Meaningful differences and changes for five Patient‐Reported Outcomes Measurement Information System domains in a large cohort of patients with cancer
Cancer · 2026
DOI: 10.1002/cncr.70219 ↗Kurz gefasst
Fachartikel in der Zeitschrift Cancer, 2026. Verfasst von Karen Llave und 6 weiteren Autor:innen. Beteiligte Einrichtung: Department of Medical Social Sciences, Northwestern University Feinberg School of Medicine, Chicago, Illinois, USA. Hauptthemen laut MEDLINE: Patient Reported Outcome Measures, Neoplasms. Die Arbeit nennt 34 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NCATS NIH HHS, NCI NIH HHS, National Center for Advancing Translational Sciences.
Abstract (Originalsprache)
Background
Patient‐reported outcomes are valuable sources of information in research and clinical care, particularly in populations experiencing long‐term effects on health‐related quality of life such as patients with cancer. Identifying meaningful differences and changes in patient‐reported outcomes can prove useful in research and clinical settings.
Methods
Data from a system‐wide symptom‐management intervention trial were analyzed. Patients with cancer completed Patient‐Reported Outcomes Measurement Information System (PROMIS) computer‐adaptive tests in pain, fatigue, depression, anxiety, and physical function at baseline and 6 months. The authors used a combination of distribution‐based and anchor‐based approaches to estimate meaningful differences in PROMIS scores at baseline. Meaningful change estimates were calculated using mean changes in PROMIS scores in response to selected anchors.
Results
The sample ( n = 3907) was predominantly female (66.1%) and non‐Hispanic (84.2%). Overall, the sample reported PROMIS scores close to the general population standard (mean ± standard deviation score, 50 ± 10). Longitudinal anchor‐based estimates ranged from 2.4 to 8.0 points for pain interference, 2.0–4.4 points for depression, 2.2–4.5 points for anxiety, 2.0–6.3 points for fatigue, and 1.9–4.4 points for physical functioning. Longitudinal, anchor‐based findings revealed smaller changes than cross‐sectional differences indicating the responsiveness of PROMIS measures to patient‐reported meaningful changes over time. Supporting distribution‐based information using SEM criteria produced generally smaller values, ranging from 1.14 for depression to 2.14 for fatigue.
Conclusions
PROMIS computer‐adaptive test measures of pain, fatigue, anxiety, depression, and physical function are sensitive to group differences and changes over time. Regular symptom monitoring in conjunction with the recommended thresholds could be useful in research and clinical settings.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Medical Social Sciences, Northwestern University Feinberg School of Medicine, Chicago, Illinois, USA.
- Department of Preventive Medicine, Northwestern University Feinberg School of Medicine, Chicago, Illinois, USA.
ORCID-Profile: Karen Llave
Förderung
- NCATS NIH HHS
- NCI NIH HHS
- National Center for Advancing Translational Sciences
- National Cancer Institute
Zitieren
APA
Llave, K., Kim, J., Kuharic, M., Jackson, K., Lancki, N., Webster, K. A., & Cella, D. (2026). Meaningful differences and changes for five Patient‐Reported Outcomes Measurement Information System domains in a large cohort of patients with cancer. Cancer. https://doi.org/10.1002/cncr.70219
Vancouver
Llave K, Kim J, Kuharic M, Jackson K, Lancki N, Webster KA, et al. Meaningful differences and changes for five Patient‐Reported Outcomes Measurement Information System domains in a large cohort of patients with cancer. Cancer. 2026. doi:10.1002/cncr.70219
BibTeX
@article{Llave2026,
author = {Llave, Karen and Kim, Jiwon and Kuharic, Maja and Jackson, Kathryn and Lancki, Nicola and Webster, Kimberly A and Cella, David},
title = {Meaningful differences and changes for five Patient‐Reported Outcomes Measurement Information System domains in a large cohort of patients with cancer},
journal = {Cancer},
year = {2026},
doi = {10.1002/cncr.70219}
}
EUVIMED