Wednesday, August 10, 2016

Observations On A New Hospital EMR At RNSH From The Point Of View Of A Curious Patient!

Note: All these notes are based on discussions with a range of staff at various levels from Consultants to Ward Porters – overall probably chatted with 10-15 different people.

Was good to see pretty strong privacy focus – to the extent of taking some fast talking to get to be able to access my record!

All this is E&OE as I may or may not have accurately grasped what I had been told. (I was a might sick for a good part of the time!)

Key Points.

Initial Implementation was in A&E and have managed to become paperless with clinical charting, patient records, results (image integrated with text etc.). Implementation in the wards is quite recent (months)

Screens are impressively clear and high resolution – and speed is really good.

System works by selecting ward census then patient and then has lots of tabs down left side to move into different functionality.

I was easily able to look up my new results with no training at all so it works pretty intuitively!

The COWs are a feature. Residents and registrars glide them around and make notes and place orders as they see patients. Battery powered and fully secured with scrub-able illuminated keyboards and full size high-res screen on really secure trollies. (COW = Computer On WheelS)

WiFi for the computer system is everywhere and fast!

There appear to be secure Wi-Fi points for clinicals, byod, mob-xray, and data. As well an open guest account I assume I was not meant to use!

Acceptance of the system is broadly aged based and seems also very much related to the level of typing skills.

Was interesting to note placing a ‘request on the system’ was almost seen as having done what was requested and needed. The workload communication was widely used.


Some staff worried about privacy of the large screen census boards on the wards.

Senior staff can find navigation (menus can be rather deep) and system freezes frustrating.

Was a bit sad to see observations which were gathered by superb and very smart trollies being manually typed into the system rather than being interfaced!


Where possible specific tasks for senior staff might be set up as scripts.

It is important to minimise system unresponsiveness as this can frustrate staff.

Organisation of free text information is still a challenge. Reports are fine and findable – progress notes seem to be a different story – suffer rather from the PCEHR pile of notes issue.

It was not clear just how useful the search functions were.


All the Clinical Ward PCs run an enterprise level version of Win 7 – but I did spot the odd Win XP screen in radiology.

Each PC is named and has a unique IP address and while many are wired at desks – the COWs all run wi-fi in the Hospital.

The number of terminals / COWS seems nicely over-provisioned – there is access everywhere that seem reasonable.


Log on is username / password to the relevant domain.

Once logged on the username does not expire but password does pretty quickly.

(user name is 8 or 10 digit staff number)

Home Nursing Service – Remote Visit Support:

Using a laptop the initial admission to the APAC Service was done on line to another copy of Cerner Millennium – over a remote WiFi link.

The laptop provided access to all information held in my record within the hospital – and worked well.


This implementation is clearly a very complex EMR with very rich functionality that is really working pretty well indeed. I have often wondered would I ever see at least some of the potential I knew was possible actually delivered but this is not bad at all! It works, is capturing rich information and has essentially got rid of paper notes – except at present for medication charts.

What is now in place is clearly a key part of the nervous system of the Hospital.

Once the more advance information management issues are solved there is the basis of a really useful and important system here that will in all likelihood make a very positive difference to care. This is all much better than I dared hope!

Well done to all those involved in getting to this point!



Bernard Robertson-Dunn said...


Good to hear you are fully awake and compos mentis.

Did anyone ever ask if you had a myhealthrecord?

If you have one, did it make any difference?

If you haven't got one, would it have made any difference?

Dr David More MB PhD FACHI said...

No ones asked if I had one and it would not have helped much. My 2 page paper summary of my history worked a treat!


Bernard Robertson-Dunn said...


How many $billions would that solution cost?

Oh course, it does mean that the government can't get at your personal data. IMHO, they can get statistical info from service providers without invading your privacy.

Have you filled in your census response?

Dr David More MB PhD FACHI said...

Census failed as I tried to submit....

