28 March 2019

Prising the secure messaging limpet from its rock

Insights

 

Two weeks ago Wild Health sent an email into the Australian Digital Health Agency (Agency) with a raft of questions enquiring as to how their secure messaging program was going. From the sidelines, it seemed like progress, initially looking promising, had ground to a fairly serious halt.  This was despite a lot of effort on the part of the Agency in the last 18 months to bring together the key industry players and sort out the key issues, finally.

What we got back was a belated invite to the launch of the Agency’s interoperability consultation, and a short note to explain that everything is, of course, in hand.

Our first question was fairly simple. Where did the Agency think it was on its own program timetable for progress on secure messaging? While the My Health Record (MHR) project had been capturing all the attention of the public (some would say oxygen) for the past 12 months, the Agency had identified the issue of secure messaging quite early in its life as critical, and had determined to twin track a program to sort out the issue alongside it’s major MHR push.

It was a sensible call. While the MHR was being feted as a life saver of the future, many industry stalwarts, including some of the secure messaging vendors themselves, have pointed out that in the absence of any governance or standards for secure messaging, it is highly likely that the errors in the system caused through bad electronic messaging could be killing patients today.

Given the risk of life and death in getting messaging wrong – its ‘clear and present danger’ –  many in the industry are wondering why is there still no firm framework requiring vendors who use messaging in their software to meet minimum standards of safety and testing for their products.

In every other part of healthcare, risk and managing patient safety is  almost universally at the heart of all patient management. Indeed, in all parts of society we see robust governance regimes (managed usually by the government in some way) to protect the public from bad products. Imagine if there was no regulation or governance around the testing of aircraft software in this day and age.

Healthcare messaging is necessarily very complex and subject to a lot of risk in the process of attempting accurate, and timely delivery. Millions of  HL7v2 messages are being exchanged daily by products which have no standards or compliance regime around them. And a lot more non HL7 formatted messages are being sent without any real compliance either (eg, PDF, CDA and even text formats).

With any governance and compliance regime comes additional cost to commercial operators so you might expect the major secure messaging vendors to be in the camp of resisting moving forward on this issue. But most of them want some form of sensible governance regime, to protect themselves, their businesses, and their ultimate clients, patients.  Mostly though, the variety of formats they have to deal with from all the medical software vendors in the country who want to send a secure message, is so wild, varied and constantly changing with each software update, that they are spending much of their time in translation at one end and then the other, in order to see messages make it through the system.

About a year ago there seemed to be a lot of optimism that we could foster a constructive relationship between the key providers for the greater good, so their protocols would at least talk to each other, and that we could have the beginnings of simple but effective standard for message formatting for software vendors, and a master phone book – a master directory of providers electronic addresses and all the details at a receivers end we needed to know to talk to any provider or supplier via any system.

But it seems the ADHA has run into a raft of problems and we are experiencing secure messaging “groundhog day”.

Most of the issues are the similar or the same to those from the past:

  • Each of the existing software vendors who want to send electronic messages are busy, usually cash strapped, and with large existing development programs in play. Being forced in some way to conform to a standard format for more sophisticated interoperable messaging isn’t on their immediate agenda, and no one is prepared to put any pressure on them to get it there. It’s Manyana mostly.
  • Although the large messaging vendors would like the software vendors to standardise, they also have commercial footprints to protect, including their own IP in their own directories, and product, so walking the line in which each agrees to a roadmap that eventually has one master directory for the system (ie, they share their IP), has to be very skilfully negotiated. Although the Agency went into this new program with high ideals, and a determination to bring the vendors along, it doesn’t seem like they’ve been able to win their trust back enough to achieve progress.
  • Industry observers say that Agency staff who are running the process simply do not have the technical know how and experience to run the process and that is driving disenchantment from the participating parties. Unreasonable deadlines are mentioned as well.
  • The technologies surrounding secure messaging are changing and no one is prepared to make decisions around how these new technologies might be incorporated into a compliance regime moving forward. A key issue is with the web sharing standard FHIR, which might form the basis of a lot of different form of secure sharing in the future, but for which there is no commitment to by any government bodies in Australia to date. And a lot of nervous vendors who for reasons above don’t wish to have to start re-architecting their systems for the future that FHIR is promising. This is despite serious commitments being announced overseas which back a future where FHIR is central in compliance.

In the light of issues like these, one of the major vendors – Medical Objects –  walked away from the project, at least for now. At the time they walked away, they had already done much of the work needed in writing a common format that software vendors could use. They did that in collaboration with Argus and Healthlink at the behest of the Agency, but when push came to shove, the Agency wasn’t prepared to put in place a regime that transitioned all vendors to such a common standard in a meaningful way. Not yet anyway. The standard is published so far only for use for PDF and CDA formats, which discounts  the massive proportion of messaging that is already done using HL7v2.

