On Monday, Health Affairs published a study that I co-wrote with Dr. Alexandra Mahkoul taking the lead position from the team consisting of researchers from Vanderbilt, Penn, Wakely Consulting and Duke. We wanted to know how did people use the no-cost sharing preventive care services in the ACA and how much same-day cost-sharing happens when someone gets one of those services. We looked at four services; flu shots, annual wellness visit, mammograms for recommended populations and colonoscopies for recommended populations. We identified individuals by the metal level that they chose (Bronze, Silver, Gold, Platinum) as well as the three Silver Cost-Sharing Reduction (CSR) levels (94%, 87%, 73%).
So what did we find?
- There are huge variances in the take-up of preventive care services
For the standard metal plans without CSR benefits, we see a pattern where the probability of using a service goes in the same direction as the actuarial value. We are not saying anything causal. It is an interesting association. I think it could be a reflection of the unobserved information and/or risk preferences of individuals. People who buy Gold plans are different than people who buy Bronze plans. However, that relationship is non-existent when we look at CSR plans. Higher values of actuarial value are mechanically linked to lower incomes. I think this could be a reflection of
That is interesting
- Cost-sharing from other services received on the same day varies a lot by metal and service
We find that as AV increases without regard for CSR or not CSR status, cost-sharing goes down for people who have any same day cost sharing. We also see that the office visit services tend to have a lot less cost sharing on the same day than the cancer screening services. These services vary a lot. Some of the common flu shot services were vaccine administration and 20 minute office visit codes. These hint that the flu shot was an “add-on” instead of the only intended service for that day. However colonoscopies had substantial anesthesia and surgical pathology charges. These were far rarer but way more expensive.
We only see people who received preventive services. We don’t see the behaviors of folks who would have gotten preventive care services if things were different somehow. We observed relatively low uptake of preventive care services as seen from claims files even when the services were cost-sharing free. People may not make that distinction if there are other cost-sharing events that happen on the same day/visit or if there are other barriers to care.
Wag
Interesting graphs. What is really striking is the level of uptake for screening colonoscopies. At first glance it looks really low, but once you realize that a routine coloscopy is a once every 10 year procedure the uptake is pretty good across all levels of insurance.
BradF
David
Outside of mammo and colonoscopy same-day costs, the variances you allude to and illustrate make interpretation difficult. That is the point I suppose, but help me understand what CMS should takeaway from these findings. What is signal and what is noise, and what would be the next steps in policy making if you were king for a day?
Brad
David Anderson
@BradF: This is a purely descriptive piece. Plan selection is a function of premium, cost-sharing expectations, network effects, and individual level risk tolerance, so seeing more preventive services as higher AV is bought aligns with unobserved personal characteristics story. I think the interesting thing is the CSR story is not the non-CSR story. There, I think more thinking about non-cash barriers to care is needed.
I’m not sure what my recommendation is if I was CMS Administrator for the day.
Ohio Mom
Congrats on being published!
BretH
A thought: ACA covers a lot of people who had limited insurance, or none before. Preventative care is as much a habit as anything, and maybe people used to scrimping on their health care are just not as likely to schedule things like checkups, etc.
David Anderson
@BretH: I could buy that story in 2015. I have a hard time buying that story by 2018.
Uninsured is not a permanent state, it is a transitory state for many people. I was uninsured for several stretches in my 20s but I transitioned to ESI at various points.
Ang
I really wonder how many people know just what services are no-cost. For folks who have trouble affording their regular co-pays, it may be the first thought is “that sounds expensive and I feel fine so I’ll skip it”. When money is really tight, you have to assume _everything_ has a price.