A fascinating recent report from the HealthLeaders about the supposed scads of errors being associated with HIT, as health systems transition to the age of electronic records. According to the report that features the results of a recent study by ECRI Institute’s Patient Safety Organization, some of the errors “are causing harm and in so many serious ways, providers are only now beginning to understand the scope.”
For example, during the 2012 study at 36 participating hospitals, computer programs truncated dosage fields leading to morphine-caused respiratory arrest; lab test and transplant surgery records sometimes didn’t talk to each other, leading to organ rejection and patient death; and an electronic systems’ misinterpretation of the time “midnight” meant an infant received antibiotics one dangerous day too late.
At the end of the nine-week study, there were 171 health information technology malfunctions and disconnects that caused or could have caused patient harm.
Essentially, the results of the study suggest that this is the tip of an iceberg related the HIT errors. Though I’m sure this isn’t the kind of iceberg that can sink a Titanic, it is one to be aware of.
We all know these errors happen more frequently than anyone would like to admit. This doesn’t excuse the fact that this is actually happening, but to claim that electronic systems are 100 percent accurate is and mitigate human error is foolish.
Two of my recent experiences related to this where in 2010 and 2011 when I was working in house as a public relations director for a major EHR vendor. In 2010, one of GE’s imaging systems inverted the images captured. So, for example, a patient’s lungs were actually reversed. The left was the right and the right was the left. In 2011, one of the major e-prescribers forwarded truncated codes for certain prescriptions to Walgreens pharmacies.
According the HealthLeaders, as health systems move from paper to electronic records, upgrade their systems from one version to the next or incorporate different vendors across various departments or service lines, mistakes are happening that are often go hidden for quite a long time.
“A forgotten lab test that wasn’t carried over and nursing and system manager work-arounds that fail are commonplace. Human error happens when systems depend on people to manually enter information from one system to another.”
One example of an error occurred when a health system did an upgrade of its health IT system, and down the line, realized a printout of a report omitted certain fields. An upgrade that caused the omission happened two years prior.
When changes are taking place to a system, organization leaders should spend time walking through three stages: planning, implementation and ongoing monitoring to review the workflow and processes.
The 171 events documented, break down like this (again, according to HealthLeaders):
- 53 percent involved a medication management system
- 25 percent involved a computerized order entry system
- 15 percent involved an electronic medication administration record
- 11 percent involved pharmacy systems
- 2 percent involved automated dispensing systems
- 17 percent were caused by clinical documentation systems
- 13 percent were caused by Lab information systems
- 9 percent were caused by computers not functioning
- 8 percent were caused by radiology or diagnostic imaging systems, including PACS
- 1 percent were caused by clinical decision support systems
According to ERCI, the group that conducted the survey, the errors are of two types:
- Those caused by humans and their interaction with electronic devices such as computers, such as when an operator fails to log off or retrieves the wrong patient record;
- Those caused by machines that move too slow, can’t communicate, malfunction and crash
The report blames 56 percent of the errors on computers and 46 percent on humans’ interface with those computers.
ECRI is currently evaluating a similar and much larger list of reports from many of the 800 hospitals that contract with ECRI.
Expanding the study across a larger population should be telling and provide greater accuracy to the actual situation at hand, but I’m sure no one will be surprised. This is an iceberg we’re talking about, and they tend to be quite large.