I’m not particularly tech savvy but, one month into the HealthCare.gov website mess, here’s some of what I think we do know now – and what we are still waiting to find out.
Traffic: Volume was high, and it was one reason for the system’s poor performance. But it was not the only reason, not by any means. Nor have I seen a convincing explanation for why the administration low-balled the traffic expectations.
Code: There are millions of lines of it and it’s a mess. It’s being rewritten/worked-around. We don’t know much more than that.
Testing: Woefully inadequate. Would end-to-end testing of the entire system – rather than just testing specific pieces as CMS did – have prevented all of this? Could it have really simulated how it would work in the real world? Maybe it wouldn’t have caught every “glitch” – which seems like too soft a word. But even the non-geeky among us can probably be fairly certain that comprehensive testing would have prevented at least some of what HHS Secretary Kathleen Sebelius called a “debacle.”
Quick fixes: There aren’t any. There have been reports focusing on this component or that one as the source of the problems. These components may be part of the problem – but it sure doesn’t seem like that’s the whole story.
It’s not just the log-in: We know that the initial steps to sign on and get started didn’t work – and we know it’s somewhat better now although that’s hard to quantify. But we also know there are problems at the backend – transferring those “834” enrollment files to insurers. How much is wrong in between? Who knows. The data hub seems to be doing OK; the two recent outages seem to have more to do with capacity and the Verizon division hosting it rather than intrinsic design flaws. State exchanges like Kentucky that are working well couldn’t sign people up without interacting with the federal data hub for premium and subsidy verification and so forth. But whether the hub will keep doing what it’s supposed to do as more people flow through the system is something that we can’t be sure of yet.
Fix or rebuild? Can HealthCare.gov really be fixed or, as some reports have suggested, does it need to be taken down and rebuilt? The administration is pushing the repair message. How long will it take to repair? The administration is predicting a relatively smooth experience by late November. We’ll know how that turns out in late November. Whether it will have to be rebuilt before the 2015 enrollment is a separate question, but not as pressing.
(One little tiny note: The administration has given out “application” figures which of course is not the same as enrollment. But although HHS spokespeople themselves sometimes say X number of people have applied, some of those applications are from families. So that 700,000 application number could really be more than 700,000 people. We don’t know how many more.)