Objective To establish a health self-rating scale for undergraduates during the online learning period and to test its reliability and validity in medical students, so as to provide reference for self-health monitoring during online learning. Methods The health self-rating scale was developed through literature review, qualitative interviews and Delphi methods. The reliability and validity of the scale were tested in 969 medical college students who were taking online courses. Results A 32-item health self-rating scale was formed. The content validity index at item level (I-CVI) was 0.824-1.000, and the content validity index at scale level (S-CVI) was 0.928. Exploratory factor analysis showed that five common factors could be extracted, namely, somatic symptoms induced by online learning, psychological symptoms induced by online learning, psychological expectation of continuing online learning, role adaptation and social support of online learning, and changes in health status during online learning, which accounted for 72.946% of the total variance. The correlation coefficients between each dimension were 0.103-0.721 (P<0.01 for all), and the correlation coefficients between the total score of the scale and each dimension were 0.398-0.924 (P<0.01 for all). The Cronbach′s α coefficient of the overall scale was 0.939,the Guttman split-half coefficient was 0.733, and the test-retest reliability was 0.962. The Cronbach′s α coefficient of each dimension was between 0.817 and 0.957. The score of the scale was finally divided into four levels: normal 0-29 points, mild abnormality 30-56 points, moderate abnormality 57-87 points, and severe abnormality 88-112 points. Conclusion The health self-rating scale for undergraduates during the online learning period has good reliability and validity in medical students and can be used to evaluate the health level of medical students during online learning.The scale is suggested to be further verified and popularized in other student groups.