Subscribe
Share
Share
Embed
Each week, Health Affairs' Rob Lott brings you in-depth conversations with leading researchers and influencers shaping the big ideas in health policy and the health care industry.
A Health Podyssey goes beyond the pages of the health policy journal Health Affairs to tell stories behind the research and share policy implications. Learn how academics and economists frame their research questions and journey to the intersection of health, health care, and policy. Health policy nerds rejoice! This podcast is for you.
The refrain feels constant. If only, they say, we could reduce the amount our system spends on administering health insurance, our nation would go a long way toward reducing spending overall, so the argument goes. Of course, there are a lot of assumptions built into this theory. One is that we already know what exactly drives such high administrative spending in the first place. Another follows that is if we know what drives administrative spending, maybe we can also uncover why that spending is high in certain states, in certain markets, and why it's more muted in other states and other markets.
Rob Lott:The bad news is that until now, the answers to these questions have remained pretty murky. The good news is that there are a lot of smart health policy researchers on the job, and I am delighted that we get to host one of them on our Humble Podcast here today. I'm here with Doctor. Jason Buxbaum, Assistant Professor of Health Services Policy and Practice at Brown University. Together with coauthors, he has a new paper in the March issue of Health Affairs that identifies, quote, substantial variation in administrative spending and profit across state insurance markets for the year 2023.
Rob Lott:Friends, this is an important paper, and, I really don't think we've seen something making these cross state, cross market comparisons of insurer administrative spending and profit in quite such a systematic and focused way. So I can't wait to dig in. Doctor. Jason Buxbaum, thanks for being here.
Jason Buxbaum:Rob, good morning. Thanks so much for having me.
Rob Lott:Let's dig in a little bit to the subject of your research. I know in most discussions of high and rising health care spending, there are these constant references to administrative spending and profit as a sort of key category. And while most people probably have a sense of what they mean, maybe you can help us start by defining our terms. In particular, what are we talking about when we talk about administrative spending?
Jason Buxbaum:Sure. By and large, we we focused our conception of administration on, on expenditures that are not for clinical care. In this particular corner of the world that gets you to a pretty common sense sort of definition. So we're talking about IT expenses, cost containment, a subset probably of quality improvement expenses that don't involve actually directly interfacing, with members, compensation that flows to agents and brokers. And then there's this huge lump that we can observe in regulatory filings.
Jason Buxbaum:It's just general and administrative. So hard pressed to know exactly all of the contents of this category. It's it's a catchall, but it's huge. So we kind of we we came at this, not necessarily from a place of first principles, but from the data that exists, and what can we what can we work with here? What do regulators commonly capture when they ask for information on administration?
Rob Lott:Got it. Alright. In your paper, you you compared state to state and market to market. And I'm curious if you could set a baseline by saying a little bit about why state to state is a good framework for assessing variation in administrative spending and profit. Why not county to county or metropolitan statistical area?
Rob Lott:What's significant about the state comparison?
Jason Buxbaum:Well, the data exists. So I I'm not sure that it's superior, but it exists. And I think every frame that I hope to come back to is just we're incredibly constrained by available information and what gets filed. So that's why we focused on the state.
Rob Lott:Got it. Alright. Well, let's dive right in. It's in that context that you use regulatory filings to estimate state level, all plan administrative spending, profit for each of three separate market segments. Those are fully insured commercial coverage, Medicaid, and the third one, self funded commercial coverage.
Rob Lott:So give us some of your key takeaways.
Jason Buxbaum:Sure. I think, couple key takeaways here are that administrative spending in this country is qualitatively high, and it's it's quite variable. So kind of some top line figures for you. In the fully insured segment, where commercial carriers kind of assume risk and are underwriting coverage on in a typical state, it's about $600 per person per year spent on these administrative functions so far as we can tell from the filings such as they exist. The self funded segment, where insurers or carriers process claims on behalf, generally on behalf of large employers or or unions, but are not necessarily assuming full financial risk.
Jason Buxbaum:In that segment, sort of a bit less than $300, per person per year, so less than half. The numbers may be I'd hesitate to to to to read great precision into these numbers. We're going again with filings such as they are and such as they exist.
