To explore the characteristics and influencing factors of family resilience among Chinese PLA Army personnel.

A total of 8 094 military personnel subjected with cluster sampling were surveyed with family resilience scale (FRS) and self-designed questionnaire for basic information on military families. Independent sample t test, one-way analysis of variance and step-up regression analysis were employed to analyze the family resilience.

① The results of demographic variable analysis showed that there were significant differences in family resilience among the personnel with different marital status, from one-child family or not, birth in rural/urban areas, personnel category, age and education level (P < 0.05).② Family factors analysis indicated that different family structure, different family relationships, family rearing styles, recent separation from family members, and frequency of contacting family members had certain effects on family resilience (P < 0.05).③ When family relationship, family rearing styles, frequency of contacting family members, birth place, recent separation from family, personnel category, from one-child family, and education level entered the regression equation, the resilience level of army personnel' families was increased by 8.197 for every 1 unit increase in family relationship, and the level of family resilience was increased by 4.688 for every 1 unit increase in family contact frequency (F=202.168, P < 0.05).

The overall level of family resilience of army personnel is good. The risk factors for family resilience are recent separation from their families for ≥1 year, officers, not from only-children family, and the protective factors are harmonious relationship with their families, democratic family rearing style, contacting with their families weekly, urban birth, and bachelor degree or above.

To explore the causal relationship between family resilience and mental health in military personnel population.

A total of 204 military personnel were recruited from an army unit stationed in Western China with cluster convenience sampling. Family Resilience Scale (FRS) and Symptom Checklist 90 (SCL-90) were used to survey them twice, in an interval of 4 months. Amos 26.0 was applied to construct a cross-lag model and analyze the data.

After controlling mental symptoms from the first survey, family resilience in the first measure significantly predicted mental symptoms in the second measure (β=-0.14, P < 0.05). After controlling for family resilience from the first survey, mental symptoms in the first measure significantly predicted family resilience in the second measure (β=-0.13, P < 0.05).

The relationship between family resilience and mental health is mutually causal in military personnel, and one predicts the other one. Our findings highlight the key dimensions of the relationship between the two.