Wednesday, July 18, 2012

The Lack Of Health Sector Wide Knowledge and Understanding Bites NEHTA Hard. Pretty Pathetic Really.

The following popped up today:

States not ready for e-Health system

GENERAL practitioners will have to wait up to three years to receive secure discharge summaries digitally signed by hospital doctors following more delays to the Gillard government's e-health system.
State and territory health departments say they are not ready to use healthcare providers' 16-digit unique identity numbers created for the national system to verify the identity of doctors or other medical staff creating a patient's discharge summary.
Healthcare providers individual identifiers - dubbed HPI-Is - were created and assigned to all registered doctors two years ago as part of the Healthcare Identifiers service launch, which also saw unique 16-digit identifiers allocated to every Australian enrolled on the Medicare database.
Use of local hospital or state health agency identity numbers instead of a uniform national identifier will impact their use for authentication and audit purposes within the personally controlled e-health record system.
The ability to accurately identify individual healthcare providers, health organisations and consumers using the PCEHR system is key to securely exchanging electronic information and reducing the potential for errors - either through assigning records to the wrong patient or sending documents to the wrong doctor.
User verification is supposed to be provided through the not-yet-available National Authentication Service for Health (NASH), which is also supposed to provide an audit trail of all access to a patient's electronic record.
Delivery of the NASH is subject to ongoing negotiations between the contracted supplier, IBM, and the Health department after the PCEHR system launched without it on July 1.
Health chief information officer Paul Madden yesterday called for comment from stakeholders on a proposal to abandon the mandatory requirement to include HPI-Is in discharge summaries in the near to medium term.
Lots more of the saga is found here:
Here is a slightly more technical summary of the issue from NEHTA:


The current Nehta specification and proposed Australian Technical specification for discharge summaries includes a mandatory requirement for an HPI-I.
Jurisdictions and lead implementation sites have difficulty in obtaining HPI-I’s for their providers, and have been unable to secure funds to store HPI-I’s in their systems.
Jurisdictions have estimated a 1-3 year horizon for the implementation of HPI-I’s.
As a result, it will take a considerable time for discharge summaries to be available in the PCEHR if the HPI-I remains a mandatory requirement.


Nehta and the NCAP have been reviewing the opportunity for a number of quick wins for increasing the clinical utility of the PCEHR through the jurisdictions.  These opportunities have been presented by the NCAP to the Nehta Board.  A key opportunity in most jurisdictions is the implementation of an interface from the hospital discharge summary process into the PCEHR.  Queensland, South Australia, Victoria, NSW, the ACT and the NT have all expressed a desire to move forward with discharge summaries.
Jurisdictions raised the issue of the steps they would have to go through to implement the HPI-I’s into the discharge summary.  A series of workshops, facilitated by the NCAP and attended by jurisdictions, Doha and Nehta concluded that by relaxing the conformance requirement of a mandatory HPI-I on discharges, that jurisdictions would then be able to build the necessary infrastructure to send discharges to the PCEHR. 
It is noted that while the requirement of a HPI-I is relaxed, there remains a mandatory requirement for a Provider Identifier of some nature.  This identifier may be an identifier supplied by the healthcare facility associated with the individual provider.  It may be of any form (e.g. Numeric, Alpha-Numeric, etc).  The identifier must be unique to the facility, for example a concatenation of the local identifier linked with an OID derived from the facility’s HPI-O or ABN.
A clinical impact and safety assessment has concluded that there is no introduced clinical risk by relaxing this conformance point.
----- End Extract
So it seems that no-one at NEHTA has noticed that, as far as most of the States are concerned, use the Health Identifier Service is distinctly optional - and they will get round to it sometime in the rather ill-defined future.
It should be noted that without Individual as well as Provider Identifiers flows of information both to AND FROM the NEHRS / PCEHR are made more problematic. How can an internal hospital system use B2B to request PCEHR records if it does not handle the HI Service?
You can read my comments on feeder systems  being an issue from last year here:
What is not clear to me - reading this - is just how a particular discharge summary - with a local identifier - can be safely merged with a PCEHR record. It is not clear to me if the IHI remains mandatory (looking at the V1.3 specs it seems not) and if all the hospital systems are able to use this identifier.
An obvious question is, of course, if the IHI is not mandatory just why is that and how is that decision justified?
Despite the words from NEHTA Clinical Safety I think this needs some careful development of test cases to make sure any interactions between the Hospitals and the PCEHR occur exactly as expected with a high degree of reliability.
The failure to take a health system wide view - when working with a so called ‘nation system’ - just shows how far out of touch some are.
What exactly are NEHTA playing at with all this?