Rob Lott:You wouldn't apply your mortgage with this data. You would
Jason Buxbaum:need to
Rob Lott:maybe build your budget for the next year.
Jason Buxbaum:Exactly. I I think that's a very fair way to look at it. These figures are directionally correct, and I think they're telling us something useful about the world. And we observed higher spend on the Medicaid side where states and Medicaid managed care plans have a host of regulatory requirements that that they must fulfill related to eligibility is particularly complicated in Medicaid. We tend to have very tight utilization management in Medicaid.
Jason Buxbaum:And so right. So if, on the fully insured commercial side, we saw about $600 per person per year in administration. Self funded side, about a bit less than $300 per person per year. Medicaid was closer to $670 per person per year. But I think the variation within each of these segments was very striking.
Jason Buxbaum:So the I gave you the medians, but the the variation was something that that jumped out to to me and to my coauthors.
Rob Lott:Great. Well, I wanna ask you about that variation in just a second, but first, let's take a quick break. And we're back. I'm here talking with Jason Buxbaum about variation in administrative spending and profit across state insurance markets. Just in the previous question, you really zeroed in on that variation.
Rob Lott:The numbers you gave us were the median, but obviously there's a spread. And of course, there's variation, right? States are different. Their populations are different. We can't all be like wake woe be gone where everyone's above average.
Rob Lott:And so I'm curious what you make of the degree of variation. Were you expecting more or less? And what does that say about our system writ large?
Jason Buxbaum:Right. So I I've done a lot of reflection on this question. I mean, to to write off a couple of numbers, Medicaid was particularly variable in our in our study, both when you are are including, you know, pry the the profits of private plans and when you aren't, about $250 variation between across the the for the interquartile mean. So right across that twenty fifth to seventy fifth percentile, somewhat less, on the the commercial side. I think, you know, there's a long tradition in our field of scrutinizing variation, saying that there's something here, worth slowing slowing down and, carefully considering.
Jason Buxbaum:I think absent clear differences in need, which aren't apparent, this should make us say, you know, is this variation does it serve the public interest? Does it suggest that that areas with with high per person per year spend are you know, perhaps have room for greater efficiency? I'm not sure if popular interpretation would be the the flip that, you know, our areas with lower administrative spend, missing out on some, you know, neglecting some important activities. But I think it's an invitation to scrutinize where are these dollars going. That's that's what jumped out for me.
Jason Buxbaum:That's what was most exciting about this work.
Rob Lott:Great. So I one of the things that's, I think, really fascinating about your paper is that there's just so much kind of under the hood and the appendix is really robust, the supplementary materials.
Jason Buxbaum:Rod, you're a glutton for punishment.
Rob Lott:I got to make sure I know what we're talking about here. So I appreciate that all of our authors put all that information out there. And, of course, one of the interesting things is sort of looking at which states are above average and which states are not. And I'm curious, I guess, a couple questions. Why didn't you put out a, you know, a clickbait, article that said, this is the state with the highest, you know, administrative spending?
Rob Lott:And then I guess my follow-up question would be, have you received a phone call yet from the good folks in Indiana asking how they can be more like the good folks in Vermont?
Jason Buxbaum:Sure. So I I think there's several good reasons not to take that that approach. I first off, I don't think it really matters from a US perspective whether, you know, whether spending is relatively high in Kansas or relatively high in in Colorado. The the point for me is the variation, and this is an invitation for for scrutiny. I think also really important to remember that this data is noisy.
Jason Buxbaum:I I absolutely believe the variation story, but the idea that that we would necessarily observe, you know, a high spending state staying, you know, at that highest you know, being observed as an outlier across years. This data is is problematic. If I had a magic wand, I would love to improve the quality of information that researchers and regulators have available to work with. So, I I don't think picking on, states at the extreme is is so useful or necessarily fair. I think the point the point here is there's a striking variation.
