Abstract
Background: Analyses of patient falls are often using traditional ways to explore factors of patient falls. Many efforts were not made in examining data dimension and then failed to made inferences about factors of patient falls associated with days of the week and hours of the day in a hospital. Using Rasch model to explore them is required. Methods: We used the Rasch rating scale model to analyse data of inpatient falls from July 2005 to June 2010, which were collected from 35 wards of three different hospitals, examined unidimensionality. Differentiate item functioning (DIF) was detected in comparison between groups. Factors associated days of the week and hours of the day to inpatient fall counts are investigated to help accomplish large improvements on a small number of key areas giving an alarm to station nurses. Results: We found that (1) there were two stages of the time data extracted by parallel analysis; (2) DIF was not found in stage II of the time unidimensional data; (3) factors associated days of the week and hours of the day regarding inpatient fall counts were in existence on Saturday and at 9 o'clock, respectively. Conclusions: Dimension checking and modeldata-fit using parallel analysis and Rasch analysis are required to overcome the drawbacks of traditional ways to explore factors of patient falls. Factors associated days of the week and hours of the day can help pay more attentions on Saturday and at 9 o'clock giving an alarm to station nurses.
Original language | English |
---|---|
Pages (from-to) | 1395-1403 |
Number of pages | 9 |
Journal | HealthMED |
Volume | 5 |
Issue number | 6 |
Publication status | Published - 2011 |
All Science Journal Classification (ASJC) codes
- General Medicine