Ed and Erika have covered most of the bases on Glitchapalooza today but I want to put this in front of you because, in its own way, it’s the most baffling Healthcare.gov failure yet. Via Peter Suderman of Reason, WaPo buried a fascinating fact about the website’s capacity at the end of a story it published a few days ago:
David Brailer, who worked as HHS’s first national coordinator for health information technology during the launch of the Medicare drug benefit in 2006, said the administration could have anticipated that the opening of the federal exchange would trigger a rush of Americans onto the Web site, either as onlookers or outright buyers.
He pointed out that the exchange was built to accommodate 50,000 to 60,000 visitors at a time — fewer than half as many as the enrollment site for the Medicare drug benefit could handle. The number of older Americans eligible for the drug benefit was far greater than the group of uninsured people who will be allowed to buy insurance through the health exchange, Brailer said, but many elderly patients didn’t have home computers at the time, compared with the near-universal access to the Web that exists across the United States today. For a new program that’s had as much advertising as the Affordable Care Act, building a Web site for just 60,000 people at a time “is weird. The math just doesn’t add up,” he said.
It’s one thing to say the feds badly, badly underestimated how much traffic volume there’d be in the first few weeks. It’s another to point out that they underestimated it despite having a volume benchmark courtesy of another federal health-care program for which there was heavy demand. Any techies out there have an explanation for that beside the obvious one of grotesque incompetence? If you knew that the Medicare Part D site could serve as many as 100-120,000 users at a time, what would be the logic for building Healthcare.gov at half the size? The only explanation that occurs to a layman like me is that capacity on the Medicare site might have ended up being wildly greater than what they ended up needing. E.g., maybe only a thousand seniors at a time nationwide ended up using the site, rendering 99 percent of the capacity redundant. But even if that were true, surely they should have expected a crush of ObamaCare curiosity-seekers to Healthcare.gov during rollout. And even beyond the rollout stage, a health-care product pitched (mainly) at young adults is bound to get many times as much traffic as one pitched to people aged 65 or over. The whole reason they rushed the site out now, when it’s still buggy, is that the White House hoped to maximize the amount of public dependency on it before the GOP could make any inroads on repeal/defund/delay. If anything, that should have meant going overboard on capacity to make sure no one was shut out — and thanks to Medicare Part D, they had a handy justification for boosting it to 100-120,000 at a minimum since that’s what the Bush administration did. Why on earth wouldn’t they do that? Is it significantly more complex to double capacity? Even if they didn’t trust Medicare Part D’s site as a benchmark, couldn’t they at least have talked with Blue Cross, say, to get a sense of what their capacity was like and then scaled up from there accordingly?
Even more baffling is the fact that, because the feds knew the site would be buggy, they also had reason to believe that individual users would be forced to make repeat visits in order to complete the sign-up process. If the site were running smoothly, a user might be done after one try, upon signing up successfully. As it is, the news this week has been nothing but horror stories of people trying every day, for hours on end, to create an account and failing at various stages of the process. That’s another reason to err on the side of more capacity rather than less. Instead, via Jim Geraghty, not only are people forced to try, try again, but some portion of the site’s successful “enrollees” are surely duplicate or triplicate accounts created by frustrated users who couldn’t even tell if they’d successfully registered on their first try. That’s what I mean in saying that the capacity problem is arguably the most mystifying disaster of the Healthcare.gov saga. It’s not an execution flaw, as the coding is; it’s a planning flaw, and a big one, and one that could have been easily avoided. With the legacy of the president’s biggest domestic achievement on the line, you might have thought they would have gotten something as simple as capacity right. But here we are.
Erika posted this the other day but it’s worth a re-watch. Here’s DNC chair Debbie Wasserman-Schultz casually noting that, yes, the site was designed to serve 50,000 people at a time. Any problems with that, Debbie?