Establishment of nursing needs scale for elderly patients with urinary incontinence and its reliability and validity test

Xiangping LiuSilian Cai^*

• Department of Respiratory and Critical Care Medicine, Ruijin Hospital, School of Medicine, Shanghai Jiao Tong University, Shanghai, China

Objective: Urinary incontinence significantly impacts the quality of life of elderly patients, yet there is a lack of specific assessment tools for their nursing needs in the Chinese healthcare context. To develop a nursing needs scale for elderly patients with urinary incontinence and test its reliability and validity, aiming to provide a quantitative assessment tool for geriatric nursing practitioners and a scientific basis for quality-of-life interventions and standardized nursing for these patients, thus promoting the standardization of nursing practice.Methods: A comprehensive approach was adopted, including literature analysis, qualitative interviews, the Delphi method, pre-testing, and questionnaire surveys. Literature was retrieved from multiple databases and relevant websites to construct the initial scale framework and item pool. Purposive sampling was used to select 12 elderly patients with urinary incontinence (aged 60 years and above) and 10 medical staff for semi-structured qualitative interviews. Twenty-two experts participated in two rounds of Delphi consultations. Convenience sampling was applied to select 30 patients for pre-testing and 530 patients for the formal questionnaire survey. Content validity was evaluated using the Item-Content Validity Index (I-CVI) and Scale-Content Validity Index (S-CVI). Kaiser-Meyer-Olkin (KMO) test and Bartlett's test of sphericity were used to assess sampling adequacy. Factor loadings and total variance explained were calculated through exploratory factor analysis. Confirmatory factor analysis was conducted to validate the factor structure. The data were analyzed using SPSS 26.0 and AMOS 25.0 software to evaluate the scale's reliability and validity.Results: The initial scale had 48 items in 5 dimensions. After expert consultations and item screening, the pre-test version with 36 items was formed. Through exploratory factor analysis on 250 patients, 5 common factors were extracted, and one item was deleted, resulting in the final scale. Confirmatory factor analysis on 280 patients showed that the model fit well (χ²/df = 1.412, RMSEA = 0.037, SRMR = 0.042, GFI = 0.901, TLI = 0.942, CFI = 0.947, NFI = 0.915, IFI = 0.948). The scale had good reliability (Cronbach's α coefficient of the total scale was 0.901, split-half reliability was 0.865) and validity (content validity index S-CVI was 0.942).