Lag time in an incident reporting system at a university hospital in Japan
- Masahiro Hirose1,2,
- Scott E Regenbogen1,3,
- Stuart Lipsitz4,
- Yuichi Imanaka5,
- Tatsuro Ishizaki5,
- Miho Sekimoto5,
- Eun-Hwan Oh5,
- Atul A Gawande1
- 1Harvard School of Public Health, Department of Health Policy and Management, Boston, Massachusetts, USA
- 2Patient Safety Division, Kyoto University Hospital, Kyoto, Japan
- 3Department of Surgery, Massachusetts General Hospital, Boston, Massachusetts, USA
- 4Brigham and Women’s Hospital, Boston, Massachusetts, USA
- 5Department of Healthcare Economics and Quality Management, School of Public Health, Kyoto University Graduate School of Medicine, Kyoto, Japan
- Correspondence to: Dr M Hirose Department of Healthcare Economics, School of Public Health, Kyoto University Graduate School of Medicine, Yoshida Konoe-Chou, Sakyou-Ku, Kyoto 606-8501, Japan; mhirose{at}kuhp.kyoto-u.ac.jp
- Accepted 19 November 2006
Abstract
Background: Delays and underreporting limit the success of hospital incident reporting systems, but little is known about the causes or implications of delayed reporting.
Setting and methods: The authors examined 6880 incident reports filed by physicians and nurses for three years at a national university hospital in Japan and evaluated the lag time between each incident and the submission of a report.
Results: Although physicians and nurses reported nearly equal numbers of events resulting in major injury (32 v 31), physicians reported far fewer minor incidents (430 v 6387) and far fewer incidents overall (462 v 6418). In univariate analyses, lag time was significantly longer for physicians than nurses (3.79 v 2.20 days; p<0.001). In multivariate analysis, physicians had adjusted reporting lag time 75% longer than nurses (p<0.001) and lag time for major injuries was 18% shorter than for minor injuries (p = 0.011). Adjusted lag time in 2002 and 2004 were 34% longer than in 2003 (p<0.001).
Conclusions: Physicians report fewer incidents than nurses and take longer to report them. Quantitative evaluation of lag time may facilitate improvements in incident reporting systems by distinguishing institutional obstacles to physician reporting from physicians’ lesser willingness to report.
