MD: 58K OFF-Exchange, 53K ON-Exchange from 1 company; graph clarified

OK, I actually do have one update today (it's 8:30 at night and my kid is getting ready for bed).

When I first learned that Washington State's off-exchange total was a whopping 2.2x that of their exchange-based enrollments back on February 20th, I casually noted that if that held true nationally, it would make the off-exchange QHP tally easily 7.7M or higher ...and this was back when the exchange QHPs were only around 3.5M.

However, I also tried to be cautious in my "official" estimate, keeping it to around 4M or so since I didn't know how many there really were beyond the 300K (later 560K) or so I had documented at the time. After all, other states would likely have a much lower ratio of Off- to On-exchange QHPs; In Wisconsin, for instance, Off-Exchange QHPs were only about 23% as high as exchange-based.

Well, thanks to contributor Esther Ferington for this link which notes that in Maryland, the largest private insurance provider on their exchange is reporting about 10% higher off-exchange QHP enrollments than on-exchange (58K to 53K).

Now, this doesn't change the potential off-exchange enrollment total (up to 7.8M according to the Rand study), nor does it really increase the documented number, since these 58K would be rolled in with the 1.7M confirmed by the BCBSA in general.

However, it does show that there are now at least two states where the off-exchange total is higher than the exchange-based total enrollment number. Basically, this is just more evidence that the 7.8M figure is not only realistic, but may actually be lowballing the 3/31 total, since the Rand study only ran through around March 16 or so.

Assuming that the ratio of on-exchange to off-exchange enrollments ends up remaining around the same from March 17 - April 15 as it appears to have from last October - March 16, it's conceivable that we could be looking at an even higher total:

• March 16 ON-Exchange QHPs: around 4.9M / OFF-exchange QHPs: (potentially) up to 7.8M
• April 15 ON-Exchange QHPs: around 7.8M / OFF-exchange QHPs: (potentially) up to 12.4M

Seeing how I've now documented over 2.09 million (up from about 308K back in mid-February), still not including heavy hitters like Aetna, United Healthcare, Kaiser Permanente and so forth, either the 7.8M or 12.4M figure seem quite possible at this point.

The vast majority of people who bought health insurance through the state's new online marketplace chose plans offered by CareFirst BlueCross BlueShield, the state's dominant insurer. Even more called the carrier and enrolled directly, according to data obtained from the insurer.

CareFirst reported for the first time that it had signed up more than 53,000 people through the health exchange as of April 9. The exchange reported Friday that almost 65,200 people had enrolled in private insurance through the four carriers selling on its website in the same time frame.

Another 58,000 people called CareFirst directly to sign up, many possibly because the exchange experienced technical problems so severe since its launch Oct. 1 that it is being scrapped and replaced with a system used in Connecticut. About 80 percent of people enrolled in CareFirst plans have paid, the insurer said.

IMPORTANT CHANGE TO THE GRAPH:

Note that I've never included the non-documented additional 5.7M or so potential off-exchange QHPs on The Graph; I only include them after they've actually been documented; the same is true of the completely out-of-left-field estimate of 8.2 million additional off-exchange employer-supplied insurance policies (ESI's).

However, I had been including both of these as "not shown" footnotes on the graph as well as part of the high-end range estimates for QHPs and ESIs.

Given the recent revelation about the Rand study not including any children in their survey, however, I've decided that until I have clearer proof of both of these potentially higher figures, they shouldn't really be included in the "high-end range" either. As such, I'm reducing the high end of the "Grand Total" from 34.1M down to 20.2M to match what actually apepars on the graph itself.

I will continue to display the extra 5.7M / 8.2M in prominent notes on both the spreadsheet and the graph; I just won't be listing them as part of the ranges for now.

In addition to making me feel more comfortable about people understanding the ranges, this also makes the graph itself less confusing, since the total displayed visually now matches the total at the high end of the range.