Not much is compelling medical software vendors to align themselves into a single standardised protocol. 

Medical Objects walking away represents nearly 25% of the market, some 60,000 healthcare provider users and  most of Queensland.  It is difficult to see how anything uniform will be achievable in the near term with such a big part of the market now off reservation.

Apparently, the bosses at Medical Objects felt that implementing the standard format they wrote for common use only in part (CDA and PDF only at this stage), wasn’t going to work. Especially given the Agency wasn’t even going to commit to enforcement of this.

They felt they were wasting their time. They are up for participating still. But only when they see a meaningful path forward.

Another commercial problem might be that although the big pathology companies are making lots of noise about supporting the ADHA in its work, it is still commercially difficult for them to collapse their bespoke messaging systems entirely and join the compliance party. If they went to a common system then they would risk disrupting the patterns of their downstream reselling channels at GP practices.

A possible side effect of this fallout is that some of the patient software system vendors are apparently discussing the possibility of joining forces with a third party directory provider to bypass the commercial vendors, at least on doctor to doctor messaging, or more specifically GP to specialist. That puts Genie in the middle of any such negotiations, likely with one or both of MedicalDirector and Best Practice. The third party directory provider is likely to be Healthshare, which is said to have the best specialist directory in the country and has the technical knowhow to help the other vendors.

It might be surprising if MedicalDirector comes in though, given that they tried their own secure messaging exchange a few years back and it failed spectacularly. A vendor industry insider said of the MedicalDirector failure that the nitty gritty of sorting messaging, even doctor to doctor was far more complex than MD bargained for.

Without a proper universal path to better messaging such a move by the patient management system vendors might end up fragmenting the system of messaging even further.

Getting to a “meaningful path”  of secure messaging compliance and governance was never going to be easy and some commentators feel that the Agency severely underestimated the complexity of the task they set themselves, both commercially and technically.

On spec, previous roadmaps developed by the much maligned NEHTA (and then added to and published by the ADHA) look sensible and reasonable. The Conformance, Compliance and Accreditation (CCA) Governance Group, originally part of NEHTA, developed a framework that had:

  • NEHTA and then the ADHA being the governing body for conformance and implementation
  • Testing being done by the vendors that met the standards set by the ADHA
  • Third party verification being required in higher risk security, privacy or clinical situations
  • Annual audits of vendors being required to update processes and product enhancements to the standards
  • A range of reasonable interventions for non compliance, but including some enforcement measures

All of this seems fairly standard for a compliance regime, and talking to some of the major vendors off the record, they maintain that despite the cost, such a regime would be a good thing, if it was sensibly introduced and maintained.

If the industry’s major players aren’t objecting to such a regime, indeed, some are even openly asking why one still isn’t in place, what is wrong with this picture?

Various views are being put on industry blogs, at event panels around the country and sometimes even  in the news media. Much of the commentary centers on an ongoing distrust of the ADHA to effectively manage such a process, a criticism which feels like it is a possible side effect of the Agency bulldozing through the MHR project through to its opt-out finale, to please political masters more than sensible health system objectives.

After all, if you had to choose between solving secure messaging, which does pose such a huge existing and real risk to patients, and which is probably the major roadblock to progress in overall system interoperability (and the safety and efficiency it might bring if sorted), would you have spent your money and time on the MHR or secure messaging?

Amid the commentaries swirling around, there is a possible answer.

Secure messaging is invisible to many of the power brokers that matter in transforming healthcare in Australia.  It simply isn’t on the radar as a serious concern. It is highly technical and backroom, and while its always been in the scope of the governments objectives to reform the system, it’s simply not been taken seriously enough by government agencies. Why won’t the government many wonder, bite the bullet, and give the software vendors a reasonable but firm ultimatum to get on board? What is holding them back?

Many of the patient management and other medical software vendors who do understand what a critical point of failure in our healthcare system that secure messaging remains, are in a difficult commercial position in one way or another. Without some sort of stick in the process, or carrot, from the government (where else will it come from?) there isn’t enough commercial or political downside presently, or penalty, to compel them to change.

And why would you risk changing even if you wanted to without having a firm guidance on what your future standards regime would look like? It would be a little like the risk our major energy companies would take in building infrastructure for our future energy needs that the government hasn’t firmly communicated will actually fit into policy.

In the words of one industry observer who is flabbergasted that we find ourselves in such a situation after so many years, “It might take a significant and potentially catastrophic event for anything really to happen here”.

Let’s hope not.

Secure messaging and standards for interoperability will be a key topic of discussion at WILD HEALTH No 4 in Sydney on June 25.

CLICK HERE for SPECIAL TICKETS OFFER and use promo code WHP1

CLICK HERE FOR FULL PROGRAM AND SPEAKERS