An Office of Inspector General report in 2010 recommended internal hospital reporting of all harms—whether considered a complication, a preventable harm, or a harm caused by system failures or errors (all-cause harm).
A new study published by the Joint Commission details how the Adventist Health System Patient Safety Organization in Altamonte Springs, Fla., developed an automated all-cause harm trigger system to enable the identification of current patients who may have experienced harm or may be at risk for harm.
The study reports a pilot test at two Adventist Health System hospitals of a clinical decision support system with 41 algorithms. Nurse reviewers analyzed the electronic health records of current patients with positive triggers to determine if harm had occurred—a process averaging approximately five minutes per trigger.
Furthermore, if harm was identified, it was classified as hospital-acquired or outside-acquired and was grouped into one of five harm categories and assigned a severity level. Nurse reviewers also had the ability to identify patients with potential harm and alert an Intervention Coordinator to evaluate the patient with the goal of limiting or preventing harm.
After the system was implemented, combined data from the two hospitals during an 11-month period indicated:
- A total of 2,696 harms were acquired, of which almost one-third were acquired outside the hospital.
- Hypoglycemia or low blood glucose was the most frequently identified harm.
- A nurse reviewer was able to analyze 20 records in 1.5 hours using the automated review process compared to 6.5 hours using the previous manual review process
The automated harm trigger system, which can serve as a model for other healthcare organizations, also provided the Adventist Health System Patient Safety Organization with the ability to identify patterns of harm as they evolved, providing hospital quality departments with the opportunity to respond proactively before the patient left the hospital by providing awareness, education and intervention training as needed.
Potential harm flagging has an inherent flaw. First it implies that the treaters need an extra push to treat effectively. If the flag does result in extra effort which in fact then prevents harm, then the flag will have been in one sense wrong because no harm occurred. Repeated wrong flags will result in a distrust of the system. There is some story about a boy and a wolf applicable here. If the system is adjusted to remove flags when no harm occurred then the original trend of harms may reoccur. This becomes cyclical.
If you are a patient you want to be there when the flags are first introduced (or reintroduced) and the staff raises their game accordingly. You don’t want to be there when the flags stop triggering extra effort.