Wednesday, December 05, 2012

This Australian Article on Health IT Safety Makes Some Good Points But One Part Worries Me.

This appeared over the weekend.

More tests vital for health IT

Date December 2, 2012
Category Opinion

Enrico Coiera and Farah Magrabi

A doctor calls up the drop-down menu on her electronic prescribing system, looking for the heart drug digoxin. The 225 options are listed in counter-intuitive alphabetical order and she clicks on the wrong dose. Her patient is given four times the amount he needs.
Another healthcare worker finds the computer's font hard to read and accidentally gives a patient 10 times the correct dose of epinephrine (adrenalin). The patient dies.
These are just two of the many serious events that we at the University of NSW Centre for Health Informatics have unearthed while analysing accidents and mishaps involving health information technology reported to the US Food and Drug Administration from January 2008 to July 2010.
The recent accidental deletion of 10-year-old Ezekiel Howard's electronically stored heart scans, along with two months' worth of similar scans on other patients' hearts at Nepean Hospital, is another example of an IT incident that has the very real potential to harm patients.
There's no formal requirement to report errors such as these - unlike adverse patient safety events from medication or medical negligence. But the few studies that have been done into errors associated with computerised health information systems are pointing to a disturbing fact: as much as technology can improve the health system, it can also have deadly side effects.
IT is transforming doctors' surgeries, pharmacies and hospitals. Over the next 10 years, more IT will be deployed in health systems worldwide than in their entire history.
You've probably already seen the start of it. GPs don't write prescriptions any more. They print them out using an electronic system, which then files away information on your medication history and your medical history, to build a growing database. Indeed, you may even have entered your own history into the new Personally Controlled Electronic Health Record, which the government launched in July.
There is no doubt technology will contribute much to healthcare - better record-keeping, better communication in a complex system, and better information about patients' medical histories. It will be evermore crucial to rein in costs and create efficiencies as governments worldwide cope with ageing populations, a shortage of healthcare workers and an increasing workload due to obesity and chronic disease - and do so with fewer resources.
But here's the rub. If we were introducing a new drug or surgical procedure, we'd be proving it worked in large clinical trials and running exhaustive tests to get it registered by a national regulatory authority. For health IT, there is no such regulation.
.....
Enrico Coiera is a professor and Farah Magrabi is a senior research fellow at the University of NSW Centre for Health Informatics.
The full article is found here:
These three paragraphs worry me a lot, especially the third one:
“IT is transforming doctors' surgeries, pharmacies and hospitals. Over the next 10 years, more IT will be deployed in health systems worldwide than in their entire history.
You've probably already seen the start of it. GPs don't write prescriptions any more. They print them out using an electronic system, which then files away information on your medication history and your medical history, to build a growing database. Indeed, you may even have entered your own history into the new Personally Controlled Electronic Health Record, which the government launched in July.
There is no doubt technology will contribute much to healthcare - better record-keeping, better communication in a complex system, and better information about patients' medical histories. It will be evermore crucial to rein in costs and create efficiencies as governments worldwide cope with ageing populations, a shortage of healthcare workers and an increasing workload due to obesity and chronic disease - and do so with fewer resources.”
My view is that while there is absolutely no doubt we need the appropriate regulation to ensure Health IT (or e-Health as DoHA and NEHTA want to call it) is both safe and effective that we also need a few other boxes ticked.
Among these are that we need the appropriate leadership and governance frameworks for the introduction of the technology and that benefits are properly assessed, tested and then proven to have been delivered through rigorous evaluation.
I would suggest that the benefits case for the NEHRS / PCEHR program is very weak, has never been subjected to independent scrutiny and is basically a fanciful fiction.
It seems to me the authors need to have gone a little further than they did to push past the mentality being pushed by the Government that if you build e-health ‘they will come’ and that it will really work safely and actually deliver the benefits claimed.
All those views are on pretty rocky ground as far as I am concerned.
On a different track and from the same source.

