Construction of a pregnancy exercise evaluation scale based on the modified Delphi method

Abstract

Background: With in-depth research on physical exercise and health promotion during pregnancy, the scientific monitoring and evaluation of physical activity during pregnancy have gained increasing importance,with research in this area-becoming more in-depth. Objective: This study aims to synthesize existing information to construct an evaluation index system for physical exercise behavior among pregnant women in China, and to develop a corresponding assessment scale. Methods: Based on literature review and preliminary clinical surveys, this study employed the modified Delphi method to conduct two rounds of questionnaire consultations with 20 experts in related fields. Data were analyzed to calculate the experts' positive coefficient, authority coefficient (Cr), coordination coefficient of variation (CV), and Kendall's coefficient of concordance. The weight of each indicator was determined using the product method. Results: The positive coefficient for both rounds of expert consultation was 100%. The first-round authority coefficients were Cs = 0.804, Ca = 0.953, and Cr = 0.878, while the second-round coefficients were Cs = 0.820, Ca = 0.966, and Cr = 0.893. The coordination degree W for the first round was 0.331 (χ² = 763.552, P < 0.001), and for the second round, it was 0.437 (χ² = 1532.916, P < 0.001). Ultimately, 4 first-level indicators (individual circumstances of pregnant women, health risk assessment, physical exercise behavior and dietary intake, and health outcomes of physical exercise during pregnancy) and 22 second-level indicators (e.g., psychological state, physiological condition, educational background, income, intrinsic motivation, family support, external social environment, built environment, selfefficacy evaluation, medical history, health status, exercise type, exercise duration, exercise frequency, exercise intensity, symptoms requiring cessation of exercise, safety precautions, healthy lifestyle behaviors, postpartum laboratory outcomes, neonatal laboratory outcomes, sleep quality, and anxiety levels) were selected. The scale demonstrated good reliability with a Cronbach's α of 0.955. Conclusion: The developed assessment scale includes 4 first-level and 22 second-level indicators, with statistical data conforming to a normal distribution. This scale provides practical guidance for evaluating physical exercise behavior during pregnancy and serves as a reference for developing a comprehensive, scientific, and operable assessment system for pregnant women's physical exercise behavior.