Quote Of The Year

Timeless Quotes - Sadly The Late Paul Shetler - "Its not Your Health Record it's a Government Record Of Your Health Information"

or

H. L. Mencken - "For every complex problem there is an answer that is clear, simple, and wrong."

Friday, October 04, 2019

You Would Really Hope The Data Separation And Controls Are Pretty Robust I Reckon!

Reports of this initiative popped up last week:
See here:

Government eyes health insurers' data to help uncover Medicare fraud

My Health Data excluded from proposed data-matching push
Rohan Pearce (Computerworld) 23 September, 2019 18:08
The Department of Health has launched a public consultation on proposed legislation that would lay the basis for increased data-matching between government entities in order uncover fraudulent or incorrect Medicare claims. The proposal could also see the data provided from private health insurers automatically matched against government data holdings to red flag potential fraud.
A consultation guide released by the department states the proposed Health Legislation Amendment (Data-Matching) Bill 2019 and associated regulations won’t give the department new powers to conduct compliance activities and will not “authorise the automation of compliance outcomes or raising of debts” (although such activity could be conducted with human oversight on the basis of anomalies flagged by data matching).
If passed, the bill would enable Medicare Benefits Scheme (MBS) and Pharmaceutical Benefits Scheme (PBS) data to be automatically cross-matched against data from the Department of Veterans’ Affairs, the Therapeutic Goods Administration, data from private health insurers and registration information from the Australian Health Practitioner Regulation Agency, as well as immigration data provided by the Department of Home Affairs.

Data from private health insurers would be provided to the department by the insurers on a voluntary basis; no Medicare data would be provided to private health insurers, however. Data from the My Health Record system would not be in the scope of the data-matching proposals.
The data-matching scheme would have three key aims: Detecting Medicare fraud, detecting incorrect claiming and detecting inappropriate practices by health providers.
The Department of Human Services earlier this year released details of a data-matching scheme combining data from Centrelink and Medicare to unearth discrepancies between records held at the two agencies in order to tackle fraud or overpayments.
A controversial government data-matching scheme dubbed ‘robodebt’ by its critics, which combines data from Centrelink and the Australian Taxation Office to flag potential overpayment of welfare benefits, is currently the subject of a prospective class action lawsuit.
Lots more here:
There is also coverage here:

Govt proposes Medicare data matching scheme for fraud crackdown

By Justin Hendry on Sep 24, 2019 11:21AM

Could source data from private health insurers.

The Department of Health is set to expand its data matching activities to better detect fraudulent or incorrect Medicare claims under proposed new laws that would make it possible to pull data from private health insurers for compliance purposes.
But unlike Service Australia’s controversial robodebt program, the department has ruled out using the changes to introduce any automation or perform any compliance activity without human oversight.
Draft laws released for consultation late on Monday are aimed at improving the existing Medicare compliance program by introducing a new “scheme of data matching for permitted Medicare compliance purposes”.
Around 1 percent of the $37 billion handed out through Medicare and other health services claims each year are estimated to be fraudulent, with up to a further 4 percent falling into what the department labels “occasional or inadvertent non-compliance”.

The department’s existing compliance program uses “tip-offs and data analysis of existing Medicare datasets” to identify fraud and incorrect claims.
But while these mechanisms are a “highly valuable source of information”, they are “limited” and the department wants to identify and recover a “greater percentage of incorrectly paid benefits”.
The proposed changes would give the department unrestricted freedom to access Medicare Benefits Schedule (MBS) and Pharmaceutical Benefits Scheme (PBS) data that is currently accessible only in “narrow circumstances”.
“The proposed legislation will exclude data matching from this restriction when the matching is undertaken for specified Medicare compliance purposes,” the consultation paper states.
Access to data from the My Health Record would, however, be restricted to the department, though much of this data is also available through the MBS and PBS datasets.
The proposed changes would also allow “certain information” to be disclosed to federal government agencies for both Medicare compliance purposes and to “assist them in performing their functions”.
This data could be sourced from the Department of Vertans’ Affairs, Department of Home Affairs, Australian Health Practitioner Regulation Agency and the Therapeutic Goods Administration.
Private Health Insurers could also provide data to the department on a “voluntary basis” for Medicare compliance activities, though Medicare data will not be shared with those insurers.
Not another robodebt
The department was keen to highlight the distinction between its proposed data matching expansion and robodebt, otherwise known as the online compliance intervention program.
More here:
You can download the 8 page discussion paper from here:
To me this proposal is perfectly reasonable for detecting fraud but I am not so sure when it comes when it comes to monitoring non-financial practitioner behaviour. Maybe that should also be out of scope?
Crucial in all this is that the boundaries remain intact and that there is no scope creep. I have not a high level of confidence this will be the case.
What do you think?
David.

2 comments:

Anonymous said...

The track record of the department of health says - do not trust them. Even if you believe them today, you can't trust them to not change things. opt-in became opt-out. Explicit consent became no consent.

Bernard Robertson-Dunn said...

Those readers who are interested in better healthcare rather than just IT might like to have a look at a book from 2016:

Secondary Analysis of Electronic Health Records
https://link.springer.com/content/pdf/10.1007%2F978-3-319-43742-2.pdf

The essence of Chapter 1 is that secondary use of EHR data is very hard and so far hasn't been done either at all or very well.

Large scale hospital EHRs present challenges for secondary use.

My conclusion is that a summary record system like myhr is orders of magnitude less useful than clinical databases (either hospital or clinical/GP), which themselves do not easily map to research use.

Those who argue that myhr will eventually be a useful research tool haven't done their homework. They should read this book and then tell us just how a summary system like myhr could ever be of use in a medical research context.