Wednesday, August 10, 2011

Medico Legal Issues, EHRs and the PCEHR. It Is Pretty Messy and Complicated.

The following appeared a little while ago.

How Will EHRs Impact Medical Liability?

Widespread use of clinical decision support may establish new legal standards of care

By David Raths

Moving from paper to electronic data is expected to have widespread patient safety benefits, but could the use of electronic health records actually heighten the liability risks clinicians and hospitals face? And does the existence of the technology change what is legally expected of physicians?

At least in the initial phase of implementation, the answer to both questions may be yes, according to Michelle Mello, JD, PhD, a professor of law and public health in the Harvard School of Public Health.

Mello is the author of more than 100 articles and book chapters on the medical malpractice system, medical errors and patient safety, research ethics, clinical ethics, and other topics.

Speaking at the World Congress 3rd Annual Leadership Summit on Mobile Health in Cambridge, Mass., on July 28, she noted that as systems mature, there will be clear opportunities to avoid errors, but providers should recognize the potential for new kinds of errors to develop.

“There’s the potential for the failure of clinicians to use the technology consistently,” she said. “There may be bugs, defective applications and system crashes.”

In the transition to an all-electronic environment, there can be documentation gaps in the interface between paper and electronic records.

Before courts have weighed in on new legal standards of care, newer modes of treatment may be problematic. For example, she said, the use of e-mail has the potential to improve communication between providers and patients and improve patient satisfaction. But the use of e-mail with patients may allow for arguments that clinicians are being less thorough. “Not seeing a patient in person might be considered negligent in some circumstances,” Mello said. Also, patients may develop expectations about what defines a reasonable response time to e-mail correspondence.

Clinical decision support (CDS) can transform and improve care and can catch errors before they harm patients.

But Mello said a significant change in malpractice law could involve CDS features. Widespread use of CDS may establish new standards of care. The existence of EHR systems means there is more documentation of clinical decisions and activity and more discoverable evidence, including metadata such as time stamps. Departures from CDS guidelines could bolster a plaintiff’s case, she added.

“Systems document when providers choose to deviate from the algorithms in CDS, and the onus will be on clinicians to legally justify why they did,” Mello said. Several chief medical information officers in the audience expressed their concerns about this topic, and said they have struggled to determine which alerts have the most value in their system and which are just having an “information overload” effect on clinicians.

More here:

This article adds a whole new dimension to the responsibility of clinicians in using any form of electronic health record.

We already know the Australian Medical Association has concerns about liability associated with the PCEHR. See here:

It is time, in thinking about all this to go back to basics. Essentially a patient medical record has three main functions.

The first is to record in reasonable detail the clinician’s assessment and diagnosis(es) of a patient along with the supporting evidence (test results, images etc.). This record is then able to be referred to in the future to ensure follow-on assessments and therapy to do not overlook previous and possibly unresolved issues. It also allows the clinician to build a picture of the patient over time and to ensure proactive prevention and screening are undertaken.

The second function is for the record to facilitate transfer of care or to facilitate collaborative care by members of a practice (e.g. GP and Nurse Assistant or GP and another GP).

The third and less important function of the clinical record is as a record of what assessments and investigations were done by the practitioner to confirm that the care provided has been in a competent way and to a standard that would be expected from peer practitioners (this means that in a specialist area there are higher standards expected from the specialist than a GP).

To ensure legal acceptability, should there be major problems, the each entry should be signed and dated when completed. The record should never be altered - but it is perfectly acceptable to add a later comment or information as long as it is also signed and dated.

Moving to the electronic domain it is ideal that at least the same, if not more complete information is recorded at each encounter, and that the system be designed to record relevant user, date and time stamps to provide an audit trail of the activity within the record. The electronic form also needs to be able to produce a ‘dump’ of the record onto paper or into a human readable file for storage and review.

To me there are two issues that critically need to be addressed - besides the obvious one of the clinical quality of what is recorded electronically. I am also assuming appropriate information integrity and security from change and revelation which seems obvious Also, of course, there is the expectation of system IT quality and ‘bug freeness’ - which can’t be assumed.

Less obvious in the electronic world is to easily be able to discover the provenance of a record. Who wrote the record, when it was written, who are they and what qualifications do they have - as well as under what circumstances was the record created. It is well known in clinical circles that a complete and contemporaneous clinical note is the best defence there is against any medical negligence charge and as such a practitioner needs to be able to fully rely on the completeness and integrity of the electronic record - should the worst come about.

Second there is also the issue raised in the lead article about how clinical decision support (CDS) is implemented, what it is based on, its reliability and so on. No good following CDS if it will mislead or only provide half the picture.

Making all this work in provider systems is a very big ask. In the shared record environment like the PCEHR is may just be too much, as the AMA seems to think.

It is worth pointing out that these issues are not new and that lots of thought has been given to most of these issues a within the GEHR project in Europe which then led to the openEHR work among other efforts. There is a lot of valuable material found on their web site:

This page especially on the early work and GEHR those interested will find interesting in following up.

(Note: this blog studiously ignores many of the non medico-legal aspects of GEHR and openEHR which are important but not easily covered in this blog format)


No comments: