as the program takes effect this year. Eligibility errors date back
to the launch of the federal marketplace.
The Health and Human Services Department announced
the extension of a special insurance program created by the
Obamacare legislation to provide coverage for those with
pre-existing conditions. The extension was intended to help
consumers that want coverage but got trapped by website glitch-es. This also includes some of the 4 million consumers who had
their individual plans cancelled because they failed to meet the
new law’s coverage standards.
In mid-January HSS Secretary Katherine Sebelius told Congress that she had asked the department’s inspector general for
an independent investigation into contracting and management
factors that contributed to the technology failure.
What to make of this?
First, the most glaringly obvious fact is that the train wreck was
inevitable and foreseeable. There never was enough time to
define, build, and test the complex software that was required.
This is no mere “website”, rather it is a complex and multi-ten-tacled system that integrates with many disparate federal, state,
and private insurance company applications, sending and
receiving complex and sensitive data in many different formats.
The aforementioned eligibility rules are complex and rely on the
retrieval and assessment of a lot of external data. The pricing
algorithm and associated coverage choices are many and varied.
Second, it was obvious months ago to both the government
and CGI that there was insufficient time to do what was necessary. Was there a pact, spoken or unspoken, between vendor
and customer to keep their collective heads down, report status
as yellow, and just keep going? Many of us have been in that exquisitely awful circumstance in which we know the piano isn’t
going through the door, at least not on the current schedule, but
we don’t know enough to answer the questions which will inevitably follow by confessing that yellow is actually red. In order to
declare red and survive the consequences it would be necessary
to stop the project long enough to evaluate where the project is,
and that would be suicidal. With so much outside pressure it
seems the project just kept going until it was obvious to anyone
watching that the delivery was a failure.
And third, it is likely the political downside of doing something rational would have been too great for logic to prevail.
Clearly it was within the power of the Obama Administration to
delay the website launch. It delayed key aspects of the pro-
gram such as the Individual Mandate, and reverse marched the
insurers into reinstating coverage (where possible) for policies
cancelled at the government’s behest. So, as is always the case
the (technology) debacle could have been avoided, but at great
political cost. Now we have a technology debacle and great
political cost, the worst of both worlds. To paraphrase the old
saying – there was not time to build it properly, but now there is
time to fix it. And apparently money! Everyone who has been
involved in an enterprise software implementation knows that
the longer it takes to identify and address a bug, the more it
costs; with production errors being the most expensive of all.
Fourth, what’s a vendor to do? CGI is a big, capable
company. Why did they fail? Ultimately because they weren’t
big enough and strong enough to control an incompetent and
irrational client. CGI knew the position they were in, they
knew there was insufficient time, they knew the testing was in-
adequate, they knew the software was unfinished and bug-rid-
den, and they knew that the website was going to blow up in
October. Maybe they tried to do the right thing but if they did
they failed. Maybe they just kept their heads down. After all
they had two hundred and ninety two million reasons to keep
quiet. But what of that small but insistent voice that nags at the
back of the mind “We’ll get fired. We might get sued.” Maybe
the saddest aspect of this whole sorry tale is that it isn’t the
first and it won’t be the last. Government IT projects have a
tendency to fail and the same small group of vendors seem to
remain in contention despite prior performance. This time may
be different, even by government standards Obamacare is huge
both in scope and visibility.
Fifth, what happens next? Hopefully Accenture is big
enough, strong enough and DHHS is chastised enough that the
fix project can be rationally estimated, funded, programmed,
tested and implemented. Hopefully Accenture is big enough
and strong enough to control an incompetent client, and hopefully the client is frightened enough to actually listen and act
accordingly.
Sixth, what of the beauracrats that foisted this on the unsuspecting public and burned hundreds of millions of dollars
doing so? The inspector general will order his investigation
and we the taxpayers may know more, or not. Certainly the
time to fire the guilty parties is now. After all, the government
just fired the vendor. Worst case CGI is only partly responsible
for this debacle. DHHS was minimally responsible for defining
its requirements, acceptance testing the software, managing
its vendor and reporting status and advising the White House.
Clearly they failed badly.
Lastly, spare a thought for the poor foot soldiers that
burned endless hours and days working around the clock to
rescue a hopeless situation. The rank and file from CGI and
the government gave up their lives for months to try and
rescue their managers and masters from their own hubris and
incompetence. I hope there is some solace in all this for them,
but it’s hard to see what. ITA
(George Grieve is a popular writer and speaker on the
subject of insurance technology solutions and is the author of the book Shop Talk. He is CEO of the consulting
firm CastleBay Consulting.)
(The views and opinions in this column are those of the
author and do not necessarily reflect the views of the
Insurance Technology Association or its members.)
