Skip to content Skip to main navigation

Lifestyle

We mean business
Contact us today to get results

How the dodgy data was found at Canberra Hospital

By johnboy - 21 August 2012 12

health statistics

An intriguing set of documents thrown up on the Open FOI website.

Someone has asked for documents relating to “Identification of issues with emergency department data in February 2012”.

First we have an email from Andrew Bailey, to Lee Martin, Phil Ghirardello, Michael Hall, Kate Jackson (the only person to resign over the health statistics fraud), and Ian Thompson.

It’s about the Australian Institute of Health and Welfare raising a stink over a large number of patients with unusually round wait times of exactly 30 minutes, 60 minutes, and 240 minutes between 6 February and 12 February.

Almost as shocking as the fraud is the ham fisted way they couldn’t even put some dither on the wait times. There are high schoolers falsifying their survey data for assignments with more sophistication than these clowns.

What’s Your opinion?


Post a comment
Please login to post your comments, or connect with
12 Responses to
How the dodgy data was found at Canberra Hospital
JessP 2:36 pm 22 Aug 12

Holy Hell – the data wasnt only manipulated – it was POORLY manipulated.

kakosi 12:18 pm 22 Aug 12

I think people are missing a really big elephant in the room. If it was their practice to change wait times what else do they change?

Years ago I had to insist the doctors in charge of my mother’s care put back details of a procedure ordered by one of their team, while my mother was in emergency, into mum’s hospital medical records.

The ONLY reason I knew it was missing from her medical files when she went upstairs to a ward was because other doctors in that “team” then tried to tell me it had never happened. And since the procedure seemed to have contributed to mum going into kidney failure it was pretty important it should be recorded. The head of the treating team asked the “doctor” involved to apologise to my mother and myself for the “clerical error” he’d made.

If I hadn’t have been with mum 24/7 I would never have known what had happened.

aidan 10:44 am 22 Aug 12

Skidbladnir said :

We’ve been doing exactly the kind of data manipulation a situation like this requires since at least the 1930s, and today you can find an automated function to generate usable values to achieve a central median in about fifteen minutes.

Would they be normally distributed though? Bounded below by zero wait time, but as we all know, there is no upper bound to how long you can wait at the ED — at least it seems like that sometimes.

NB: My ED wait times (0m, 45m, 90m) have been completely acceptable considering the seriousness of my injuries (burst appendix, lacerated forehead – bone visible, deep cut to a finger), and the care I have received has been top notch. Thanks for that.

Skidbladnir said :

(But doing it well enough, for long enough, and in a way that outlasts your employment, to fool both auditors and statisticians at AIHW would be a considerable feat)

Indeed. A chap I knew working on meteorological data developed some methods for checking data quality for long time series. One particular station on the NSW south coast would always pop up as being aberrant. When he investigated it turned out the temperature measurements had been contracted out to a local business person after the Post Office stopped doing the measurements. They decided the tourist traffic might be increased if their town was a little warmer.

(For those inclined to conspiracy theories, no this does not invalidate observed global warming)

dpm 9:54 am 22 Aug 12

Thumper said :

Why in that case would you fake it so badly?

Compliant Greens and Labor have made sure that it makes no difference if anyone found out or not.’

Just think, every single bit of information the public gets fed by the current government could be as corrupt as this and there is nothing anyone can do about it.

Absolutely nothing at all.

Come now, it’s not all that bad.

You’re forgetting this initiative:
http://www.cmd.act.gov.au/open_government

Hahahaha! 🙂

bitzermaloney 9:44 am 22 Aug 12

Thumper said :

Why in that case would you fake it so badly?

Compliant Greens and Labor have made sure that it makes no difference if anyone found out or not.’

Just think, every single bit of information the public gets fed by the current government could be as corrupt as this and there is nothing anyone can do about it.

Absolutely nothing at all.

And yet they’ll all probably be re-elected in Oct.

Interesting article in the crimes from the other Kate (who was shafted over the Bruce Stadium Affair). Seems there was a hint more integrity in the assembly back then with the minister taking responsiblity for their portfolio.

http://www.canberratimes.com.au/act-news/assembly-gone-soft-carnell-20120821-24l1e.html

Oh how I wish the Greens would have the guts to actually stand for something, instead of being a pseudo-faction of the ALP.

Thumper 8:43 am 22 Aug 12

Why in that case would you fake it so badly?

Compliant Greens and Labor have made sure that it makes no difference if anyone found out or not.’

Just think, every single bit of information the public gets fed by the current government could be as corrupt as this and there is nothing anyone can do about it.

Absolutely nothing at all.

Skidbladnir 8:51 pm 21 Aug 12

F***ing amateurs.

We’ve been doing exactly the kind of data manipulation a situation like this requires since at least the 1930s, and today you can find an automated function to generate usable values to achieve a central median in about fifteen minutes.

(But doing it well enough, for long enough, and in a way that outlasts your employment, to fool both auditors and statisticians at AIHW would be a considerable feat)

johnboy 4:52 pm 21 Aug 12

There is of course, nothing to suggest the involvement of Mssrs Martin, Ghirardello, Hall and Thompson.

Chop71 4:35 pm 21 Aug 12

lol, just imagine the figures on their tax returns.

cranky 4:31 pm 21 Aug 12

Has anyone asked Mssrs Martin, Ghirardello, Hall and Thompson for comment?

But having said that, is it possible that all these individuals were uninventive enough to use the fairly precise 30, 60 and 240 minute cutoffs when manipulating the data? That is, of course, if they were involved at all.

VYBerlinaV8_is_back 4:14 pm 21 Aug 12

harvyk1 said :

Why in that case would you fake it so badly?

Because you’re hopeless.

harvyk1 2:15 pm 21 Aug 12

You can purchase programs for around the $300 mark and takes about 2 hours to set up properly which can generate good high quality fake data which when done right would be nigh impossible to tell apart from the real thing. (I personally use such a program for stress testing databases, need to have pretty realistic data in there otherwise it is difficult to spot problems).

Hell you could probably find a freeware or open source program which would at least put in semi believable data and only take 5 minutes to set up.

Why in that case would you fake it so badly?

Related Articles

CBR Tweets

Sign up to our newsletter

Top
Copyright © 2017 Riot ACT Holdings Pty Ltd. All rights reserved.
www.the-riotact.com | www.b2bmagazine.com.au | www.thisiscanberra.com

Search across the site