Sunday, September 09, 2012

It Seems I Am The Bug Finder In Chief for the NEHRS. I Just Don’t Believe What A Mess This Is!

Tried the routine look up of my PCEHR at 11:49am on Sunday 9 Sep. 2012.
Logged in OK and got to the screen where you click to actually see the record.
-----

Welcome DAVID G MORE

-----
Here was the link:
(Pleasingly the link no longer works - Error 404)
Rather than going to my record I got this (In the corner of a white screen.):
-----

Unknown Relationship

The relationship to the eHealth Record could not be resolved.
If you continue to have problems please contact the Helpline on 1800 723 471 for further assistance.
-----
Here was the link.

This also does not work: The Oracle Access Manager (V 11g) says:

Error

System error. Please re-try your action. If you continue to get this error, please contact the Administrator.

I went back and tried my Personal Details this worked
-----

Your Personal Details

Information in this view is based on Medicare information held by the Department of Human Services (DHS). To update this information, please contact DHS on 132 011.
Note: Your healthcare professionals will be able to see your personal details except your address.
First Name: xxx
Last Name: xxx
Individual Health Identifier (IHI) Card Number: xxx
Date of Birth: xxx
Age: xxx
Sex: xxx
Address: xxx
-----
But when I came back from this then asking to go the NEHRS just does nothing at all. The system just sat there and did nothing. System broken again!
What to say other than to let people know this project is being run by a collection of incompetent nincompoops. Surely using the users to test a system that seems to be fatally flawed is just not good enough for a billion dollar system.
Walks off shaking head in despair.

David.

Tried again at 2:43pm. Guess what.

-----

Notification of planned outage

The Personally Controlled Electronic Health Record System is currently unavailable.

-----

This is found here:

http://ehealth.gov.au/internet/ehealth/publishing.nsf/content/availability#.UEweGFFAXak

Clearly an unplanned outage - has suddenly become 'planned' with no end time specified. I hope I didn't cause it. And if you believe the outage was planned you are a strong believer in flying piglets!

Some has suggested that the NEHRS has specific behaviours when I log on. I am beginning to suspect that may be true! <big grin>

Surely the time has come for an adult to be put in charge of this mess, take it off line and fix it before starting again after proper testing?

D.

Update 2: 5:31 pm Sep. 9, 2012

System seems to be back and very interestingly my medications are now updated to June 30, 2012. - only 2.4 months behind. More interesting is that a medication for my wife that was on a different script has now gone from the update - having been placed in my record in the earlier records.

This is just astonishing inaccuracy and rubbish.

They need to start from scratch and sort all this out!

We all know the system just crashed - so why just not resign and have adults fix the mess up?

D.


24 comments:

Bernard Robertson-Dunn said...

On 10/19/2011 12:13:00 PM, on this blog, I said...

"Are there any standards for exceptions and error handling?

It is one thing to define how a system is to work normally, it is far more difficult and time consuming to define what happens when the system goes wrong."

When you get an error like this from the system software:

"The Oracle Access Manager (V 11g) says: Error System error. Please re-try your action. If you continue to get this error, please contact the Administrator."

it suggests very strongly to me that the application developers have not implemented appropriate error or exception handling. If they had, they would have trapped the error and taken appropriate application level action. They obviously haven't. My guess is that the implementation has been too rushed, or they never even considered it important.

That is not a good sign. In fact it is a very bad sign.

Once again, I'll willingly change my mind, or my guess, given evidence.

Anonymous said...

Your medications get updated as they arrive at Medicare. Is your pharmacist perhaps tardy in claiming? The system can only display data that has been provided to it.

On the medications for your wife, were they on your record? If so, did you perhaps correct that with Medicare, or perhaps your pharmacist corrected it? If so, then the system will reflect that once it comes through.

In all cases, PCEHR displays the same data that Medicare has. As with any data source, it can have errors, the right process to address those errors is to contact Medicare or your pharmacist.

Dr David More MB PhD FACHI said...

@9/09/2012 06:45:00 PM

Sorry. Like all pharmacists his / her cashflow means quick submission.

The issue is that the data is just wrong - I have not been asked - there is no edit functionality on this feed and I have a life to lead rather than discussing nonsense with Medicare.

The system just does not work and is not reliable - despite what you say. Blaming the victim in rubbish policy.

David.

Bernard Robertson-Dunn said...

Anonymous said...

"As with any data source, it can have errors, the right process to address those errors is to contact Medicare or your pharmacist."

Where is this data correction capability documented? And what is the process if there is a disagreement/dispute?

Anonymous said...

Did you mean that you went into the future 12 months and it was no better, not something I would be surpised about, or did you mean 2012.

In fact I am a bit surprised that it would be even running in 2013 when I think about it. Back to the Tardis David

Dr David More MB PhD FACHI said...

