According to the Ya’an Statistical Year Book(2018), the permanent population of Ya 'an at the end of 2017 was about 1,538 million (Bai et al., 2018),and the prevalence of mental disorders in China published by the National Health Commission(NHC) in 2017 was 17.5% (Phillips et al., 2009),with an estimated tolerance error of 20%, the minimum sample size for the epidemiological survey of mental disorders in Ya 'an was about 6200 cases. Considering the small population size of some ethnic minorities, the sample size was expanded to 9000 cases. According to the proportion of permanent residents in urban and rural areas of the city was about 1:2.6, the sample size was divided into 2500 cases in urban areas and 6500 cases in rural areas.
Sampling method
Using a multi-stage stratified cluster random sampling method, both urban and rural areas of Ya 'an City were included in the sampling frame, and their demographic characteristics were basically consistent with the permanent population of Ya 'an area. In the first stage, taking towns (streets) as sampling units, a total of 153 towns (streets) from the city's two districts and six counties were listed and 33 towns (streets) were randomly selected. In the second stage, the village (neighborhood) was selected as the sampling unit, and the list of villages (neighborhood) under the jurisdiction of the selected township (street) was discharged and 27 villages (neighborhood) were randomly selected. In the third stage, in the selected villages (neighborhood committees), according to the list of household registration population provided by the local public security department, all the people who met the conditions were included in the survey.
Inclusion criteria: age ≥15 years old; Living in Ya 'an area for ≥6 months at the time of investigation; Experienced Lushan earthquake. Exclusion criteria: age < 15 years old; Those who refuse to participate in the survey, or who are not fully conscious (or mentally ill) and unable to cooperate with the data collection; Not personally experienced Lushan earthquake. Exclusion criteria: The contents of the questionnaire were unqualified, which affected the analysis of the results of this study. If the main sample could not be found for three consecutive times during the survey, it was defined as the dropped sample, and the replacement sample was selected as the first replacement sample in the framework of the same village (neighborhood committee) by matching the principle of gender and age (±2 years old).
Preparation and investigators training
All investigators were provided with the information of the project aims, procedures and detailed content. Those who were responsible for recruitment and assessment were trained for on-site communication skills and the measuring tools.
Measuring tools
Self-made basic information questionnaire: general demographic data such as gender, age, ethnicity, residence, marriage, employment, income, education level, substance abuse, physical health status, and distribution of disaster areas were collected.
General Health Questionnaire (GHQ-12): developed by Goldberg, it is a screening tool for general health level. The supplemented GHQ-12 has good sensitivity (Chen et al., 2023).There are 12 items with a total score of 0 -12, which can be summarized into 3 factors: lack of pleasure and sleep disturbance, social role and loss of confidence. Yang Ting-zhong et al. (Ting-zhong et al., 2003) confirmed that GHQ-12 can be used as a screening tool for the epidemiological investigation of mental disorders in China, and considered that 3/4 is the best cut-off value in the Chinese population. According to the total GHQ-12 score, the subjects were divided into high-risk group (GHQ≥4), Moderate-risk group (GHQ score of 2 or 3); Low-risk population (GHQ score of 0 or 1). The internal consistency is α = 0.91.
Neurosis Questionnaire: It is provided by World Health Organization (WHO), containing a total of 10 questions, and Shen Yucun and Professor Cooper added a total of 12 questions about neurasthenia and hysteria after consultation.Each question contains two parts, the first is the presence of symptoms, the score is 0-2;the second is the moderate or severe symptoms, the score is 1-2, the maximum score for each question is 2 points.The standard cutoff scores were used to define:≥2 as positive,<2 as negative. Questionnaire has good sensitivity and specificity (Yu et al., 1985).
Social Support Rating Scale (SSRS): It is designed by Xiao Shuiyuan et al on the basis of foreign scales and according to the actual situation in our country, which can comprehensively evaluate people's social support status (Shuiyuan X, 1994).The scale had 10 items, including three dimensions: objective support (3 items), subjective support (4 items) and the utilization of social support (3 items). The higher the total score and the score of each subscale, the better the degree of social support.
The MOS item short from health survey (SF-36):It is one of the most common standardized measurement tools for quality of life in the world, containing a total of 36 questions,8 health domains: physical functioning, role limitation due to physical health problems, bodily pain, general health, vitality, social functioning, role limitation due to emotional problems, and mental health. Each subscale was scored from 0-100. The above eight subscales can be further grouped into two categories: physical summary of health and mental summary of health. Higher scores indicate better functional status in that area and higher quality of life (Kubo et al., 2020).
In the second stage, the 5th edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-V) was used as the diagnostic criteria, and the Structured Clinical Examination for DSM-5 Axis Disorders-Research Edition (SCID-5-RV) was used as the diagnostic tool.
Investigation procedure: the investigators in this study were doctors and nurses engaged in psychiatry, and the field investigation time was from January to May 2019. Before the investigation, the project team contacted in advance by telephone to determine the investigation date and investigation cooperation, and guided by village cadres or primary health workers, 3 investigators (1 doctor and 2 nurses) carried out household investigation. All subjects with positive results in the high-risk and intermediate-risk neuroses screening table, and subjects with study number 1 in the low-risk group were diagnosed according to DSM-5 by attending psychiatrists who received unified training in SCID-5-RV.
Quality control: in order to ensure the standardization of the survey steps and the consistency of the survey tools, all the investigators were trained before the survey began, and the test was conducted after the training, and the consistency isα=0.875.The survey quality control group and quality control personnel were set up, and the completion of various scales was counted every day, and the accuracy and completeness of the scales were verified and signed for confirmation.Feedback the existing problems to the investigators in time, eliminate or verify and supplement the non-conforming or incomplete data.
Statistical analysis
EpiData3.1 software was used to input data, and SPSS 22.0 software was used for statistical processing of data results. Continuous variable data were presented as and categorical variable data were presented as numbers or percentages. Adjusted prevalence rates were calculated based on the current (last 12 months) and lifetime number of patients diagnosed with anxiety disorders in the three risk groups. Single factor analysis of qualitative data using chi-square test; After stepwise manual screening of independent variables, multivariate logistic regression was used to analyze the influencing factors of anxiety disorder. The test level α=0.05.
Ethical permission
The study was approved by the medical ethics committee of the Ya’an Fourth People’s Hospital on July 17, 2018 (No: 2/2018).