Psychometric properties of the EQ-5D-3L in South Australia: a multi-method non-preference-based validation study.

OBJECTIVE: Although HRQoL tools such as the EQ-5D-3L are significant in determining health status, these measures have not been validated in general populations in Australia. This study aims to psychometrically validate the EQ-5D-3L in a large population sample in Australia for the first time. METHODS: The EQ-5D-3L was included in the Dental Care and Oral Health study (DCOHS), conducted in a South Australian population sample. The participants were 23-91 years old, and 44.1% were male. The EQ-5D-3L was responded to on a three-point rating scale ("none"/"no", "some" and "extremely"/"unable"/"confined"). We employed the area under the receiver operating characteristic curve (AUROC) to evaluate whether the EQ-5D-3L total score could identify participants with diagnosed diseases and mental health disorders. Psychometric validation of the EQ-5D-3L investigated dimensionality with Exploratory Graph Analysis, model fit, floor/ceiling effects and criterion validity. RESULTS: The EQ-5D-3L comprised two dimensions, Activities and Symptoms. According to Root Mean Squared Error of Approximation (RMSEA) (<.05) and Comparative Fit Index (CFI) (>.950), the 2-dimensional structure showed excellent model fit with good reliability for the Activities subscale (Ωc = 0.80-95% CI [0.77, 0.83]), and poor reliability for the Symptom subscale (Ωc = 0.56-95% CI [0.53, 0.58]). The EQ-5D-3L showed adequate reliability (Ωc = 0.70-95% CI [0.67, 0.72]). The EQ-5D-3L showed good discrimination for diagnosed diseases (ranging from 64.3% to 86.3%). Floor/ceiling effects were observed across all items. The EQ-5D-3L total score discriminated between respondents who were experiencing health conditions (e.g. cancer, cardiovascular disease, stroke) from healthy individuals. DISCUSSION: Despite the ceiling effects, the EQ-5D-3L displayed good psychometric properties as an HRQoL measure and discriminated between health states in the general South Australian population. Further research should investigate the psychometric properties of the EQ-5D-5L in South Australia and whether an increased number of response categories can mitigate the observed ceiling effects.

Comparison of the Tree-Based Machine Learning Algorithms to Cox Regression in Predicting the Survival of Oral and Pharyngeal Cancers: Analyses Based on SEER Database.

This study aims to demonstrate the use of the tree-based machine learning algorithms to predict the 3- and 5-year disease-specific survival of oral and pharyngeal cancers (OPCs) and compare their performance with the traditional Cox regression. A total of 21,154 individuals diagnosed with OPCs between 2004 and 2009 were obtained from the Surveillance, Epidemiology, and End Results (SEER) database. Three tree-based machine learning algorithms (survival tree (ST), random forest (RF) and conditional inference forest (CF)), together with a reference technique (Cox proportional hazard models (Cox)), were used to develop the survival prediction models. To handle the missing values in predictors, we applied the substantive model compatible version of the fully conditional specification imputation approach to the Cox model, whereas we used RF to impute missing data for the ST, RF and CF models. For internal validation, we used 10-fold cross-validation with 50 iterations in the model development datasets. Following this, model performance was evaluated using the C-index, integrated Brier score (IBS) and calibration curves in the test datasets. For predicting the 3-year survival of OPCs with the complete cases, the C-index in the development sets were 0.77 (0.77, 0.77), 0.70 (0.70, 0.70), 0.83 (0.83, 0.84) and 0.83 (0.83, 0.86) for Cox, ST, RF and CF, respectively. Similar results were observed in the 5-year survival prediction models, with C-index for Cox, ST, RF and CF being 0.76 (0.76, 0.76), 0.69 (0.69, 0.70), 0.83 (0.83, 0.83) and 0.85 (0.84, 0.86), respectively, in development datasets. The prediction error curves based on IBS showed a similar pattern for these models. The predictive performance remained unchanged in the analyses with imputed data. Additionally, a free web-based calculator was developed for potential clinical use. In conclusion, compared to Cox regression, ST had a lower and RF and CF had a higher predictive accuracy in predicting the 3- and 5-year OPCs survival using SEER data. The RF and CF algorithms provide non-parametric alternatives to Cox regression to be of clinical use for estimating the survival probability of OPCs patients.

Direct effect of common mental disorders on xerostomia in adults estimated by marginal structural models: A population-based study.

OBJECTIVE: To investigate the controlled direct effect (CDE) of common mental disorders (CMD) on xerostomia which is not mediated by psychoactive medication intake. METHODS: A population-based cross-sectional study was conducted with 1720 individuals aged 20-59 years from Florianópolis, Brazil, in 2009. A two-stage cluster sample design was adopted, with a systematic selection of census tracts, followed by a random selection of households. All inhabitants of the selected households belonging to the age group of interest were considered eligible to participate. CMD were assessed using the Self-Reporting Questionnaire (SRQ > 7). Xerostomia was the outcome assessed by asking: "How often do you feel dry mouth?" with responses "often" and "always" taken as indicating xerostomia. Psychoactive medication intake was set as the mediator. Confounding factors included sex, age, per capita family income, educational attainment and smoking status. Marginal structural models were performed to quantify the CDE of CMD on xerostomia. Missing data were handled by multiple imputation by chained equations, and sensitivity analysis for unmeasured confounding was carried out. RESULTS: The overall prevalence of xerostomia was 8.8% and 15.3% of the individuals presented with CMD. After adjustment, the prevalence of xerostomia among individuals with CMD was 3.2 times higher than that among individuals without CMD (PR = 3.2 [95% CI: 2.3; 4.5]). The CDE of CMD on xerostomia which is not mediated by the medication intake resulted in a three times greater prevalence of xerostomia (PR = 3.0 95% CI: 2.0; 4.5) among those individuals with CMD. CONCLUSIONS: Common mental disorders is directly and negatively associated with xerostomia. Clinicians should bear in mind the emotional state when investigating the underlying factors and managing patients with xerostomia.