This is an interesting report but as usual what is said is less important than what is omitted.
The report – such as it is – is found here:
An Evaluation of the Healthelink Electronic Health Record Pilot (Summary Report) (272K)
A comment made on the announcement in many ways said it all – to paraphrase “one or two clinician evaluations is worth more than a host of accountants”!
The random points I found interesting were (given the trial period reported is from March 2006 to September 2008):
First any thought of reporting costs of the pilot was blocked by the terms of reference.
Second there was no attempt reported to make any assessment of clinical impact.
Third, despite the time allowed for adoption, only 165 of 800 clinicians had actually used the system at all in the last 90 days.
Fourth, despite 2.5 years of trialling, and compulsory enrolment, a useful critical mass of patients and information is yet to be gathered.
Fifth, it seem pretty clear that while there are claims of technology success the time allocated has not permitted development of any really useful seamless integration of Healthelink and GP client systems. Clinician access and workflow problems remain major issues.
Sixth is appears the project team have, for whatever reason, failed to get GP software providers to co-operate and successfully integrate Healthelink into their client systems.
Seventh the benefits claimed are anecdotal and not evaluated in any hard way in terms of time saving, clinical improvements, patient satisfaction or anything else I can spot.
Eighth what has been done has been rather ‘National Standards Free’ and it is suggested this should be addressed sooner rather than later.
Ninth, despite the time available for improvement, the system is functionally poor and does not yet even provide a basic patient summary for each patient. (This really should be the core of any Shared EHR in my view as it is in Denmark, Scotland etc).
Tenth 35% of patients wanted an opt-in – not opt-out consent model – hardly a number to be ignored despite the clinicians being happy with opt-out. (Amazing that 66% of patients first found out they had been enrolled when the information pack turned up in the mail and then almost ½ did not read it fully!)
The scope of the pilot involved approximately 40,000 at the end of the trial with ½ being enrolled in the last 4-5 months.
Of the records created 95% of the Maitland records and 98% of the Western Sydney records were not accessed during the trial – hardly a heavy use!
Patients ignored internet access to their record in droves – only one in 492 looking at them in Maitland and one in 1078 in Western Sydney.
Overall it really seems to me this is one of those trials which should have been conducted as a learning experiment – given the length of time it ran – to get things really working. It seems that has not really happened and that the opportunity to really find out what might works has for now been lost.
Even the report we have should force a total re-think and a much improved approach before broader implementation is planned. The recommendations for improvement in Section 6.0 of the document need to be taken very much to heart.
The final point is, of course, if this pilot was actually a success, why are we not allowed to see the whole evaluation report - and if less than that why can't the most learnings be obtained through full disclosure?
Right now we can only say ‘Healthelink and NSW Health needs to try much harder’
David.
7 comments:
So much the Booz & Co Report which says:
"Healthelink – $40m program to roll out a Medical Record system for the state's eight area health services by 2009." !!
Having project buy in by the top 5 GP clinical system providers should be required before any tax payer funds are allocated to these pilots. GPs are happy to use IT functionality if it is integrated tightly within their existing clinical software product, but as soon as they have to "click outside the square", poor uptake is the only certainty.
Likewise, unless the shared/personal/individual health record scheme is integrated into a patient's existing electronic lifestyle (Facebook/Google/Netbank etc), single figure % interest is all that these pilots will ever achieve.
David,
Popular understanding is that $22 million was spent on this trial. With a NEW RFT for a statewide EHR due out in the new year (if it survives the NSW Government scalpel), NSW Health/Treasury confidence in the quality of the trial appears to be low; it would certainly seem to be $22m not well spent.
The Opt-in/Opt-out issue cannot be assessed without knowing the questions that were put to patients during the "interviews", the nature of the interview sample and the rigour that was applied to get an unbiased response. I simply do not believe the approval rate quoted for the opt-out model.
A previous commentator (why does everybody want to be anonymous?) said: "Having project buy in by the top 5 GP clinical system providers should be required before any tax payer funds are allocated to these pilots."
Initially I sympathised with this idea, because in theory at least it would make any new scheme or system more likely to be used by GPs. However, on reflection, I suggest that what would be better to propose that any new State or national programs such as Healthelink should be designed to conform to established standards for clinical information system used in general practice, so that developers of *all* current and future software for general practice will be able to link to or integrate with the new State or national system or program. We should avoid favouring or dealing only with vendor X or with vendors V, W, X, Y and Z to the neglect of other vendors currently in the market or who may enter the market.
All that we need now is some established standards for clinical information system used in general practice. Does anybody know of a taxpayer-funded national e-health organisation whose task it is to develop standards for clinical information systems or to have them developed?
Oliver,
The CCHIT in the USA might fit your bill (www.cchit.org).
David.
ps The anonymity seems to help as it means a lot of people who could be fired for what they say can tell the truth among other things.
D.
Hi all
What might be needed for GP software vendor buy-in is to offer them accreditation to a minimum standard required for interconnectivity to work, and let them market their product as "Accredited to AS xxxx" or some such. This still enables them to sell their product with value adds above the standard.
In regards to anonymity, I personally don't use it, but I can imagine some situations would require its use.
Thank you for this blog post, David. Your assessment of this 'evaluation' is very much in line with the views of the Australian Privacy Foundation. Our response paper is available at:
http://www.privacy.org.au/Papers/Healthelink-Roxon-0812.pdf
as sent to the Federal Minister, with copies also sent to various authorities in NSW.
I have also learned recently that Health Records and Information Privacy Regulation 2006
is to be repealed on 1 March 2009, which allowed the opt-out exception for Healthelink. Presumably this change will once again require opt-in or informed consent rather than the 'sign 'em up' approach in this "trial"???
I will be bookmarking your blog. Good work!
Jan Whitaker,
Australian Privacy Foundation
Post a Comment