Validity and reliability test for assessing the bleeding factor in breast cancer wounds

Background: Breast cancer is the cancer with the highest incidence in women in the world. In Indonesia, breast cancer also ranks first in terms of number and cause of death. Bleeding is a common complication of advanced breast cancer. Identification of bleeding factors in breast cancer wounds is very important. Research instruments are needed to analyze various factors related to bleeding and their indicators Purpose: To determine the validity and reliability of the construct of factors related to bleeding in breast cancer wounds and their indicators. Method: A cross-sectional approach, specifically focusing on observational study. The study took place from January to May 2024, in Dr. Soedraso Hospital Pontianak. The independent variable was factors affecting breast cancer wound bleeding the dependent variables is assessment of breast cancer wound bleeding. Using purposive sampling and the Slovin formula, 60 respondents were selected. The validity and reliability tests in this study were conducted using the outer model evaluation in SEM-PLS with SmartPLS 4.0. Validity was assessed through convergent validity using the AVE value and discriminant validity with the Fornell-Larcker criterion. Reliability was evaluated based on outer loading values and composite reliability (Rho-C) to measure internal consistency. Results: the outer loading value, composite reliability (Rho-C), and average variance extracted (AVE) have met the validity and reliability test standards. The outer loading value of the indicators ranged from 0.628 to 1.000, with the highest value in the indicators of cancer stage, leukocytes, nutritional status (MNA), and bleeding at 1.000. Composite reliability for the variables of wound classification, dressing use and wound care, chemotherapy, therapy, and haematology ranged from 0.776 to 0.903, while the variables of cancer stage, infection, nutritional status, and bleeding each amounted to 1.000. AVE values ranged from 0.599 to 0.822, with the highest value of 1.000 for the variables of cancer stage, infection, nutritional status, and bleeding. Conclusion: This research conducted validity and reliability tests using the measurement model (outer model) evaluation in SEM-PLS. The assessment included outer loading, composite reliability (Rho-C), AVE, and the Fornell-Larcker criteria. The results confirmed that the research instrument was valid and reliable, making it suitable for analyzing bleeding factors in breast cancer wounds.