Learning from our mistakes Using Near-miss reports to

Learning from our mistakes: Using Near-miss reports to improve patient safety and communication STFM Spring Conference April 2012 Steve Crane, MD Natascha Lautenschlager MD

MAHEC Hendersonville, NC Rural Training Track http: //www. mahec. net/resident/fhch. aspx Steve. crane@pardeehospital. org

Why this is important

What is a near-miss event? l A deviation from a planned or expected process of care that COULD have resulted in harm to the patient, but was averted before ANY harm was done. The original concept from aviation: Near-misses are when aircraft come within 1, 000 feet of each other vertically or three miles horizontally

Something happens that we didn’t intend, and. . . There is SOME potential for harm, but. . . NO HARM (even minor) to the patient

Not a near miss!

A Near Miss (can be complicated)

So what? l Most medical encounters take place in ambulatory settings, i. e. not in the hospital or ER. l We don’t know very much about near-miss events in practice: l l l they’re probably very common. Probably under-reported Near misses may represent low-risk opportunity to find and fix problems BEFORE patients are harmed.

Examples l Rx for amoxcillin sent to pharmacy; pharmacist calls to remind prescriber of reported penicillin allergy; Rx changed. l Lab report misfiled in another patient’s chart without review from the ordering provider; found by chance when other patient seen in clinic the following day Was this a deviation? Potential for harm? No patient harm?

Why study near-misses? l They are PREVENTABLE errors l They are accessible (as opposed to adverse events where a patient is harmed) l They numerous

What we did l Implemented a near-miss event reporting system in 7 primary care practices l Observed how practices used this information to improve patient safety l Designed and administered a point-of-care survey of patient and clinician attitudes and feelings around disclosure of near-miss errors.

Features of successful near miss reporting systems Anonymous l Keep it simple—point of care, quick (<2 minutes) l Practice leader buy-in l Engage all staff l Share results of reports and remediation l

The Near Miss Desktop Icon

What did we discover about near-miss events? l l They are common--770 events reported in a year. Many are potentially serious l EMR a significant source of errors (15% of total) l Knowledge gap only 2% of errors

Near Miss Events by Type (%)

Most common near miss events l Office process 47% (filing 50%) l Investigations 26% (reporting results 50%) l Meds/treatments 14% (ordering meds 60%) l Communication 8% (60% with patients)

Seriousness and potential for harm Error Type Seriousness % “Very likely” % “Very costly” Reporting results 72/100 38% 17% Dispensing meds 63/100 16% 11% Ordering meds 59/100 17% 6%

EMR-related errors l Significant portion of errors attributed to EMR l l l Ordering meds— 40% Filing— 22% Two of the top 5 causes of errors l l l Human error Taking time NOS

How did practices use information? l Reports frequently uncovered a previously unknown process error. l Serious errors usually fixed on the spot. l Practice leaders found the reporting process easy and the information useful

What practice leaders think about near-miss reporting l l “If everyone is aware of what is going on, it makes you more apt to change. ” (results of focus groups)

Attitudes about disclosure of near miss events l l Parallel patient/provider surveys 99 patients; 53 clinicians l l l Two scenarios l l l 132 approach/103 started/99 completed 45/75 participating in study; 8/39 community PCPs Both serious One where patient knows; other where they don’t Open/closed ended questions

What we found Expectation/Attitude Patients (%) Clinicians (%) Disclosure 95% 77% What happened 92% 81% Why 92% 55% How it will be fixed 96% 76^ Expect apology 82% 85% Lose confidence 38% 51% Less upset if told 96% 89%

Bottom line l l l Near-miss errors are common, and potentially serious. Staff are willing to report, and practices are willing to try fixing problems they know about. Patients want to know about serious errors that affect their care—physicians are less likely to disclose errors, and may not include the elements patients expect.

So what? l Error reporting should be part of our DNA. l We need to get better about having conversations with patients about errors. l There is a lot we still don’t know (particularly about how to fix these problems).

What do you think? What’s next? http: //www. mahec. net/resident/fhch. aspx Steve. Crane@pardeehospital. org