Sunday, May 07, 2017

Guest Blog - My Health Record. The end of the beginning, or the beginning of the end?

If My Health Record stays opt-in or goes opt-out, the system will need a complete rethink.
Many people have indicated that they are in favour of a national health record system, however most reviews and recent reports indicate that there many fundamental problems with the My Health Record system as it currently exists. It is also worth noting that the government consistently promotes the claimed benefits and advantages of the system without mentioning or explaining the many risks and shortcomings inherent in the original design of the system. Here are a few:

  1.   My Health Record was supposed to reduce data fragmentation, however by adding yet another repository all it does is the opposite. The whole concept is fatally flawed; it will not and cannot achieve this goal.
  2. It’s a summary system. Patients and clinicians would get more value out of full access to all their health data. The UK and Sweden are both moving in this direction.(https://www.digitalhealth.net/2017/04/personal-health-records-accessible-on-nhs-uk-in-september-2017/  http://www.bmj.com/content/357/bmj.j2069).
  3. The access controls are weak, not as complete as the government claims (the evidence is on the government’s own website) and should be on a need-to-know basis.
  4.  As a secondary system and one to which many providers can upload data, there is no guarantee of, or even mechanisms to achieve, data accuracy, consistency and/or completeness. Many, if not most of the claimed benefits make the assumption that data will be accurate and complete. The opt-out trials did not explain to healthcare providers and/or patients how easily errors and omissions can occur and how difficult it is to rectify them. The subsequent assessment also assumed accurate and complete data; this error makes the conclusions suspect at least and totally unfounded at worst. It should be remembered that bad data can kill.
  5. It can never be more than a secondary system and was never intended to be highly-available 24/7. Making it so would incur huge additional development and running costs.
  6.  GP's haven't used it because it provides no value to them, just extra costs. Any attempt to make it compulsory for GPs to upload health data to My Health Record is likely to result in even less reliable data in the record.
  7.  As a source of data for research it has very limited value compared with full health record data. 
  8. It is a privacy nightmare and the legislation that enables opt-out allows for data to be collected without patient consent.
  9. That the government has full access to all the data in My Health Record is a huge disincentive for patients.
  10. My Health Record was designed to be opt-in and as such assumes certain relationships between a patient, their nominated service provider (there can only be one at any one time and is the only person permitted to upload a Shared Health Summary) who may or may not be their GP. Patients who have no regular GP or who see multiple GPs will have major problems keeping their health record current, accurate and consistent. If the system is made opt-out the size of this latter group will increase significantly. 
  11.  Health care and the systems that support it will need constant and rapid updating to accommodate changes in data acquisition and analysis driven by personalised and precision medicine. This will be a very expensive on-going and endless commitment for the government. Selling such a system to the private sector is likely to meet strong opposition.

It is strongly suggested that the only realistic, workable strategy is to scrap My Health Record and re‑evaluate the whole approach to a national health record system. Any future initiative should aim for patient access to full health data on a clinical need to know basis and no government access or control of the data.
Anything short of that is going to get massive pushback from the activist crowd concerned about recent government IT problems involving the census, data linking, CentreLink and its robo-debt system, mandatory data retention and the perceived persecution of vocal dissenters.
To put it in a very few words: What has been done so far with My Health Record and many earlier initiatives have been total and un-mitigated disasters and a waste of well over $2b. To proceed with My Health Record would be reckless, futile and incur huge costs remedying fundamental and basic design flaws.
There are other and better ways of achieving a national heath record capability for all Australians based on individual circumstances and need, not a single, one-size-fits-all, government owned and run repository of secondary data. Data fragmentation will not be reduced by creating another system. Data fragmentation will only be achieved through fewer systems and/or better integration and interoperability.

Advice to ADHA:
There would seem to be two alternatives:
  1. Proceed with My Health Record, make it opt-out, spend a significant amount of time and money making it appropriate for opt-out use, another $1b making it 24/7, spend another $500million/year forever maintaining and updating the system, fight the activists and advocates, bribe and/or coerce GPs into wasting consultation time uploading secondary data into a privacy risky system and run the risk of a very public and un-necessary failure, especially if there is a privacy breach or problem because of inaccurate or incomplete data.
  2. Re-think the My Health Record strategy in the light of experience. Take a totally different approach based upon distributed GP and Hospital systems, facilitated, but not owned by the government. An approach similar to that in the UK and Sweden. This would be very cheap for the government, adds zero costs to GP time, is much more privacy friendly and if it fails, then it is not the government's fault. Learn the lessons of failure and do it properly this time.
Option 1 is on a hiding to nothing. Option 2 is much more acceptable and achievable. Unless of course it is the government's intention to grab as much health data as possible for non-health reasons. In which case option 1 is the only alternative.
There is no logical or defensible reason why government should be a participant in owning or managing patient health data. It does need some information because of the role it plays in health funding, but it has no part to play in the delivery of health care. It could have a role in facilitating data exchange, setting and enforcing standards and in ensuring patient safety. Its actions should enhance patient privacy and the quality of health care, not threaten it.

Disclosure:
I am chair of the health committee and a board member of the Australian Privacy Foundation. We are an advocacy group, not activists.
We are not against health records, we support the idea totally. However, the value of the system must be balanced by the privacy risk. My Health Record has delivered no value, threatens patient safety and the quality of health care, and is a major privacy risk, a risk which will only increase exponentially if it is made opt-out.
Dr Bernard Robertson-Dunn
Email: Bernard.Robertson-Dunn@privacy.org.au

12 comments:

Paul Fitzgerald said...

