Reliability and validity of a pattern identification questionnaire for liver-qi stagnation infertility

• Published in: European journal of integrative medicine 2019-08, Vol.29, p.100922, Article 100922
• Main Authors: Lee, Ji-Yeon, Ahn, Soo-Yeon, Jun, Da-Hee, Ha, Ki-Tae, Yoo, Jeong-Eun
• Format: Article
• Language: English
• Subjects: Construct validity; Cutoff value; Liver-qi stagnation infertility; Pattern identification questionnaire; Reliability; Traditional Korean medicine
• ISSN: 1876-3820; 1876-3839
• DOI: 10.1016/j.eujim.2019.05.008

Traditional Korean medicine recognizes six types of female infertility. One of these types is liver-qi stagnation infertility. A pattern identification questionnaire for liver-qi stagnation infertility was first developed to objectively assess the diagnosis and treatment of liver-qi stagnation according to traditional Korean medicine. This study examined the reliability and validity of that questionnaire. Forty-two infertile women identically diagnosed by two liver-qi stagnation diagnostic experts were studied. The reliability of the questionnaire was measured using Cronbach’s α, and validity was examined using the McNemar test and kappa statistic to assess the agreement between the diagnoses by the two experts. Construct validity was tested using factor analysis. The cutoff value for the diagnosis of liver-qi stagnation was determined by the Youden index and concordance probability. The questionnaire had relatively high reliability, with a Cronbach’s α of 0.858. The kappa statistic for validity was 0.6383, indicating significant agreement between the two experts. Factor analysis to test construct validity revealed four major factors. The cutoff score for the diagnosis of liver-qi stagnation was 37, with a sensitivity of 82.35% and specificity of 92.00%. The questionnaire for liver-qi stagnation infertility may be appropriate for identifying and assessing liver-qi stagnation and has statistical reliability and validity. Follow-up studies with large samples should be performed to review potential problems and confirm statistical significance.