Modeling Patient Falls in Hospitals Based on Survival Analysis and Recurrent Events Analysis  

Oleh:

Kato, Shogo ; Fukumura, Fumio ; Tsuru, Satoko ; Iizuka, Yoshinori

Jenis: Article from Proceeding
Dalam koleksi: 12th ANQ Congress in Singapore, 5-8 Agustus 2014, page 1-11.
Topik: cox proportional hazard model; regression analysis; survival analysis; quality management system; safety management system
Fulltext: RL1-1.5-P0243.pdf (396.7KB)

Isi artikel

Medical accidents are generally classified into two types: accidents caused by healthcare staff while providing services, and accidents caused by patients. Among the latter, patient falls are a significant problem. The goal of our research here is to establish a methodology to prevent patient falls by identifying situations that are dangerous for each patient and determining concrete countermeasures to prevent dangerous situations based on assessment results. Kato et al. (2013) have evaluated each risk factor through Cox regression analysis by using about 1,000 actual case data at Iizuka Hospital. However, the data are insufficient for a detailed analysis. It was needed to arrange the data format, consulting models for repeated events. Furthermore, it was need more consideration on statistical model and covariates to be included in the model. In this paper, we aim to statistically model patient falls based on the Cox proportional hazard model as well as on recurrent event models. We first discuss the statistical model and covariates to be included in it. We then develop a multiple scoring system based on the results of the Cox regression analysis and a recurrent events analysis using data from an acute hospital from 2009 to 2011. Finally, we evaluate each scoring system by a Concordance Index

Opini Anda

Klik untuk menuliskan opini Anda tentang koleksi ini!

Kembali