Response shift results of quantitative research using patient-reported outcome measures: a meta-regression analysis

Abstract

Purpose: Our objectives were to identify characteristics of response shift studies using patient-reported outcomes (PROMs) that explain variability in (1) the detection and (2) the magnitude of response shift effects.

Methods: We conducted a systematic review of quantitative studies published before June 2023. First, two-level multivariable logistic regression models (effect- and sample-levels) were used to explain variability in the probability of finding a response shift effect. Second, variability in effect sizes (standardized mean differences) was investigated with 3-level meta-regression models (participant-, effect- and sample-levels). Explanatory variables identified via the purposeful selection methodology included response shift method and type, and population-, study design-, PROM- and study-quality characteristics.

Results: First, logistic regression analysis of 5597 effects from 206 samples in 171 studies identified variables explaining 41.5% of the effect-level variance, while no variables explained sample-level variance. The average probability of response shift detection is 0.20 (95% CI: 0.17-0.28). Variation in detection was predominantly explained by response shift methods and type (recalibration vs. reprioritization/reconceptualization). Second, effect sizes were analyzed for 769 effects from 114 samples and 96 studies based on the then-test and structural equation modeling methods. Meta-regression analysis identified variables explaining 11.6% of the effect-level variance and 26.4% of the sample-level variance, with an average effect size of 0.30 (95% CI: 0.26-0.34).

Conclusion: Response shift detection is influenced by study design and methods. Insights into the variables explaining response shift effects can be used to interpret results of other comparable studies using PROMs and inform the design of future response shift studies.

Keywords: Effect sizes; Effects; Meta-regression; Patient reported outcomes; Response shift.

Fig. 1

PRISMA flow diagram. Notes: RS = Response shift. PROM = Patient-reported outcome measure. HRQOL = Health-related quality of life. ^a)Reasons are ranked by the first identified reason for exclusion. ^b) When method- and result- sections were not explicitly related to RS analyses, these studies were excluded (e.g., longitudinal measurement invariance studies were excluded if they did not explicitly follow a recognized SEM method for response shift detection (e.g., Oort’s SEM approach)

Fig. 2

Relationship between study-level observations, variance levels of the modeling procedure, and explanatory/control variables. Notes: ¹The samples are assumed to be independent. ²Accounts for instances where multiple effects are based on the same sample. ³Accounts for individual-level variation among the participants. These are based on the variances (and sample sizes) of the primary studies. ⁴ The study design and study quality characteristics are effect-level variables because some samples are investigated in more than one study and therefore there may be variability in these characteristics within samples

Fig. 3

Partitioning of explained variation in detection of response shift effects. Notes: The size of each slice and its corresponding value indicates the percentage of effect-level explained variance. The explained variance attributable to the different response shift methods, when compared to the then-test method, are partitioned out in the outer ring. The percentages do not add up exactly to their corresponding totals due to rounding errors of parameter estimates. Sample-level explained variance is not depicted because the sample-level variables (medical conditions and interventions) did not statistically significantly explain any of the sample-level variance (R² = 7.2%, p = 0.19)