Jason Buxbaum:We don't know what we're getting, in exchange for the extra spend. And I think there's a really great tradition in our field of, you know, going back to studies of tonsillectomies in Vermont, saying what's going on in one community where, you know, almost all the kids are missing their tonsils and another where the kids, by and large, kept their tonsils. And it's an invitation to say we we have to know more about when the tonsils need to be out. I view this work in the same vein. This is an invitation to understand, you know, what is the the right level of of admin spend and to the extent that that spend ought to vary or reasonably could vary with some you know, in accordance with some rational public policy goals, what's kind of the the magnitude of variation that you'd expect?
Jason Buxbaum:And and I'd I'd really encourage folks to take a look at it at the the figures in our work, and and I think you would tend to agree that that the dispersion here is suspicious and invites further questions about what are we getting for these dollars.
Rob Lott:Great. I'm curious if you had this data during your earlier work, what do you think you would have done with it? Oh,
Jason Buxbaum:this is a hard question, but I know it's I I know it's so important. I mean, right away, I I think it invitation to ask for better information in many of these reports. You'll see a plurality of administrative dollars attributed to this admin in general category. And and when you see variation across states in or across particular health plans and admin in general, you immediately wanna know what is in there, what is driving that difference. Some instances, health plans will will pay consultants for particular benchmarks, to see, you know, how they compare to an industry average, but there's, you know, there's limited public information that allows that that sort of drill down.
Jason Buxbaum:I So think to understand, you know, where you know, the it's, again, an invitation to ask questions about, you know, if my spend on utilization management or my spend on fraud detection or whatnot, is is variable, you know, is is an outlier compared to to my peer plans or my peer states. I think it's an invitation to ask, you know, how much are we spending? You know, how much do we pay our utilization management vendors? How much, you know, is our is our spend on on sales or on customer service? Are we as efficient as we can be?
Jason Buxbaum:At the same time, I wanna recognize that, you know, that The US is is is extreme. And so in any other country, you'd be like, what is with this level? Whereas here, I do think that there's opportunity for us to focus on variation given given where we are. But at the same time, I feel like it's important to acknowledge that in absolute terms, our admin spend is exceptional.
Rob Lott:Great. Well, perhaps a great note to end on, with quite a few questions to for us, for you, for other researchers in the field to follow-up on. One more one more comment from you.
Jason Buxbaum:If you would. Sure. So at the time when when you're an academic and you're going through the you've submitted work you think is is important and, you know, squarely answers the question, Sometimes the the reviewer comments here and there, you know, I will admit, you know, drive you a little nuts, but sometimes the work ends up, you know, substantially better for it. And one ask that the reviewers made of us or was to was to consider profit and how profit relates to administration. I think there are, like, interesting philosophical questions about, you know, should we consider positive profits to to be a form of of administrative spend?
Jason Buxbaum:Setting aside the the philosophical element of it, I really hope readers take a look at our work and see our analyses that include that are inclusive of profit and exclusive of profit. And I think what you'll see is that profit is a relatively modest proportion of overall admin spend. Profit is quite variable across states. Probably, you know, we're capturing some year to year variation. But so I think you can to say that that there's concerning variation in nonclinical spend is not to say that there's necessarily, that that variation is is accruing as as profit.
Jason Buxbaum:I think a reasonable interpretation of our work is that there's concerning bloat, but we should not necessarily conflate profit and bloat. They're they're different issues.
Rob Lott:Well, that's great. And I'm envisioning a follow-up commentary with a some kind of headline of, like, profit versus bloat. What are we talking about here? So perhaps with an illustration with a Wall Street fat cat, perhaps a monocle and a top hat as well.
Jason Buxbaum:That'd be fun, Rob. Let's do it. Let's do it together.
Rob Lott:All right. We're on it. Jason, thanks so much for taking the time to speak with us. This was a really fascinating conversation about a really fascinating paper. So thanks so much for being here today.
Jason Buxbaum:Rob, thanks so much to you and the team for for welcoming me. It's a treat.
Rob Lott:And to our listeners, thanks for tuning in. If you enjoyed the episode, tune in again, leave a review, recommend it to a friend, and, hit that subscribe button. Of course, tune in again next week. Thanks.