Is there anybody listening?

Anonymous said...

Excellent blog Bernard, is anyone listening? we can but hope. Sadly though no one is talking. The Opt out seems to be a very quiet affair compared to the fan fare of PCEHR days, the budget big ticket items will overshadow this, which in the grand scheme of things has failed to make much of headline, if it in fact even reaches a mention in a footnote. Perhaps that is a reflection of the ADHA teams performance, was it not there to reignite eHealth?

If the MyHR is funded for more than just operations then it will be down to other business cases calling it out as a dependency, such as Homecare.

The big question is, will this be good or bad debt?

john scott said...

Bernard, an excellent summary of the strategic position.

This comes at a time when there are a growing number of innovative and successful clinical reforms that embrace digital pathways in various ways.

As these come more into the light, the contrast will be even more striking.

Just how lasting is the ignorance and arrogance of the leadership of Commonwealth Health?--to borrow a phrase from the Secretary of the Prime Minister's Department.

Anonymous said...

Friends, be not deceived. The Review of the Opt-Out trials was kept out of the public eye form months until it could be used to support the case to COAG for more funding.

Only when the Department knew COAG had given the green light for a national roll-out (as recommended in the Report - not in stages but all at once) did the Department then decide to release the Report. The Budget will show how right I am.

Anonymous said...

John, could not argue. As a subscriber to what is proving a useful health app that talks directly to my GP's system I am concerned the GovHR system is going to put me at risk, either by providing out of date information in an emergency, cause issues with my GP system where neither data source can be trusted, or that through this system my technology provider cannot stay in business. I would be far more comfortable if the Government simply oversaw the health information market to ensure they obey the law on my behalf, I have never found any one public or private being good at self regulation.

The Pharmacy Guild getting into bed with the MyHR is extremely concerning.

Anonymous said...

Dear Dr More, the articles in PulseIT are worth noting.
1. Guild, CHF upbeat about opt-out My Health Record, and
2. Blog: Opt-out is now the only option, and
3. My Health Record to go opt-out in 2019, support for automatic records “almost universal” and
4. Majority of practices have met the ePIP upload target: DoH .

Is it possible you have been flogging a dead horse? In the face of all this evidence should you not now move to embrace and support the My Health Record? With evidence in abundance from the Evaluation of the Trials shouldn’t you and your commentators be adding your voices to those of the AMA, the Guild, CHF and others in support of rolling out the My Health Record nationally without delay.

Dr David More MB PhD FACHI said...

As I have personally told Tim Kelsey I believe the myHR project is misguided, privacy invasive and fatally flawed architecturally. I am not likely to have my views changed by article in Pulse+IT. The AMA is less than enthusiastic, the RACGP even less so and the others are Government paid lackeys who are not independent.

The AMA has a lot of reservations - see this note from yesterday.

https://ama.com.au/ausmed/my-health-record

I will leave others to comment more fully.

David

Anonymous said...

There are many important viewpoints and considerations to make, none more important than people, technology has a place, but is not healthcare - https://motherboard.vice.com/en_us/article/whats-digitization-doing-to-health-care

Perhaps certain organisations need to be more honest about their intentions. Computers are also very black in white and operate according to there designed purpose, a calculators purpose is not that of a typewriter

Bernard Robertson-Dunn said...

IMHO, there are only three important questions re MyHR:

1. Is is useful?
2. Is it safe?
3. Is it worth it?

Answers:

1. As far as we know it has never been used to make healthcare in Australia either more effective or more efficient.

2. The quality of the data in the system is doubtful (the government says it needs to be corroborated by the patient - as if they would understand the medical niceties). It can never be relied upon to be either accurate or up-to-date. The access controls are pathetic and are not what the government claims.

3. Given the answers to 1 and 2, I can't see any other response than No, it's not worth $2b and counting.

IMHO, none of the bodies referenced in Pulse+IT know enough about the details of MyHR to make informed comments. They believe what the government says. Unfortunately what the government says in its headlines is at odds with the information on its own website.

If you read the Privacy Policy (all 10,000 odd words of it), the section For Healthcare Providers and the legislation, you get a clearer and more comprehensive view of MyHR. Very little, if any of which was explained to those participating in the opt-out trials.

Andrew McIntyre said...

It seems that some people feel that MyEHR will succeed because the minister says so, so just accept it. Reminds me of this quote from an investigation into NASAs management failings by Richard Feynman.

"For a successful technology," Feynman concluded, "reality must take precedence over public relations, for nature cannot be fooled."

john scott said...

Andrew, we can also add the following from Alan McDonald the rocket maker's engineer who stood steadfast against the decision to launch the ill-fated Challenger.

Even when NASA overrode his recommendation against the launch, MacDonald said to Mulloy, the NASA manager who was pressuring him to change his recommendation, “If something goes wrong with this flight, I wouldn’t want to stand up in front of a board of enquiry and say that I went ahead and told them to go ahead and fly this thing outside what it was qualified to.” MacDonald later reported this conversation verbatim to the incredulous President’s Commission investigating the Challenger disaster.

Bernard Robertson-Dunn said...

Look up the backfire effect:

When Corrections Fail: The persistence of political misperceptions

http://www.dartmouth.edu/~nyhan/nyhan-reifler.pdf

"Results indicate that corrections frequently fail to reduce misperceptions among the targeted ideological group. We also document several instances of a “backfire effect” in which corrections actually increase misperceptions among the group in question."