Fixed,

David

Anonymous said...

I strongly agree with Bernard and David. Any system that gives you an error message such as "The Oracle Access Manager (V 11g) says: Error System error. Please re-try your action. If you continue to get this error, please contact the Administrator." has failed just about every web system design basic. A perfect advert for anyone wishing to know what is in use so they can more easily subvert it.

Fail, fail, fail.

Jim Cocks said...

Notice that this error is coming (by the look of it) from Oracle 11g (R2 I suspect). This is a fundamental DB error at the database level. Which says to me that the application is not communicating with the Database correctly - not good for a $467m system

Anonymous said...

As with all data in the system, there is the ability to remove documents or to set security controls on individual Medicare documents.

As with all documents, only the author can edit them, so the patient themselves cannot change the content, only remove the entire document. In this case, the author is considered Medicare, although ultimately Medicare only alter them if the pharmacist or Medicare alter the claim.

There are a number of controls in the system about who can edit what, you'd appreciate no doubt that it isn't desirable for a patient, for example, to edit a document that a clinician has submitted, particularly when that document is being asserted as correct by that clinician.

The correction mechanism for all documents is to either:
- request the source of that document to correct it, OR
- to remove that document from your record.

Dr David More MB PhD FACHI said...

So, unbeknownst to me - unless I log on daily to check - the information in my NEHRS is just incorrectly (potentially) updated at Medicare's whim.

Joke Joyce - really huge one!

David.

Anonymous said...

Actually, it is probably a blessing when the system is in outage, in terms of quality and safety. Makes a farce of the "medications" data sourced from the PBS. I also found mbs items in my record that were claims for my children- obviously a keying error when entering data for our Medicare card, but dangerous if a clinician acted on this data. We can only hope that the next phases with the Gp and hospital data has integrity and currency. Not a good start! Even for a baby system as they are fond of describing it.

Anonymous said...

This is what you get from a the $1m a day team, thanks! It is broken, unreliable and now we are defending it by blaming the users! What sort of useability testing has happened here?

The software integrity of this system is very low. We were told this system was working in Singapore. What did that mean, did anyone check? What did they do to it to break it?

We are about to pour more money in by paying GPs to use this thing. I wonder how many GPs will "Please retry your action" let alone "contact the Administrator". Hmm! Well thought out that one!

Oh and please don't say that will be a different interface!!

When will we stop and think sensibly about this! Either take it offline and think about it properly then fix it and test it properly, or just take it offline.

I think this is a just poor quality system, no taxpayer value here!

Anonymous said...

I am confused now. I would have thought once a document or data item was published in the PCEHR then it stayed there? After all that is why we were sold this particular expensive architecture, with centralised repositories, instead of the distributed HIE approach that many would have liked to see, where data stays local and is just viewed.

If the problem is Medicare changing its database then this puts up a new question. Is the medicare database, which is fit no doubt for its designed purpose, also fit as the primary feeder source for the PCEHR? Looks not. What are probably acceptable behaviours to Medicare for its purposes become inexplicable behaviours to a consumer looking at it.

Anonymous said...

I qualified first as pharmacist then studied medicine. I currently practice in both capacities to break the boredom of each domain. I see escripts at both ends of the spectrum and I ask myself why is government and NEHTA messing around like this instead of using the script exchanges which are already installed in almost every pharmacy and medical practice.

Anonymous said...

9/10/2012 09:47:00 AM why is government and NEHTA messing around like this instead of using the script exchanges which are already installed in almost every pharmacy and medical practice.

Good question. Maybe it's because they wish the script exchanges would go away and take David with them.

Anonymous said...

1. Medicare is not the primary feeder system. It is a conformant repository.

2. As with all conformant repositories, the data that PCEHR displays is a window into that repository. As some have noted, this is part of having a federated architecture, which is what PCEHR was designed to be. And as others here are fond of noting, this is their desired architecture. It is not clear to me what the concern is over this.

As with any document in the system, if the source determines that the document is incorrect, then they are able to amend it. A new version arrives, so audit to the previous version is retained. Or alternatively the document is removed, in which case audit of the removal is retained.

It seems to me that some are jumping at shadows here, and assuming that something is happening that there is no evidence of. Exactly what is it that is supposed to be happening in the Medicare data that people are concerned about?

Dr David More MB PhD FACHI said...

"Exactly what is it that is supposed to be happening in the Medicare data that people are concerned about?"

The concern is that even on the evidence of a really tiny sample there are some significant errors. Who knows how much of the data is current, reliable and correct?

David.


Anonymous said...

". As some have noted, this is part of having a federated architecture, which is what PCEHR was designed to be. And as others here are fond of noting, this is their desired architecture. It is not clear to me what the concern is over this."

Ah yes garbage in garbage out, not our fault, we just display it.

