Overall performance of Oort's procedure for response shift detection at item level: a pilot simulation study

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.

Using structural equation modeling to investigate change and response shift in patient-reported outcomes: practical considerations and recommendations

BACKGROUND

Patient-reported outcomes (PROs) are of increasing importance for health-care evaluations. However, the interpretation of change in PROs may be obfuscated due to changes in the meaning of the self-evaluation, i.e., response shift. Structural equation modeling (SEM) is the most widely used statistical approach for the investigation of response shift. Yet, non-technical descriptions of SEM for response shift investigation are lacking. Moreover, application of SEM is not straightforward and requires sequential decision-making practices that have not received much attention in the literature.

AIMS

To stimulate appropriate applications and interpretations of SEM for the investigation of response shift, the current paper aims to (1) provide an accessible description of the SEM operationalizations of change that are relevant for response shift investigation; (2) discuss practical considerations in applying SEM; and (3) provide guidelines and recommendations for researchers who want to use SEM for the investigation and interpretation of change and response shift in PROs.

CONCLUSION

Appropriate applications and interpretations of SEM for the detection of response shift will help to improve our understanding of response shift phenomena and thus change in PROs. Better understanding of patients' perceived health trajectories will ultimately help to adopt more effective treatments and thus enhance patients' wellbeing.

Response shift in health-related quality of life measures in the presence of formative indicators

Background

Response shift (RS) has been defined as a change in the meaning of an individual’s self-evaluation that needs to be accounted for when assessing longitudinal changes in health-related quality of life (HRQoL). RS detection through structural equation modeling is accomplished by adopting Oort’s procedure based on a measurement model in which the observed variables are defined as reflective indicators of the HRQoL latent variable; that is, the latent variable causes the variation in the reflective indicators. This study aims to propose a procedure that assesses RS when formative indicators are used in measuring HRQoL; in this last case, the latent variable is considered to be a function of some formative indicators. A secondary aim is to compare the new procedure with Oort’s procedure to highlight similarities and differences.

Methods

The data were retrieved from a consecutive series of 258 patients newly diagnosed with colorectal cancer and undergoing chemotherapy and/or surgery. The European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire (EORTC QOL-C30) was administered twice, once before and once six months after treatment. Structural equation modeling was used to evaluate RS and true change with the newly proposed method (in which fatigue and pain were defined as formative indicators) and with Oort’s procedure (in which fatigue and pain were defined as reflective indicators).

Results

According to the new procedure, there was no measurement bias, and on average, patients’ quality of life improved by 3.53 points (on a scale ranging from 0 to 100) at the 6-month follow-up. With Oort’s procedure, the loading of the pain indicator was not invariant across the two time points, suggesting the presence of reprioritization, whereas the estimation of true change was very similar to the previous one: 3.87.

Conclusions

RS and true change in HRQoL can be evaluated in the presence of formative indicators. Defining a measurement model by formative or reflective indicators can lead to different results.

Comparison of structural equation modelling, item response theory and Rasch measurement theory-based methods for response shift detection at item level: A simulation study

When assessing change in patient-reported outcomes, the meaning in patients’ self-evaluations of the target construct is likely to change over time. Therefore, methods evaluating longitudinal measurement non-invariance or response shift at item-level were proposed, based on structural equation modelling or on item response theory. Methods coming from Rasch measurement theory could also be valuable. The lack of evaluation of these approaches prevents determining the best strategy to adopt. A simulation study was performed to compare and evaluate the performance of structural equation modelling, item response theory and Rasch measurement theory approaches for item-level response shift detection. Performances of these three methods in different situations were evaluated with the rate of false detection of response shift (when response shift was not simulated) and the rate of correct response shift detection (when response shift was simulated). The Rasch measurement theory-based method performs better than the structural equation modelling and item response theory-based methods when recalibration was simulated. Consequently, the Rasch measurement theory-based approach should be preferred for studies investigating only recalibration response shift at item-level. For structural equation modelling and item response theory, the low rates of reprioritization detection raise issues on the potential different meaning and interpretation of reprioritization at item-level.

Scoping review of response shift methods: current reporting practices and recommendations

BACKGROUND

Response shift (RS) has been defined as a change in the meaning of an individual's self-evaluation of his/her health status and quality of life. Several statistical model- and design-based methods have been developed to test for RS in longitudinal data. We reviewed the uptake of these methods in patient-reported outcomes (PRO) literature.

METHODS

CINHAHL, EMBASE, Medline, ProQuest, PsycINFO, and Web of Science were searched to identify English-language articles about RS published until 2016. Data on year and country of publication, PRO measure adopted, RS detection method, type of RS detected, and testing of underlying model assumptions were extracted from the included articles.

RESULTS

Of the 1032 articles identified, 101 (9.8%) articles were included in the study. While 54.5 of the articles reported on the Then-test, 30.7% of the articles reported on Oort's or Schmitt's structural equation modeling (SEM) procedure. Newer RS detection methods, such as relative importance analysis and random forest regression, have been used less frequently. Less than 25% reported on testing the assumptions underlying the adopted RS detection method(s).

CONCLUSIONS

Despite rapid methodological advancements in RS research, this review highlights the need for further research about RS detection methods for complex longitudinal data and standardized reporting guidelines.

A systematic review of the quality of reporting of simulation studies about methods for the analysis of complex longitudinal patient-reported outcomes data

PURPOSE

This study describes the characteristics and quality of reporting for published computer simulation studies about statistical methods to analyze complex longitudinal (i.e., repeated measures) patient-reported outcomes (PROs); we included methods for longitudinal latent variable measurement and growth models and response shift.

METHODS

Scopus, PsycINFO, PubMed, EMBASE, and Social Science Citation Index were searched for English-language studies published between 1999 and 2016 using selected keywords. Extracted information included characteristics of the study purpose/objectives, simulation design, software, execution, performance, and results. The quality of reporting was evaluated using published best-practice guidelines.

SYNTHESIS

A total of 1470 articles were reviewed and 42 articles met the inclusion criteria. The majority of the included studies (73.8%) investigated an existing statistical method, primarily a latent variable model (95.2%). Most studies specified the population model, including variable distributions, mean parameters, and correlation/covariances. The number of time points and sample size(s) were reported by all studies, but justification for the selected values was rarely provided. The majority of the studies (52.4%) did not report on model non-convergence. Bias, accuracy, and model fit were commonly reported performance metrics. All studies reported results descriptively, and 26.2% also used an inferential method.

CONCLUSIONS

While methodological research on statistical analyses of complex longitudinal PRO data is informed by computer simulation studies, current reporting practices of these studies have not been consistent with best-practice guidelines. Comprehensive reporting of simulation methods and results ensures that the strengths and limitations of the investigated statistical methods are thoroughly explored.