Rasch analysis of the pre-sleep arousal scale in patients with acute insomnia disorder.

BACKGROUND: Hyperarousal plays an essential role in the initiation and maintenance of insomnia, highlighting the need for a tool that measure the hyperarousal state during the early course of insomnia. Pre-sleep Arousal Scale (PSAS) is a self-report questionnaire to evaluate subjective pre-sleep arousal of insomnia. The aim of this study was to examine the psychometric properties of the PSAS among patients with acute insomnia disorder (AID) using Rasch analysis. METHODS: Totally 170 patients with AID from 31 public hospitals in China were recruited and completed the test. The psychometric properties of the PSAS were tested using Rasch analysis by Winsteps v5.4.1.0, including unidimensionality, local item independence, item fit, category diagnostics, reliability, item-person maps and differential item functioning (DIF) by age and gender. RESULTS: According to Rasch analysis, somatic and cognitive subscales were unidimensional and basically demonstrated good item-fit statistics. 4-point Likert scale may be more appropriate for PSAS. All items have a good reliability and separation. No gender and age bias were detected for the scale. However, the person arousal level is not well matched to item difficulty. CONCLUSION: The present study further reveals appropriate psychometric properties of the PSAS in patients with AID and provides suggestions for refinements and supplements to the PSAS.

Perceived social support and suicide ideation in Chinese rural left-behind children: A possible mediating role of depression.

OBJECTIVE: Psychologically distressed rural Chinese left-behind children (LBC) are at an increased risk of suicide. However, within this population, suicide and potential associated factors are severely under-discussed. This study primarily aims to explore the association between perceived social support and suicide ideation (SI), and to discuss the possible mediating role of depression in this association. METHOD: A population representative sampling survey was carried out with 2,898 LBC in southwest China. Relevant information was collected by means of face-to-face interviews. Binary and ordinal logistic regression models were employed to screen factors associated with perceived social support, depression and SI. Path analysis was applied to estimate the direct and indirect associations between social support and SI. RESULTS: Multivariate logistic regression models revealed that, among the four sources, perceived social support from relatives living together, friends, and parents, but not teachers, was inversely associated with SI in LBC. Moreover, depression and SI were strongly associated with each other: Compared with non-depressed LBC, depressed LBC were 6.76 (95%CI: 3.85, 11.87) times more likely to report prominent SI. Path analysis revealed that it is possible that the association between perceived social support and SI was almost entirely mediated by depression. CONCLUSIONS: Perceived social support from relatives living together, friends, and parents may be a protective factor against SI among rural Chinese LBC. Intervention strategies and measures that are targeted towards improving social support, especially cultivating closeness between LBC and their grandparents, could be effective in reducing SI in LBC.