We need better data to understand the ACA in North Carolina
August 13, 2015 Leave a comment
Callie Gable and I have an op-ed in today’s Raleigh, N.C. News and Observer outlining why we believe that we don’t have enough information to evaluate the functioning of the North Carolina’s ACA insurance market (~500,000 persons in N.C. are enrolled in private plans). What do Callie and I want to see?
- A population based tracking poll that answers basic questions about persons who are–and are not–enrolled in ACA plans.
The Kaiser Family Foundation California Longitudinal Panel Survey is a series of surveys that, over time, tracks the experiences and views of a representative, randomly selected sample of Californians who were uninsured prior to the major coverage expansions under the Affordable Care Act (ACA). The initial baseline survey was conducted with a representative sample of 2,001 nonelderly uninsured Californian adults in summer 2013, prior to the ACA’s initial open enrollment period…..The third in the series…after the second open enrollment period in spring 2015 to find out whether more have gained coverage, lost coverage, or remained uninsured, what barriers to coverage remain, how those who now have insurance view their coverage, and to assess the impacts that gaining health insurance may have had on financial security and access to care. A fourth survey in the series will keep…The surveys are designed and analyzed by researchers at KFF and the fieldwork costs associated with the spring 2014 and spring 2015 surveys were paid for by The California Endowment.
We need to start with the population of North Carolina for whom the ACA exchange is most relevant–the uninsured–and understand their experience (anyone could become uninsured, of course). We can never do what California did which is select a random sample of the uninsured prior to the first ACA open enrollment period and follow them, but we could design a survey of uninsured at a future point plus those now covered in ACA plans and ask them retrospectively about their experience prior to the ACA (imperfect methodologically, but we can only start from where we are now).
This chartbook provides the type of information that this survey has answered in California. For example, transition in insurance status for the first to the second open enrollment in Cali (all of whom were uninsured prior to the first open enrollment):
The key thing to understand is that this starts with the population of California, and then looks at how they interact with the ACA. We need that in North Carolina.
- What about premium increases for BCBSNC and other insurers?
California’s tracking poll asks about premiums, choices and options. For example, you can estimate how important premiums were to those who remained uninsured as compared to other reasons, because they asked people. The thing we most need is a population based understanding of how the ACA is affecting N.C.
I wish we had plan level enrollment data by county, on a retrospective basis. If insurer X listed 20 plans in county Y (5 bronze, 5 sliver, 3 gold, 2 platinum), then after the fact say how many people signed up in each plan, in each county. This is the point about a range of premium increases being incomplete information–it is hard to know what they mean without knowing what people actually choose, which is of course a function of the choices they have. Because N.C. has a federal exchange, this information is not released. I wish the federal government would release the data. And as an aside, about a year ago I (Don Taylor, not Callie who co-authored the op-ed) spent some time asking for this plan by county level sign up information, and one federal official told me the State Insurance Commissioner could release the data, but that they could not. Some contacts in the State Insurance Commissioners office told me if was the opposite (feds could release, but they could not). I dropped asking about it, and didn’t make any sort of official request.
The big picture questions people have about the ACA are best answered by a population-based tracking poll, so that is where my focus is going. We are actually trying to raise money and develop such a N.C. tracking poll. Let me know if you are interested in being involved–especially if you want to help pay for it! A credible effort will be in the $500,000-$800,000 (per year) range, we think.