The purpose of collecting reports of near misses in a patient safety reporting system is to identify weaknesses in a healthcare delivery system before a patient is harmed. The analysis of near-miss reports, roughly comparable to “Incidents” in the Pennsylvania Patient Safety Reporting System (PA-PSRS), can be as valuable, or more valuable, than analyses of adverse events, or “Serious Events,” that harm patients. A single near-miss report involving delay in cardiopulmonary resuscitation based on a misinterpretation of the meaning of a colored wristband has led to an international campaign to standardize the meaning of wristband colors.1 Near-miss reports provide useful information not present in adverse-event reports—namely, the action that prevented a medical error from harming the patient. Comparisons of the root-cause analyses of wrong-site surgery and near misses in which the potential wrong-site error was corrected before the patient was harmed has revealed processes that catch errors before they reach the patient (see the article “Quarterly Update on the Preventing Wrong-Site Surgery Project” in this Advisory issue).
If your institution is not collecting and analyzing reports of near misses, or “Incidents” under Pennsylvania Mcare Act 13, you are missing valuable information that could make your healthcare system safer. Your competitors may not be missing this opportunity to deliver quality care more reliably. If you are a chief executive or board member of a healthcare facility, an intelligent question to ask your patient safety officer, risk manager, and legal counsel is “What is the number of near-miss (or Incident) reports in our facility per 1,000 patient days?”
PA-PSRS clinical staff reviewed all patient safety reports submitted for the calendar year 2007. We calculated the number of Incident reports per 1,000 patient days to compare hospitals. We then grouped the hospitals by types: acute care hospitals, behavioral health hospitals, children’s hospitals, critical access community hospitals, long-term acute care hospitals, and rehabilitation hospitals. We identified the average number of Incident reports per 1,000 patient days for each hospital group (see Table).
|Hospital Type||Incident (Near-Miss) Reports/1,000 Patient Days|
|Acute Care ||24|
|Behavioral Health ||2|
|Critical Access Community ||42|
|Long-Term Acute Care ||40|
We noticed that some hospitals reported fewer than 10% of the average for their group. This means the average hospitals in their group gets 10 times the information about weaknesses in their systems as these low-reporting hospitals.
- Of the 150 acute care hospitals, 13 had fewer than 2.7 Incident reports per 1,000 patient days, with 4 of them reporting fewer than 0.27 (1% of the average for the others) and another 2 reporting none.
- Of the 11 behavioral health hospitals, 1 reported fewer than 10% of the average for the others.
- Of the 7 children’s hospitals, 2 reported fewer than 5% of the average for the others.
- Of the 13 critical access community hospitals, 3 reported fewer than 1% of the average for the others.
- Of the 21 long-term acute care hospitals, 1 submitted less than 10% of the average for the others.
- Of the 16 rehabilitation hospitals, 1 reported fewer than 1% of the average of the others.
Hospitals that are not capturing near-miss, or Incident, events are hurting their ability to identify and correct problems before they harm patients. Hospitals with 10, 20, and 100 times more information are going to learn ways to improve their systems much faster. A wise leader will ask: “What information about patient safety are we not collecting?” If the answer is that the average hospital of your type is collecting many more reports than you are, you need to improve your collection of near-miss reports.
- Pennsylvania Patient Safety Reporting System. Use of color-coded patient wristbands creates unnecessary risk. Pa Patient Saf Advis [online]. 2005 Dec 14 [cited 2008 Nov 24]. Available from Internet: http://patientsafetyauthority.org/ADVISORIES/AdvisoryLibrary/2005/dec14_2(suppl2)/Pages/home.aspx.