Arizona

Every summer I root through a mountain of health insurance policy premium rate filings, mostly by using the SERFF (System for Electronic Rates & Forms Filing) database from the National Association of Insurance Commissioners (NAIC).

Depending on the state & carrier, some of these can be found easily; others are either heavily redacted, partial, not available until later in the year; and some are never made available at all.

In addition to the filings for the upcoming year, however, the SERFF database also includes a mountain of other filing forms, from non-ACA compliant insurance policies (short-term, indemnity, etc.) and from previous years (I have no idea how far back they go, but I'm guessing it's at least since the turn of the century). This also includes "grandfathered" and "grandmothered" policies.

Arizona

Back in mid-August I posted my analysis of Arizona's preliminary 2022 rate filings for the 2022 individual & small group health insurance markets. At the time, I wasn't able to get ahold of the actual carrier actuarial memos which include the number of people enrolled in those policies in 2021, which means I wasn't able to run a weighted average rate change for either market.

As a result, I had to go with unweighted averages, which are far less useful since how much a carrier with a huge market share changes their premiums will have a much bigger impact on the statewide average than one with only a handful of enrollees.

Arizona is a case in point: At the time, the unweighted average was a 4.8% reduction on the individual market and a 7.7% increase for small group plans.

Arizona

The good news is that the federal Rate Review database has now posted the preliminary avg. 2022 rate filings for the individual and small group markets for every state. This makes it very easy to plug in the average requested rate changes in 2021 for every carrier participating in both markets.

The bad news is that most of the underlying filing forms are heavily redacted, meaning I can't use the RR database to acquire the other critical data I need in order to run a proper weighted average: The number of people actually enrolled in the policies for each carrier.

This means that in cases where this data isn't available elsewhere (either the state's insurance department website, the SERFF database or otherwise), I'm limited to running an unweighted average. This can make a huge difference...if one carrier is requesting a 10% increase and the other is keeping prices flat, that's a 5.0% unweighted average rate hike...but if the first carrier has 99,000 enrollees and the second only has 1,000, that means the weighted average is actually 9.9%.

Arizona

I've once again relaunched my project from last fall to track Medicaid enrollment (both standard and expansion alike) on a monthly basis for every state dating back to the ACA being signed into law.

For the various enrollment data, I'm using data from Medicaid.gov's Medicaid Enrollment Data Collected Through MBES reports. Unfortunately, they've only published enrollment data through December 2020. In some states I've been able to get more recent enrollment data from state websites and other sources.

Today I'm presenting Arizona. For enrollment data from January 2021 on, I'm relying on adjusted estimates based on raw data from the Arizona Health Care Cost Containment System (AHCCCS).

Arizona total Medicaid enrollment hovered around 1.1 million people (including ACA expansion) for several years until the COVID pandemic hit last spring.

Arizona

Now that I've developed a standardized format/layout & methodology for tracking both state- and county-level COVID vaccination levels by partisan lean (which can also be easily applied to other variables like education level, median income, population density, ethnicity, etc), I've started moving beyond my home state of Michigan.

Here's Arizona:

Note: The CDC says that around 69,000 vaccinated state residents (2.7% of the total) have unknown home counties.

Arizona Icon

As I noted recently, I've relaunched my project from last fall to track Medicaid enrollment (both standard and expansion alike) on a monthly basis for every state dating back to the ACA being signed into law.

For total monthly Medicaid enrollment, the official Medicaid.gov monthly enrollment data is only available dating back to late 2013, and it's only current through November 2020. The Kaiser Family Foundation has also compiled the pre-2014 average enrollment for each state based on the 3rd quarter of 2013. In some states I've been able to find more recent enrollment data for December 2020 or later.

Over at Xpostfactoid, my colleague Andrew Sprung has been doing a great job of tracking ACA Medicaid expansion enrollment growth since the onset of the COVID-19 pandemic this past February/March at the macro (national) level, by looking at around a dozen states which have monthly reports available. He puts the overall enrollment growth rate at 23.6% from February thorugh October 2020.

I've decided to take a closer look at individual states. The graph below shows how many Arizonans have been actively enrolled their Medicaid expansion program (awkward named the Arizona Health Care Cost Containment System, or AHCCCS):

The good news is that the federal Rate Review database has now posted the preliminary avg. 2021 rate filings for the individual and small group markets for every state. This makes it very easy to plug in the average requested rate changes in 2021 for every carrier participating in both markets.

The bad news is that most of the underlying filing forms are heavily redacted, meaning I can't use the RR database to acquire the other critical data I need in order to run a proper weighted average: The number of people actually enrolled in the policies for each carrier.

This means that in cases where this data isn't available elsewhere (either the state's insurance department website, the SERFF database or otherwise), I'm limited to running an unweighted average. This can make a huge difference...if one carrier is requesting a 10% increase and the other is keeping prices flat, that's a 5.0% unweighted average rate hike...but if the first carrier has 99,000 enrollees and the second only has 1,000, that means the weighted average is actually 9.9%.

Louisiana's 2020 Presidential primary was scheduled for April 4th, but the other day Democratic Governor John Bel Edwards and Republican Secretary of State Kyle Ardoin agreed to reschedule it for June 20th...which is actually later than the last previously-scheduled primary in the U.S. Virgin Islands on June 6th:

The presidential primary elections in Louisiana slated for April will be delayed by two months, the latest in a series of dramatic steps government leaders have taken to slow the spread of the new coronavirus.

Secretary of State Kyle Ardoin, Republican, and Gov. John Bel Edwards, a Democrat, both said Friday they would use a provision of state law that allows them to move any election in an emergency situation to delay the primary.

The presidential primary elections, initially scheduled for April 4th, will now be held June 20th. Ardoin said in a press conference he does not know of any other states that have moved elections because of the new coronavirus, or COVID-19.

Back in early August, I ran the preliminary average unsubsidized 2020 individual market rate changes in Arizona. At the time, I had the requested rate changes for both the individual and small group markets, but not the actual enrollment numbers for each carrier, so I had no way of calculating the weighted average. I instead settled for a simple unweighted average, which came in at around a 2.4% reduction in premiums on the individual market and a 5.2% increase on the small group market.

A few days ago, the Arizona Insurance Dept. released the final/approved 2020 rate changes, and there was only one significant change: Health Net of AZ (dba Arizona Complete Health), which had requested a 2.9% rate reduction, will instead be keeping their premiums flat year over year on average. With Health Net holding over 50% of the market share, this meant that the statewide average is a bit higher than I had it previously.

Pages

Advertisement