Error Messages

Your request could not be completed as a problem was encountered. If you continue to receive this message, please call the Census Inquiry Service on 1300 214 531. [19:43:49, status -1, code 102]

Your request could not be completed as a problem was encountered. If you continue to receive this message, please call the Census Inquiry Service on 1300 214 531. [19:50:46, status -1, code 101]

A problem was encountered with your session. If you continue to receive this message, please call the Census Inquiry Service on 1300 214 531. [19:51:59, status -1, code 103]



Bernard Robertson-Dunn said...

I've been drawing a distinction between the Census, who are trying to keep things very secure and private (I'll let you judge if they are succeeding) and the Department of Health, who are trying to make your health data easier to get at.

Not only that, but the census is only every five years whereas your health data is constantly being updated.

And then there's this failure of the website (which doesn't appear to have a good explanation).

Furthermore, the MyHR system is not 24*7 (they have scheduled downtime), is not High Availability (they have a DR site but are in a single data centre - which is a no-no for good HA) and the data is by no means real-time.

So what good would it be a in a national emergency, a bio-terrorism attack or a health scare? Never mind resisting a DDoS.

The census is hosted by Softlayer who specialise in this sort of thing. And they have stated that a DDoS is highly unlikely. More likely that it's just undersized by ABS. It never occurred to ABS that so many people would sit down after dinner and try and fill in their census.

Back to comparisons. ABS has stuffed up and they only do it once every five years, they know when it will happen and how many people will need to access their system.

What are the chances that Accenture, working to requirements defined by Health can do better? Given the uncertainties of emergencies and the deficiencies in the data in the myhealthrecords?

They've spent $billion so far. IMHO, to re-write the software to make it realtime and get the technology up to a reliable 24*7 - HA standard would cost another $2billion.

And then it probably wouldn't work because GPs would be so busy putting data in, they wouldn't have time to care for patients.

On that cheery note, I'll let you get to your hospital dinner.

Get better soon David, we need you and this blog working well.

Peter said...

I have been wondering about access to these sort of systems. One trick used in many retail outlets - where staff move about a bit and use different terminals all the time - is a wrist band which is detected by the computer when it is within a certain range. That is - hands on the keyboard or mouse
Hence, move away from the machine and it locks out, move close and it registers the user from the wrist-band and restores their specific session.
I don't know if this would work in a hospital though. It would depend on the nature of the band, how easy it is to keep clean, and practicalities about people wearing them.

Bernard Robertson-Dunn said...

Suppose two doctors and a nurse are looking at the screen.

Whose ID are you going to record?

How do you know if the nurse is actually looking at the screen and not standing close but just talking with one of the doctors?

How do you know if one of the doctors reads stuff from the screen to his six students?

It's not as easy as it first appears.

And it all applies to MyHR as well as hospital eHR/eMR systems.

Bernard Robertson-Dunn said...

Does anyone know if the IHI is based on the SLK581 cluster?

This article might be suggesting that it is and that there are some maajor problems with it.

"The SLK581 has been registered for use for health, housing, and early childhood records. It is also designed so that the SLK can be issued once, and then follow Jane Smith around for the rest of her life. Because it's relatively unique and nonsensical, it can be used to combine records without giving away Jane Smith's name – except it’s not secure at all."

Peter said...

I don't see how SLK581 could be used for IHI. But then I can't see how it would be a useful identifier at all according to how it is formed ( To start with it cannot handle identical twins with similar names. I have worked with many databases of people (customers, staff, partners etc.) and this sort of approach is *not* recommended.
Without commenting on HIS, the normal solution is for each database to hold their own unique identifier and, if (and only if) linking is required, then a 'correlation id' is held which provides a connection to the other source.
Establishing a linkage in the first place is, of course, not trivial but it is a solved problem. There are tools available (we provide one) that can heuristically match customer records from multiple sources with different data formats etc. Once the match is made a standard linkage can be formed. If the match is uncertain, it is escalated for human review, but generally the machine is better than a person at connecting the dots.

Anonymous said...

Hope you are feeling better David!

It's interesting that the "better" EMR sites in NSW (including RNSH) are those where the local district has embraced the state programs. No-one likes those big ugly state programs run by eHealth, but they are realistically the only way to get the people and dollars to do anything beyond the basics.

