Targets and the emergency medical system--intended and unintended consequences

There is interest in health service reform and efficiencies; health service providers collect statistics, set targets and compare institutions. In January 2009, in Ireland, a national waiting time target of 6 h was set from registration in the emergency department (ED) to admission or discharge. The aim of this study was to assess the consequences of the introduction of this target on our institution and the Acute Medical Admission Unit.

All emergency medical admissions were tracked over 7 years and in-hospital mortality, length of stay and ED 'wait' numbers and times were summarized.

There were 43 471 admissions in 28 862 patients. In-hospital mortality for 2006-2008 averaged 5.9% [95% confidence interval (CI) 5.5-6.2%] compared with 4.8% (95% CI 4.6-5.1%) for 2009-2012 - a relative risk reduction of 18.3% (95% CI 11.5-24.5%) (P<0.001). The median length of stay was unaltered: 5.1 days (interquartile range 2.1-9.8) versus 5.0 days (interquartile range 2.0-9.5) (P=0.16). An ED 'first ward' allocation decreased six-fold with redistribution to the Acute Medical Admission Unit (two-fold increase) and the medical wards (four-fold increase). The time to on-call medical assessment decreased (time to team pre/post 4.5 vs. 4.2 h, P<0.001). However, calculations directly on the real-time log of arrival and first in-patient time showed a worsening of the position (time to ward pre/post 7.1 vs. 8.4 h, P<0.001).

Target setting may result in unintended consequences in other areas in addition to its stated goal. These unintentional consequences of targets should be borne in mind by those planning and instituting healthcare reform.