Abstract

Objectives

To further evaluate the psychometric properties of the National Comprehensive Cancer Network/Functional Assessment of Cancer Therapy Lymphoma Symptom Index-18 (NFLymSI-18), including estimates of meaningful within-patient change thresholds, to support its ability to reliably and validly measure clinical- and patient-relevant outcomes in diverse research settings.

Methods

Data from a longitudinal multicenter clinical trial in patients with indolent B cell lymphoma and a single-institution observational study of patients with Hodgkin’s and non-Hodgkin’s lymphoma were examined using inferential statistics. Available data were analyzed separately and then used collectively to summarize evidence of NFLymSI-18 reliability and validity. Effect sizes for known-groups comparisons and responsiveness analyses were calculated to determine meaningful score thresholds.

Results

Analyses indicated that the NFLymSI-18 demonstrated good internal consistency reliability (Cronbach’s alpha > 0.70) for all 3 subscales (with 2 of them > 0.80), and excellent (> 0.90) for the NFLymSI-18 Total in the observational study. Stability (test-retest reliability) exceeded preset thresholds for good reliability for total score and all but one subscale. Known-groups analysis confirmed construct validity, often with moderate-to-large effect sizes, particularly between categories of performance status ratings in the observational study (effect size between −0.33 to −1.06). Responsiveness to changes in clinical status was significant, indicating a sensitivity to within-patient clinical changes. Clinically meaningful individual patient change thresholds were estimated to be 5 to 6 points for the NFLymSI-18 Total score and 2-4 points for the Disease-Related Symptoms-Physical score.

Conclusion

The NFLymSI-18 is reliable, valid, and responsive and includes suggested values for individual patient change when planning lymphoma clinical trials.