Resilience in rural left-behind middle school students in Yunyang county of the Three Gorges area in China: a prospective cohort study

BACKGROUND The number of left-behind children in China is gradually increasing. This study aimed to understand the mental health status and changes in resilience of rural, left-behind middle school students in Yunyang County of the Three Gorges area in China. METHODS A prospective cohort study, including two follow-up surveys, with a frequency of once every 6 months was conducted among middle school students in Yunyang County. A self-designed questionnaire was used to obtain socio-demographic factors of participants, the Mental Health Test (MHT) scale was used to assess their mental health statuses, and resilience levels were collected using Resilience Youth Development Module (RYDM) scale at baseline and at the first and second follow-up investigations. RESULTS Of the 1401 students who completed the baseline survey, 1322 students were eligible for the cohort study, of whom 1160 were investigated in the first follow-up survey. Ultimately, 1101 students completed the 1-year cohort. The detection rate of mental health problems for middle school students in the rural Three Gorges area was 5.64 %, and there was no significant difference between left-behind students (LBS) and non-left-behind students (NLBS) (χ (2) = 1.056, P = 0.304). The detection rates of medium resilience rose gradually (Z = 4.185, P = 0.000), while that of high resilience declined gradually (Z = 4.192, P = 0.000) in the baseline, first and second follow-up investigations. There was no significant difference between LBS and NLBS in resilience level (P > 0.05). The average RYDM scores were 2.990, 2.926, 2.904 among LBS in the baseline, first and second follow-up investigation, respectively, and the effect of time on the average RYDM scores was significant (F = 14.873, P = 0.000). The average MHT scores in LBS were 41.54, 39.79, 38.84 in the baseline, first and second follow-up investigations, respectively, and the detection rates of students who had psychological problems increased gradually (Z = 4.651, P = 0.000). The simple correlation coefficients between the RYDM and MHT scores were-0.227,-0.158, and-0.204 in the baseline, first and second follow-up survey, respectively (P = 0.000). CONCLUSIONS The detection rate of mental health problems for middle school students in the rural Three Gorges area is relatively low and most of the LBS in area have medium or high resilience. The mental health status of LBS positively correlated with their resilience levels. Resilience declined gradually as time went on, but further studies with longer follow-up durations are needed to confirm the variation of resilience.