Does Differential Item Functioning Jeopardize the Comparability of Health-Related Quality of Life Assessment Between Patients and Proxies in Patients with Moderate-to-Severe Traumatic Brain Injury?

BACKGROUND: Health-related quality of life (HRQoL) is clearly recognized as a patient-important outcome in patients with traumatic brain injury (TBI). Patient-reported outcomes are therefore often used and supposed to be directly reported by the patients without interpretation of their responses by a physician or anyone else. However, patients with TBI are often unable to self-report because of physical and/or cognitive impairments. Thus, proxy-reported measures, e.g., family members, are often used on the patient's behalf. Yet, many studies have reported that proxy and patient ratings differ and are noncomparable. However, most studies usually do not account for other potential confounding factors that may be associated with HRQoL. In addition, patients and proxies can interpret some items of the patient-reported outcomes differently. As a result, item responses may not only reflect patients' HRQoL but also the respondent's (patient or proxy) own perception of the items. This phenomenon, called differential item functioning (DIF), can lead to substantial differences between patient-reported and proxy-reported measures and compromise their comparability, leading to highly biased HRQoL estimates. Using data from the prospective multicenter continuous hyperosmolar therapy in traumatic brain-injured patients study (240 patients with HRQoL measured with the Short Form-36 (SF-36)), we assessed the comparability of patients' and proxies' reports by evaluating the extent to which items perception differs (i.e., DIF) between patients and proxies after controlling for potential confounders. METHODS: Items at risk of DIF adjusting for confounders were examined on the items of the role physical and role emotional domains of the SF-36. RESULTS: Differential item functioning was evidenced in three out of the four items of the role physical domain measuring role limitations due to physical health problems and in one out of the three items of the role emotional domain measuring role limitations due to personal or emotional problems. Overall, despite an expected similar level of role limitations between patients who were able to respond and those for whom proxies responded, proxies tend to give more pessimistic responses than patients in the case of major role limitations and more optimistic responses than patients in the case of minor limitations. CONCLUSIONS: Patients with moderate-to-severe TBI and proxies seem to have different perceptions of the items measuring role limitations due to physical or emotional problems, questioning the comparability of patient and proxy data. Therefore, aggregating proxy and patient responses may bias HRQoL estimates and alter medical decision-making based on these patient-important outcomes.

Posttraumatic growth inventory: challenges with its validation among French cancer patients.

BACKGROUND: The Posttraumatic growth inventory (PTGI) aims to assess the positive psychological changes that individuals can perceive after a traumatic life event such as a cancer diagnosis. Several French translations of the PTGI have been proposed, but comprehensive data on their psychometric properties are lacking. This study aimed to provide a more complete assessment of the psychometric properties of one of the most used PTGI translations in early-stage breast cancer and melanoma patients. METHODS: A sample of 379 patients completed the PTGI two years after their cancer diagnosis. A confirmatory analysis was first performed to determine whether the initial five-factor structure of the PTGI was adequate for this French version. As issues were identified in the translation and in the questionnaire structure, we performed an exploratory analysis to determine the most suitable structure for this questionnaire. Validity and reliability of the evidenced structured were then assessed. RESULTS: The exploratory analysis evidenced a four-factor structure close to the initial structure: four of the five initial domains were recovered, and items from the unrecovered domain were split into the other domains. This new structure showed good internal consistency and acceptable validity. CONCLUSIONS: This study highlights that the process of translation and cross-cultural validation of questionnaires is crucial to obtain valid and reliable psychometric instruments. We advise French psycho-oncology researchers and psychotherapists to (i) use the revised translation of Lelorain et al. (2010) proposed in this manuscript and (ii) use the four scores newly evidenced with a grouping of two response categories.

Evaluation of the link between the Guttman errors and response shift at the individual level.

PURPOSE: Methods for response shift (RS) detection at the individual level could be of great interest when analyzing changes in patient-reported outcome data. Guttman errors (GEs), which measure discrepancies in respondents' answers compared to the average sample responses, might be useful for detecting RS at the individual level between two time points, as RS may induce an increase in the number of discrepancies over time. This study aims to establish the link between recalibration RS and the change in the number of GEs over time (denoted index [Formula: see text]) via simulations and explores the discriminating ability of this index. METHODS: We simulated the responses of individuals affected or not affected by recalibration RS (defined as changes in the patients' standard of measurement) to determine whether simulated individuals with recalibration had a greater change in the number of GEs over time than individuals without recalibration. The effects of factors related to the sample, the questionnaire structure and recalibration were investigated. As an illustrative example, the change in the number of GEs was computed in patients suffering from eating disorders. RESULTS: Within simulations, simulated individuals affected by recalibration had, on average, a greater change in the number of GEs over time than did individuals without RS. Some of the parameters related to the questionnaire structure and recalibration magnitude appeared to have substantial effects on the values of [Formula: see text]. Discriminating abilities appeared, however, globally low. CONCLUSION: Some evidence of the link between recalibration and the change in GEs was found in this study. GEs could be a valuable nonparametric tool for RS detection at a more individual level, but further investigation is needed.

Identification of sources of DIF using covariates in patient-reported outcome measures: a simulation study comparing two approaches based on Rasch family models.

When analyzing patient-reported outcome (PRO) data, sources of differential item functioning (DIF) can be multiple and there may be more than one covariate of interest. Hence, it could be of great interest to disentangle their effects. Yet, in the literature on PRO measures, there are many studies where DIF detection is applied separately and independently for each covariate under examination. With such an approach, the covariates under investigation are not introduced together in the analysis, preventing from simultaneously studying their potential DIF effects on the questionnaire items. One issue, among others, is that it may lead to the detection of false-positive effects when covariates are correlated. To overcome this issue, we developed two new algorithms (namely ROSALI-DIF FORWARD and ROSALI-DIF BACKWARD). Our aim was to obtain an iterative item-by-item DIF detection method based on Rasch family models that enable to adjust group comparisons for DIF in presence of two binary covariates. Both algorithms were evaluated through a simulation study under various conditions aiming to be representative of health research contexts. The performance of the algorithms was assessed using: (i) the rates of false and correct detection of DIF, (ii) the DIF size and form recovery, and (iii) the bias in the latent variable level estimation. We compared the performance of the ROSALI-DIF algorithms to the one of another approach based on likelihood penalization. For both algorithms, the rate of false detection of DIF was close to 5%. The DIF size and form influenced the rates of correct detection of DIF. Rates of correct detection was higher with increasing DIF size. Besides, the algorithm fairly identified homogeneous differences in the item threshold parameters, but had more difficulties identifying non-homogeneous differences. Over all, the ROSALI-DIF algorithms performed better than the penalized likelihood approach. Integrating several covariates during the DIF detection process may allow a better assessment and understanding of DIF. This study provides valuable insights regarding the performance of different approaches that could be undertaken to fulfill this aim.