CIBER en Epidemiología y Salud Pública (CIBERESP), Barcelona, Spain.
J Adolesc Health. 2011 Aug;49(2):199-205. doi: 10.1016/j.jadohealth.2010.11.260. Epub 2011 Feb 11.
To assess the contribution of life events (LEs) on psychosomatic complaints in adolescents/youths taking into account a set of socioeconomic variables.
We tested a conceptual model implemented with structural equation modeling on longitudinal data from a representative sample of adolescents/youths and parents. Psychosomatic complaints were measured by the Health Behaviour in School-aged Children scale and hypothesized to be affected by: (a) contextual factors at distal level: financial resources, home life and social support (KIDSCREEN), and parent baseline mental health (SF-12); (b) triggering factors: LEs (Coddington Life Events Scales, with two typologies: desirability and familiarity); (c) intermediate factors: same as distal level but measured at follow-up; (d) immediate cause: mental health at proximal level (Strengths and Difficulties Questionnaire at baseline and follow-up); and (e) gender.
The structural model yielded a good fit (Comparative Fit Index = .95, Tucker-Lewis Index = .93, Root Mean Square Error = .04). Boys showed more psychosomatic complaints than girls (β = .40, p < .05). Girls reported experiencing more LEs (p < .05). Only undesirable LEs showed a significant direct negative effect on psychosomatic complaints, which became nonsignificant when mediated by home life and mental health. Undesirable LEs had a remaining indirect effects on psychosomatic complaints (βindirect = -.10, p < .05) via Home Life and Mental health, which were protective factors (β = .41 and β = -.15, p < .05).
The experience of undesirable LEs increases the probability of psychosomatic complaints, but the final effect would be determined by previous levels of home life and mental health stability.
评估生活事件（LEs）对青少年/青年身心症状的影响，同时考虑一系列社会经济变量。
我们使用结构方程模型对具有代表性的青少年/青年及其父母的纵向数据进行了概念模型测试。身心症状通过《儿童健康行为问卷》进行测量，并假设其受到以下因素的影响：（a）远端背景因素：经济资源、家庭生活和社会支持（KIDSCREEN），以及父母基线心理健康（SF-12）；（b）触发因素：LEs（Coddington 生活事件量表，具有两种类型：期望性和熟悉性）；（c）中间因素：与远端因素相同，但在随访时进行测量；（d）直接原因：近端心理健康（基线和随访时的《长处和困难问卷》）；和（e）性别。
结构模型拟合良好（比较拟合指数=.95，Tucker-Lewis 指数=.93，均方根误差=.04）。男孩比女孩表现出更多的身心症状（β=.40，p<.05）。女孩报告经历了更多的 LEs（p<.05）。只有不期望的 LEs 对身心症状有显著的直接负向影响，而通过家庭生活和心理健康的中介作用后，其影响变得不显著。不期望的 LEs 通过家庭生活和心理健康对身心症状仍具有间接影响（β间接= -.10，p<.05），这两者是保护因素（β=.41 和 β= -.15，p<.05）。
不期望的 LEs 的经历增加了身心症状的可能性，但最终的影响将取决于家庭生活和心理健康的先前水平。
