occur following treatment of the condition in question. Furthermore, existing self-report OHRQoL measures have not involved children at all stages of development of the measure. The aim of this study was to develop a caries-specific measure of quality of life for children.

METHODS

The first stage of the study involved a qualitative enquiry with children, aged 5-16 years, to inform the development of the measure. Children generated the potential items, contributed to item reduction and questionnaire design and participated in the testing of face and content validity. The resulting measure was evaluated in a cross-sectional validation study. Ethical approval was granted for the study.

RESULTS

The qualitative study found that children discussed a number of caries-related impacts which affected their daily lives. These were incorporated into a draft measure which was further refined following testing of face and content validity. This resulted in the production of the Caries Impacts and Experiences Questionnaire for Children (CARIES-QC), comprising 16 items and one global question. Two hundred participants with a mean (range) age of 8.1 (5-16) years took part in the further evaluation of CARIES-QC. Four items, which did not fit the Rasch model, were removed from further analysis. The remaining 12 items demonstrated good internal consistency (alpha = 0.9) and the total score showed significant correlations with the number of decayed teeth, presence of pain, pulpal involvement, the Child Perceptions Questionnaire (16-item short form) and the global score (p < 0.01, Spearman's rho).

CONCLUSION

In conclusion, children's input allowed the development of a valid and reliable child-centred caries-specific quality of life measure. CARIES-QC can now be used to evaluate which interventions for dental caries are most effective in reducing impacts from the child's perspective.

Longitudinal measurement invariance and explanatory IRT models for adolescents' oral health-related quality of life

BACKGROUND

Longitudinal invariance is a perquisite for a valid comparison of oral health-related quality of life (OHRQoL) scores over time. Item response theory (IRT) models can assess measurement invariance and allow better estimation of the associations between predictors and latent construct. By extending IRT models, this study aimed to investigate the longitudinal invariance of the two 8-item short forms of the Child Perception Questionnaire (CPQ_11-14) regression short form (RSF:8) and item-impact short form (ISF:8) and identify factors associated with adolescents' OHRQoL and its change.

METHODS

All students from S1 and S2 (equivalent to US grades 6 and 7) who were born in April 1997 and May 1997 (at age 12) from 45 randomly selected secondary schools were invited to participate in this study and followed up after 3 years. Data on the CPQ_11-14 RSF:8 and CPQ_11-14 ISF:8, demographics, oral health behavior and status were collected. Explanatory graded response models were fitted to both short forms of the CPQ_11-14 data for assessing longitudinal invariance and factors associated with OHRQoL. The Bayesian estimation method - Monte Carlo Markov Chain (MCMC) with Gibbs sampling was adopted for parameter estimation and the credible intervals were used for inference.

RESULTS

Data from 649 children at age 12 at baseline and 415 children at age 15 at follow up were analyzed. For the 12 years old children, healthier oral health behavior, better gum status, families with both parents employed and parents' education level were found to be associated with better OHRQoL. Four items among the 2 short forms lacked longitudinal invariance. With statistical adjustment of longitudinal invariance, OHRQoL were found improved in general over the 3 years but no predictor was associated with OHRQoL in follow-up. For those with decreased family income, their OHRQoL had worsened over 3 years.

CONCLUSIONS

IRT explanatory analysis enables a more valid identification of the factors associated with OHRQoL and its changes over time. It provides important information to oral healthcare researchers and policymakers.

Evaluation of the Fecal Incontinence Quality of Life Scale (FIQL) using item response theory reveals limitations and suggests revisions

PURPOSE

The Fecal Incontinence Quality of Life Scale (FIQL) is a commonly used patient-reported outcome measure for fecal incontinence, often used in clinical trials, yet has not been validated in English since its initial development. This study uses modern methods to thoroughly evaluate the psychometric characteristics of the FIQL and its potential for differential functioning by gender.

METHODS

This study analyzed prospectively collected patient-reported outcome data from a sample of patients prior to colorectal surgery. Patients were recruited from 14 general and colorectal surgeons in Vancouver Coastal Health hospitals in Vancouver, Canada. Confirmatory factor analysis was used to assess construct validity. Item response theory was used to evaluate test reliability, describe item-level characteristics, identify local item dependence, and test for differential functioning by gender.

RESULTS

236 patients were included for analysis, with mean age 58 and approximately half female. Factor analysis failed to identify the lifestyle, coping, depression, and embarrassment domains, suggesting lack of construct validity. Items demonstrated low difficulty, indicating that the test has the highest reliability among individuals who have low quality of life. Five items are suggested for removal or replacement. Differential test functioning was minimal.

CONCLUSIONS

This study has identified specific improvements that can be made to each domain of the Fecal Incontinence Quality of Life Scale and to the instrument overall. Formatting, scoring, and instructions may be simplified, and items with higher difficulty developed. The lifestyle domain can be used as is. The embarrassment domain should be significantly revised before use.

A Rasch Analysis between Schizophrenic Patients and the General Population

Background

The aim of this study was to test the General Oral Health Assessment Index Questionnaire (GOHAI) items for differential item functioning (DIF) according to demographic characteristics (gender, age) and mental health status (schizophrenic disorders versus general population) using Rasch analysis.

Methods

This is a cross-sectional study using aggregated baseline data from a validation study of the GOHAI in the French general population (GP) and similar validation study in persons with schizophrenia (PWS). DIF was tested using the Partial Credit Model. DIF were estimated in different groups of patients.

Results

The cohort comprised 363 persons: 65% were female, 83% were 25-45 years old and 30% were PWS.Five of the 12 items exhibited DIF. DIF effects were observed with schizophrenia for 3 items, with age for 3 items and with gender for 1 item. The variable “age” gave a significant explanation of the latent variable: the latent variable decreased with age (-0.40±0.08-p<0.001 for each increase of the age of 10 years). This decrease represented an effect size of 0.27 which can be qualified of a small to medium effect. The status of the individuals (GP versus PWS) and the gender did not significantly explain differences in the values of the latent variable.

Conclusion

The GOHAI scores may not be comparable across sub-groups defined by health status, age and gender without accounting for DIF. In the future, other studies should explore this way with other Oral Health related Quality of Life assessment tools and populations with mental illness.