Information available about how individual health market risk, and Affordable Care Act risk adjustment, has worked is still limited, and the results have varied so much from state to state that it’s hard to make any predictions.
The drafters of the ACA wanted people to be able to buy major medical coverage without having to go through medical underwriting. They prohibited insurers from considering personal health status when issuing individual and small-group coverage, starting in January 2014, and they prohibited insurers from using personal health information other than age, location and tobacco use when pricing coverage.
Insurers and regulators worried that sick people would converge on specific plans with great benefits and great doctors and drown those plans with big claims. To keep unexpected swings in enrollee claim risk from capsizing plans, the ACA drafters and implementers created the risk-adjustment program.
The program, based partly on risk-adjustment programs now in use in Medicare product markets, is supposed to take cash from issuers with enrollees who come in with what appear to be low risk scores, based on their medical history and other factors, and send the cash to issuers with what appear to be high risk scores.
Some insurers, and especially small, new insurers, have argued that program rules and formulas have favored large insurers.
Owen used many different methods to analyze how risk scores and ACA risk-adjustment transfers change from 2014 to 2015.
Risk scores tended to increase, but the relationship between risk scores and number of covered lives in each state’s individual health market was unclear, and the year-over-year change patterns varied widely from state to state, Owen says.
In Oregon, for example, many of the issuers have much different risk transfer amounts in 2014 and 2015.
In New York state, 2014 and 2015 risk-adjustment transfer amounts appeared to have a close, predictable kind of relationship, and the sizes of the transfer amounts were clustered close together.
“Time and more data will lend perspective, but still it is clear that effective analysis of a particular health plan offering in a state will require analysis of directly applicable data,” Owen writes. “National averages or results from other states, and even other time periods, will be challenging to translate.”
Have you followed us on Facebook?