Surveillance shows potential in detecting HIT system failures

November 26, 2012 | By Susan D. Hall
An Australian study finds potential value in applying a syndromic surveillance system to health IT systems to detect early system failures.
Such surveillance typically is used in public health to monitor the spread of infectious diseases. The system was used in research at the University of New South Wales in Sydney to monitor four factors in a tertiary hospital laboratory: total number of records being created, the number of records with missing results, average serum potassium results, and total duplicated tests on a patient.
The researchers, led by Dr. Mei-Sing Ong, wanted to detect HIT system failures causing: data loss at the record level, data loss at the field level, erroneous data, and unintended duplication of data, according to a paper published at the Journal of the American Medical Informatics Association. Statistical models were used to detect system failures using simulated outages lasting 24 hours, with error rates from 1 percent to 35 percent.
More here with links:
Keep the work coming guys (and gals).
David.

6 comments:

Anonymous said...

Hi David,

The second article seems to provide a promising approach to address the issues of human errors in using HIT systems. Mandating the inclusion of such early-warning features for HIT systems seems to be a good idea.

However, such approach would not be of any use to reduce the risk of technical faults. A rather simple case of a technical fault is that a HIT is not available.

What would happen to a GP practice or a hospital in the case of system failure on a scale that we could recently witness at airports when booking systems fail? Stripping processes bare and reducing costs wherever possible by introducing HIT systems does come with a heightened risk profile resulting from reliance on technical systems.

Coincidentally, the latest such airline booking system outage was caused by a hosting company that is a subsidiary of Accenture who are running the PCEHR...

Anonymous said...

Its OK - we are probably all safer if the PCEHR is down.

Anonymous said...

"A doctor calls ... Her patient is given four times the amount he needs."
Good eMM systems have features that alert based on the height and weight, and also on maximum dose lparameters configurable in the system. Also, pharmacists are pretty good at reviewing medications orders in hospitals and other settings, and the eMM system can be set to ensure that a review is performed before the dose can be administered. Sounds more like a work process error than just an IT error.

"Another healthcare worker finds the computer's font hard to read and accidentally gives a patient 10 times the correct dose of epinephrine (adrenalin). The patient dies."
And good eMM systems can be configured to have another nurse/clinician sign off before the dose is administered - depending on system configuration. You just have to buy the right system!

Bernard Robertson-Dunn said...

Anon said: "The second article seems to provide a promising approach to address the issues of human errors in using HIT systems. Mandating the inclusion of such early-warning features for HIT systems seems to be a good idea."

It seems a strange way to go about developing eHealth systems - try different approaches and solutions discarding those that don't work, keeping those that do.

Of course, the approaches and solutions that don't work would result in people suffering and dying, but hey, we would be making progress.

That approach might work in some industries and on a small scale in laboratories under carefully controlled conditions, but in the context of national health?

Would this be an acceptable approach in the transport industry? the food industry?

IMHO, somebody is living in cloud cuckoo land if they think this would be acceptable.

I prefer the approach suggested by Enrico Coiera and Farah Magrabi - make sure a system works properly before deploying it nationally.

But that requires understanding everything about a system in order to test it. I think that in the case of the PCEHR, that's what is missing - understanding what the PCEHR is supposed to be doing, and the consequences of doing it.

Terry Hannan said...

Okay where does one start on this topic? We have e-health systems in place that store > 8 millions patients, have CDS recall times measured in "blink" times. All patient data is stored forever and have been designed to have all this data and information secured so it is not lost. another feature of these systems is that 'every keystroke under a given password/ID is recorded for ever and NO data - new or old is lost. How long have these systems been in place? > 30 years! Some people are not learning from history.
On another topic of the tragic cases reported like the girl with the golf ball injury or the lost records they make great press and are TRAGIC. What about all the others harmed by our lack of effective CDS/CPOE/e-health tools? These do not make headlines yets are just as tragic.

Anonymous said...

I honestly do not understand how electronic data can be deleted on a patient....after the patient has passed away and the family are seeking answers as to why. Seems to me as though it is not an IT problem, but a DR problem, not tending to the patient correctly! Then trying to cover their tracks. Yes there needs to be a chance in this neglectful system. I just hope families that suffer because of such neglect will see justice.