Prevalence and predictors of high preoperative anxiety among adults undergoing elective surgery in Eastern Nepal: a multicenter cross-sectional study

Background: Preoperative anxiety (PA) is a common yet under-assessed psychological state that affects surgical outcomes and patient well-being. This study aimed to determine the prevalence of high PA and its associated factors in adult patients undergoing elective surgery in Eastern Nepal. Methods: A multi-center analytical cross-sectional study was conducted at two referral hospitals in eastern Nepal. Patients completed the Amsterdam Perioperative Anxiety and Information Scale (APAIS) in the preoperative holding area, and an attending anesthesiologist provided a clinician-rated anxiety scale (CRS). High anxiety was defined as an APAIS or CRS score ≥11. Demographics and clinical variables were recorded. Descriptive statistics summarized the sample, and the prevalence of high anxiety was calculated with 95% confidence intervals. The chi-square test and multivariable logistic regression were used to identify associations and predictors of high anxiety and categorical variables. Internal reliability (Cronbach’s alpha) and correlation were assessed for both scales. Results: Of the 512 patients included in the study, the mean age was 42.2 ± 14.8 years, and 59.8% were female. High preoperative anxiety was identified in 11.1% of patients using the APAIS and in 14.3% using the CRS. A high need for information was reported in 6.8% (APAIS) and 7.8% (CRS) of participants. Multivariable analysis revealed that younger age, female sex, gynecological surgery, public hospital setting, previous anesthesia exposure, and high information need were independently associated with higher anxiety. The strongest predictor was high information need (odds ratio = 12.9). APAIS and CRS displayed significant agreement (κ = 0.754), with CRS exhibiting strong diagnostic precision. Conclusion: In this Eastern Nepal cohort, the prevalence of high preoperative anxiety was significantly lower compared to many reports in other settings. Tailored interventions are needed, especially for young patients and those with higher information requirements. High concordance between patient self-reports and clinician ratings suggests both methods reliably detect high PA.