Multilevel Models Analysis On The Influencing Factors Of Chronic Fatigue Syndrome Among High School Teachers In Nanchang City

Objective: To understand the present prevalence and distribution characteristicsof chronic fatigue syndrome (CFS) in high school teachers in Nanchang city. To findthe influencing factors of CFS using multilevel model, so as to provide the basis orreference for prevention and treatment of CFS.Methods: Using stratified cluster sampling,5key and5ordinary high schoolswere chosen in Nanchang city. The questionnaire investigation was conducted amongthe teachers of these schools. Then the suspected patients were definitely diagnosedby doctors. The statistical methods were descriptive analysis, chi-square test, the gen-eral logistic analysis and multilevel models. The inspection level of alpha was0.05intwo sides.Results：(1) The rate of CFS among high school teachers in Nanchang city was13.4%(11.7%~15.1%). The rates of CFS among different marital status had significant dif-ferences showed by signal factor analyses. However, there were no significant differ-ences in different age groups, genders, education background and economic income.(2) Multilevel model showed that the school types, professional titles, the aver-age daily sleeping time, working pressure and neighborhood relations were the influ-ence factors of CFS.(3) Multivariate logistic regression analysis and multilevel model all showed thatthe influencing factors of CFS were the school types, the average daily sleep time,working pressure and neighborhood relations. The professional title was a significantvariate in the equation of multilevel model, but was not in the equation of logistic re-gression analysis.Conclusions:(1) The rate of CFS among high school teachers in Nanchang city was13.4%, sothat high school teachers were high risk populations of CFS. (2) The school types, professional titles, the average daily sleep time, workingpressure and neighborhood relations were the influencing factors of CFS.(3) Multilevel model was more reasonable than the general logistic regression inanalyzing the influence factors of CFS among high school teachers, so that couldovercome the shortage of the general logistic regression in considering hierarchy ofdata.