Noohi Sima, Kalantari Siavash, Hasanvandi Saba, Elikaei Mahdi
Associate Professor of Psychiatry, Behavioral Sciences Research Center, Baqiyatallah University of Medical Sciences, Tehran, Iran.
Freelance researcher, Tehran, Iran.
Psychiatr Q. 2020 Jun;91(2):273-287. doi: 10.1007/s11126-019-09699-0.
Considering the limited resources for providing inpatient services, identification of the factors influencing length of stay (LOS) is of great importance. The current study is a retrospective chart review which was planned to investigate the determinants of LOS in two gender-specific psychiatric wards within Baqiatallah Hospital (BQH) located in Tehran. The observation period was between March 21, 2011 and March 19, 2016. 3203 patients were recruited in terms of inclusion and exclusion criteria. Next, required data on 25 explanatory variables were extracted from their case-files. Descriptive measures were used for analysis and Independent Samples T-test, one-way ANOVA, Pearson's correlation coefficient and Bonferroni's post-hoc test for inferential analysis. Lastly, a multiple linear regression model was run to determine significant predictors of psychiatric LOS. Variables significantly correlated with patients' LOS included gender, age, employment status, marital status, number of divorces, disability rate, discharge diagnosis, physical comorbidity, number of previous hospitalizations, suicide ideation, number of suicide attempts, history of assault, tobacco consumption, a history of narcotic drug abuse and number of ECT sessions. Through the analysis of multiple linear regression, it came to light that significant predictors of LOS in the final model could account for 37.9% of the variance in LOS. From the findings of current study, it can be inferred that clinical aspects as well as treatment procedures have major effects on LOS. Although the factors examined here could not explain an acceptable variance in LOS, the results are useful for the treatment team when they want to devise a care plan or give discharge to a patient.
考虑到提供住院服务的资源有限，识别影响住院时间（LOS）的因素至关重要。本研究是一项回顾性病历审查，旨在调查位于德黑兰的巴基塔勒医院（BQH）两个针对性别的精神科病房中住院时间的决定因素。观察期为2011年3月21日至2016年3月19日。根据纳入和排除标准招募了3203名患者。接下来，从他们的病历中提取了25个解释变量的所需数据。描述性测量用于分析，独立样本t检验、单因素方差分析、Pearson相关系数和Bonferroni事后检验用于推断分析。最后，运行多元线性回归模型以确定精神科住院时间的显著预测因素。与患者住院时间显著相关的变量包括性别、年龄、就业状况、婚姻状况、离婚次数、残疾率、出院诊断、躯体合并症、既往住院次数、自杀意念、自杀未遂次数、攻击史、烟草消费、麻醉药品滥用史和ECT治疗次数。通过多元线性回归分析发现，最终模型中住院时间的显著预测因素可解释住院时间方差的37.9%。从本研究的结果可以推断，临床方面以及治疗程序对住院时间有重大影响。尽管此处研究的因素无法解释住院时间的可接受方差，但这些结果对于治疗团队制定护理计划或让患者出院时是有用的。