Ok, well how about I am concerned that the taxpayer has so far payed$150m for a PVC pipe and a flushing system then!

Paper anyone?

Anonymous said...

Does this mean that any clinician accessing the PCEHR for clinical purposes, should ensure that their system takes a snapshot of what they viewed through the windows to conformant repositories and the national repository - that way they can confirm what they saw, as opposed to what they might see another day. The source data may change, and the consumer may change what the clinician can see. One clinician may see a different view to another clinician, depending on the privacy controls exercised by the patient.

Anonymous said...

Does a clinician in a hospital take a snapshot of the data they see at a point in time, knowing that any of a myriad of other clinicians in that hospital (or hospital system) may change the data?

Does a GP in a large practice take a snapshot of the data they see on a patient's record, knowing that other GPs in that practice might change the records?

The PCEHR system retains audit of changes, including keeping prior versions of any document (no document is every amended, only a new document submitted), and history of any access level changes and document removals. If this level of audit isn't sufficient for the clinical decisions that a particular clinician is making, then they always have the option to take a copy. In general that doesn't appear to be the practice even in very large institutions.

Anonymous said...

You missed the point. A clinician viewing a pcehr record may see a different set of data to another clinician because a patient can control what is seen by whom. But a clinician cannot see ithe audit log for a parient's pcehr- only the patient can see this. In a hospital, all clinicians involved in a patients care will see the same view, and will if needed have access to the history of changes - this is the essence of an electronic medical record, and also the rules around paper records where they still exist.

Anonymous said...

All clinicians in a provider organisation will see the same data, same as they do today within a hospital.

Access controls are set by a consumer for a provider organisation, not individual clinicans.

The broader question here goes back to the nature of PCEHR, and the tension between the various interest groups. Patients want access to and control over their record. Clinicians generally believe they should have unfettered access to records.

The tradeoff was that the system could be a system that many patients chose not to use (and therefore had no data for those patients), but that clinicians had full access to and control over. Or a system that patient groups were happy to recommend using, and that gives extensive patient control, but that therefore has some limitations on access for some clinicians in some circumstances.

Is that perfect for all groups? No. Is that a compromise that allowed the different stakeholders to all get benefit? Yes. There was no political path to a system where we mandated that all patients used it, and gave them no control over the data in it.

Anonymous said...

This seemingly reasonable comment conceals a plethora of assumptions required to make it seem reasonable, but nonetheless betrays the flawed justification for the PCEHR in the 1st place.

Please spell out the assumed Benefits, the evidence that these Benefits are desirable and actively pursued by your defined set of stakeholders and the current evidence these "assumptions" have any basis in fact!

Unfortunately time will be our only guardian to expose the flaws and lies pushed forth to justify criminal misapplication of Tax Payers' funds to this "wet dream" of "someone's" intellectually bankrupt PCEHR baby.

Anonymous said...

"The tradeoff was that the system could be a system that many patients chose not to use (and therefore had no data for those patients), but that clinicians had full access to and control over. Or a system that patient groups were happy to recommend using, and that gives extensive patient control, but that therefore has some limitations on access for some clinicians in some circumstances."

For my money, sure, under any normal project methodology, a fair point to discuss and the arguments of someone that might have invested hard work at the doing end of this. I don't mean to be dismissive of the point and I feel for the argument and would discuss it had the stakeholder engagement methodology and the system uptake had been anything other than abysmal.

Both are a manifest indication that it has not been and will not be a success but a failure in terms of meeting stakeholders needs.

We could discuss the implementation as though there was some merit in it, but it still appears from evidence posted here that:

- The number of patients it is currently supporting is negligible.
- It is not at all intuitive to use with a UI that throws cryptic messages at users;
- The real clinical content feeds are openly questionable and of limited quality;
- These sorts of discussions didn't stop it being implemented whilst rebuffing advice given from the local industry, so why bother engaging now;
- The price of what has been delivered is a disgraceful gouge from the point of view of taxpayer value and fitness for use;
- If it wasn't for this forum's voice it would no doubt be celebrated as a great success by all involved, with much rewriting of history!

The MSIA and advocacy groups were ignored and openly rebuffed by those with all the power to implement this system. They thought they knew better and "picked their winners" so no wonder it isn't working!

For the money they have given out on our behalf, I for one as a taxpayer want to see some value, not weasel words about why it isn't working due to tradeoffs. Lets have the conversation about tradeoffs with stakeholders beforehand, or when they have got 200,000 patients on board. Lets not post-rationalise this thing and put more good money after bad without seeing a return on the initial taxpayer investment!

Someone needs to stand up and cop it sweet here, with an audit of why we spent over a hundred million dollars in twelve months for nothing more than a portal that didn't work. We continue to spend millions of dollars on support for it. I wonder how many patients are using it now?

Disappointed!