Exploratory Psychometric Assessment of the Endler Multidimensional Anxiety Scales in Romanian Hemodialysis Patients: Reliability, Convergent Validity, and Domain-Level Structure.

Background and

Objectives

Chronic kidney disease (CKD) is an increasingly important global health challenge and is frequently accompanied by psychiatric symptoms, including anxiety. A multidimensional assessment of anxiety in hemodialysis (HD) using the Endler Multidimensional Anxiety Scales (EMAS) has not, to our knowledge, been previously reported. We aim to evaluate the reliability, convergent validity, and exploratory domain-level structure of EMAS in HD patients treated at a dialysis center in Craiova, Romania. Materials and

Methods

A total of 103 HD patients underwent clinical and sociodemographic/socioeconomic profiling, cognitive screening using the Mini-Mental State Examination (MMSE), and EMAS administration at two time points (4-week interval) for test-retest evaluation. The anxiety subscale of the Depression, Anxiety, and Stress Scale-21R (DASS-21R) was administered to assess convergent validity. Internal consistency (Cronbach's α), temporal stability (test-retest correlations and intraclass correlation coefficients), and convergent validity (Pearson correlations) were computed. Exploratory factor analyses were conducted on EMAS domain scores (state, trait, and perceived anxiety domains) as an exploratory structural check.

Results

EMAS state and trait anxiety scores were higher in women than in men, while perceived anxiety showed a more heterogeneous pattern across dimensions. Total state anxiety increased with age, particularly after 50 years. Domain-level internal consistency was good for state and acceptable for trait components (standardized α ≈ 0.84 and 0.78 across administrations), whereas perceived anxiety domains showed low cross-domain coherence, consistent with context-specific appraisal. The DASS-21R anxiety subscale showed good internal consistency (α = 0.863). Convergent validity analyses indicated small, domain-specific associations between EMAS scores and DASS-21R anxiety. Domain-level EFA supported a theoretically coherent pattern in which state and trait domains clustered distinctly, while perceived anxiety domains formed a partially separable factor; this pattern was broadly consistent across both administrations.

Conclusions

In this HD cohort, EMAS demonstrated good reliability and limited but domain-specific evidence of convergent validity, and exploratory domain-level analyses supported its multidimensional organization. Further studies with larger samples are warranted for item-level structural testing and to inform feasibility-oriented shortening for potential clinical use.