Contrast North Shore and POW say, with RPA which was once a national leader in EMR, but under the current CEO has drifted for years tinkering with smallish local projects, partly due to a "we know best" philosophy as well as lack of funding. Whereas those willing to get into bed with the central bodies have made way more progress.

Bernard Robertson-Dunn said...


according to NEHTA documentation:

6.4.4 IHI
An IHI may be assigned to any person who receives healthcare.

To uniquely assign the IHI, a limited amount of identifying
information will be used. This will include(9):
• Name
• Date of birth (DOB)
• Date of birth accuracy indicator
• Sex

(9) Information associated with the IHI is compliant with AS5017 compliant –
Health Care Client Identification.

Other information "may" be included.

This doesn't seem enough to make it unique, but, by using the word "include" they haven't told us everything.

The critical part seems to be that DHS information, held by Medicare as a Trusted Data Source, can be used to recreate the SLK581 used in census data. So ABS claims that names and addresses will be deleted and only the SLK kept is vacuous spin. The government can always recreate SLKs with a small error rate, which could probably be corrected manually.

Linking ABS data with myhealthrecord data will be/is trivial and without time limits.

Anonymous said...

"Touch-on/Touch-off" has been well implemented in Princess Alexandra with Cerna. This is seems to work well, rather than a proximity system that makes assumptions about people nearby.

Bernard Robertson-Dunn said...

I think it's worth looking at a couple of websites in the UK. Australia could learn a lot from this approach. I wonder if Mr Kelsey will try and change the MyHR to make it a bit more like the UK's SCR?

The first is a medical practice:

This is where a patient's health record is held and where the majority of their health information is stored.

It is interesting to note how much choice and control a patient has.

The second is an NHS site "Your health and care records"

The centralised, NHS record system is a Summary Care Record.

The site says:

"Your SCR contains the following basic information:

the medicines you are taking
your allergies
bad reactions you may have to certain medicines

It also includes your name, address, date of birth and unique NHS Number which helps to identify you correctly.


You can choose to add any information to your SCR that you think will help improve your care. This can be of particular benefit to patients with detailed and complex health problems. You and/or your carer should discuss anything you wish to add with your GP."

Re access control:

"Who can access or view my SCR?

Only authorised healthcare professionals directly involved in your care can access your SCR. Your SCR will not be used for any other purposes. The person viewing your SCR:

needs to have an NHS Smartcard with a chip and passcode
will only see the information they need to do their job
will have their details recorded every time they look at your record

In addition, the healthcare professional must seek your permission if they need to look at your SCR. If they cannot ask you because you are unconscious or otherwise unable to communicate, they may decide to look at your record because doing so is in your best interest. This access is recorded and checked by the Privacy Officer of the organisation to ensure it is appropriate."

"Can I opt out of having a Summary Care Record?

You can choose to opt out of having an SCR at any time. If you do opt out, you need to let your GP practice know by filling in an opt-out form. If you are unsure whether you have already opted out, you should talk to the staff at your GP practice."

Anonymous said...

One cannot OPT-OUT of My Health Record if it is going to be made COMPULSORY as stated by Jim Birch at HIC16.!!

I doubt Mr Kelsey, now the holder of the poison chalice, will be able to change much as he would have to tell Mr Madden and Mr Royle their approach and strategies have been wrong and the Minister would not like hearing that.

Bernard Robertson-Dunn said...

The current eHealth legislation states:

"2 Minister may apply the opt out model to all healthcare recipients after trial
(1) If, having applied the opt out model under clause 1, the Minister decides that the opt out model results in participation in the My Health Record system at a level that provides value for those using the My Health Record system, the Minister may make My Health Records Rules applying the opt out model to all healthcare recipients in Australia."

My reading of this is that there will always be an opt-out choice. Unless they change the legislation.

Talking about the legislation, my comment yesterday about health service providers being able to adopt other people's identity was inaccurate. The advice I've been given is that it allows a health service provider to use health identifiers in their own systems, not that they can take on another person's identity.

My Bad. Apologies to all.