Using Internet Cloud Platform To Characterize Neurosurgical Patients in an Emergency Hospital Admission Setting: Retrospective Observational Study (Preprint)

Abstract

Neurosurgical patients are admitted to hospital via emergency department admissions is common; however, studies designed to describe their features are not available.

This study aims to investigate the characteristics of patients admitted to the neurosurgery department in an emergency manner by using medical big data integration and application platform (internet cloud platform).

We derived data from the internet cloud platform of West China Hospital, Sichuan University. The data of consecutive patients admitted to the department of neurosurgery as emergency admissions in a non-profit tertiary care university hospital was collected. Data on demographic information, clinical characteristics and outcomes were collected and evaluated through the platform. Patients were stratified into five disease groups (vascular disease, trauma, oncology, spine and others) according to their main diagnoses at the time of admission.

A total of 4,149 cases (median age 52 years, 54.5% male) were identified in this study. Vascular disease was the most common reason for emergency admission (73.5%). Significant differences were found among the five disease groups in sex (P<.001), age (P<.001), surgery (P<.001) and season (P=.009) but not in the length of stay (P=.784). Multivariate logistic regression analysis identified male sex, older age, short length of stay, surgery not performed and disease type (particularly trauma) as independently associated with in-hospital mortality.

By using internet cloud platform, we identified significant demographic and clinical differences among neurosurgical patients admitted to the neurosurgery department as an emergency. These findings may assist health care practitioners in shifting tasks and clinical resources toward those patients who may benefit from particular intervention approaches.