Enrollment reports aren’t terribly useful

I’m trying not to get swept up in the periodic frenzy that seizes the media every time HHS releases new information about exchange enrollment. The data they provided most recently paints a more complete picture of the enrollee population than anything else to date, but it’s still not a very useful picture. That’s true now, and it’ll be true in April, when the administration releases data capturing the entire open enrollment period.

Since the application itself doesn’t require any health information, we won’t know how much adverse selection the insurers will encounter until beneficiaries start filing claims. Among other things, this means that even insurers won’t even have a clear idea of their risk pool until after they need to set initial premium rates for 2015 (April May 1 of this year, if I recall correctly).

Contra the media’s laser focus on young adults, we shouldn’t lean too heavily on enrollee age distributions as a health indicator for state risk pools. Age is the best proxy we have for health—but that doesn’t mean it’s a good proxy. The 3:1 age band goes a long way toward correcting age-based discrepancies in health spending. It doesn’t correct completely, of course, but it does a fair amount of heavy lifting. This week’s estimates, which show enrollment skews toward individuals over 45, would be much more alarming if premiums were fully community-rated—like, say, employer-sponsored insurance.

I’m not convinced that the distribution of metal tier enrollment tells us much, either. It’s possible that signing up for higher-deductible plans (bronze or silver) indicates good health, and that sick people would concentrate in gold and platinum plans. Evidence relevant to this is mixed. There have certainly been studies that show healthier people will select into less generous plans. But we also know that in Australia, people who opted to purchase more generous supplemental coverage (on top of the nation’s universal system) were, on average, healthier than those who maintained the basic level of coverage in the country’s universal system. These findings are mirrored in “advantageous selection” in Medigap policies. Factors like income, education, and cognitive ability are correlated with purchasing more robust coverage, regardless of health status.

For low-income individuals, silver plans are also the cheapest option eligible for cost-sharing reductions, which are effectively a second form of subsidy for individuals below 250% FPL; these assist with deductibles and copays. We do know from the HHS report that about 80% of marketplace enrollees are receiving some form of tax credits, but without a breakdown of enrollee income, inferences are tough to draw.

The risk pools were always going to skew unhealthy (or “high acuity”) in their first year. Exchanges may not meet the CBO’s “7 million” target—not that this matters substantively—but the CBO also projected that enrollment would more than triple by 2016, as the mandate grows more robust. The same logic about young healthies waiting to sign up should hold. That is, the exchanges should be healthier by the end of open enrollment than they are now, and they should be healthier in 2016 than they are in 2014.

Insurers needed to estimate a baseline risk pool to set premiums. It’s those projections that matter when it comes to assessing the state of the risk pools, not media hype. When insurers say demographics are looking better than expected, they know the appropriate reference point; reporters and bloggers probably don’t.

Adrianna (@onceuponA)

Hidden information below

Subscribe

Email Address*