RESUMO
Objective: To develop a three-dimensional body image stimuli suitable for middle-aged people in China, and verify the validity and reliability of the body image stimuli. Method: According to China and World Health Organization body mass index classification standards of adults, a set of three-dimensional body image stimuli of Chinese middle-aged males and females with different body size was developed by using 3D Studio Max and Adobe Photoshop CC based on the literature and expert consultation method. Forty-two 45- and 59-year-old middle-aged people in Zhengzhou City, Henan Province were recruited to verify the three-dimensional body image stimuli. Through questionnaire survey and physical examination, the coincidence between the selected body type and the actual body type was tested; the body composition was measured by dual-energy absorptiometry (DXA), and the structure validity of the image was tested; the body size satisfaction was investigated by the body image stimuli and the standard questionnaire, and the empirical validity of the image was tested. The repeated survey was conducted 14 days after the initial survey, and three experts were invited to score the current somatotype of the subjects to test the test-retest reliability and inter-rater reliability of the body image stimuli. Pearson, Spearman, Kendall correlation and Kappa consistency analysis were used to evaluate the validity and reliability of the body image stimuli. Results: The average age of 42 subjects was 52.7 years old, including 13 males and 29 females. A group of three-dimensional body image stimuli of middle-aged men and women were developed, and each group included 8 images. 73.8% of the subjects chose the body size consistent with the actual body type, and the weighted Kappa coefficient was 0.755 (P<0.01). The selected somatotype was positively correlated with body weight and body composition indexes such as fat content, and the Pearson correlation coefficient of construct validity was 0.623-0.717 (P<0.05). The results of the two surveys were positively correlated, and the Spearman correlation coefficient of test-retest reliability was 0.784-0.821 (P<0.05). The scores of the three experts on the current somatotype of the subjects were positively correlated, and the Kendall correlation coefficient of inter-rater reliability was 0.818-0.878 (P<0.05). Conclusion: The development of principle and reference basis of three-dimensional body image stimuli of middle-aged people is reliable, and the validity and reliability of the body image stimuli are good.