Internal Consistency Reliability of Patient-Reported Outcomes Measurement Information System Measures Used in Symptom Cluster Research: A Systematic Review and Reliability Generalization Meta-Analysis

Jeehye  Jun; Choa  Sung; Chang Gi  Park

doi:10.18502/ijph.v55i1.20968

Jeehye Jun Red Cross College of Nursing, Chung-Ang University, Seoul, Republic of Korea
Choa Sung Biobehavioral Nursing Science, College of Nursing, University of Illinois Chicago, Chicago, USA
Chang Gi Park Department of Population Health Nursing Science, College of Nursing, University of Illinois Chicago, Chicago, USA

DOI: https://doi.org/10.18502/ijph.v55i1.20968

Keywords: Symptom management; Internal consistency reliability; Systematic review; Meta-analyses

Abstract

Background: Patients with acute and chronic conditions often experience multiple symptoms known as symptom clusters. The Patient-Reported Outcomes Measurement Information System (PROMIS) is widely used to assess health status across various conditions, but its suitability for identifying symptom clusters remains unclear. Therefore, we examined the internal consistency reliability of PROMIS tools used to measure symptom clusters in adults through a systematic review and a reliability generalization meta-analysis.

Methods: We searched four electronic databases (PubMed, CINAHL, ProQuest, and Embase) for relevant articles published through December 31, 2024, including studies that measured symptom clusters in adults using at least one PROMIS measure. Meta-regression using a random effects model was performed to assess study heterogeneity, and funnel plots were employed to evaluate publication bias.

Results: The systematic review included 24 studies of 27,982 subjects with or without diseases in community, inpatient, and outpatient settings. Twenty PROMIS domains were used for symptom cluster research, and anxiety and depression were the most frequently used domains. In our reliability generalization meta-analysis of four studies, Cronbach’s alpha coefficients indicated good internal consistency reliability across five PROMIS domains (anxiety, depression, fatigue, pain, and sleep disturbance), with an average reliability of 0.91.

Conclusion: PROMIS measures may be reliable for assessing symptom clusters in adults and could serve as valuable tools for researchers and clinicians in patient assessment and symptom management. Nevertheless, future research should rigorously examine the reliability and validity of PROMIS tools in this context.