Network manipulations

Networks are a key feature of all insurance plans as every provider has a network. A network is a group of providers who have contracted with an insurer to get paid a certain set of rates for a certain set of services performed on the people who are insured by the company. Networks can be manipulated for positive public gains and for evil.

Networks are regulated at the state level for most insurance products and have some federal guidance for Medicare, Medicaid and Exchange products. The regulations are rather loose. In PPPACA, the networks for the Exchange must have significant concentration of providers in a wide variety of specialties so that residents of a county where a network is sold has “reasonable” access. This is a very soft definition because assembling a rule that makes sense for Loving County, Texas and Cook County in Illinois is extremely difficult. States will define network adequacy differently. Most states will have multiple standards depending on population density as urban areas will require more providers that are closer to the average resident while rural areas will allow long drives.

When networks are working for the public good, they provide a means of cost control as an insurance company will include in a network a provider who will take 120% of Medicare while it may exclude a provider who wants 220% of Medicare.  This works because an insurance company can guarantee volume of patients at the lower rate.  Networks are also a quality control metric.  For instance, insurance companies may refuse to contract with providers who are not board certified or who have multiple large malpractice settlements against them in the past ten years.  Contracts can be terminated for the loss of licensure or large malpractice settlements as well.

However, networks can be used against the public interest as well.

In a world where medical underwriting is no longer directly allowed, insurance companies will still want to find a way to avoid taking on risk or even worse, guaranteed high cost individuals.  Network manipulation is one of the few viable ways of doing this that does not guarantee a very nasty series of investigative reports from the local television station and a seven part series in the daily news paper.

How does it work?

Let’s work through an example of a plan that is trying to minimize the number of Type 2 diabetics it is covering on the individual or Exchange risk pools.  Type 2 diabetics tend to use a disproportionate number of  endocrinologist, primary care provider, dietitian, ophthalmologist and podiatrist appointments.  Manipulating primary care provider locations and access does not help out much as primary care providers are relatively cheap and are too useful for the rest of the risk pool.  Futzing around with dietitians and ophthalmologists are an option, but again, they are reasonably cheap and unlike PCPs, they are reasonable low usage services.

The network manipulation to make a plan less attractive to Type 2 diabetics is two fold.    The first is to recruit and maintain a large number of endocrinologists, but have a focus on reproductive endocrinologists or adrenal endocrinologists or whatever subspecialty of endocrinology that can treat diabetes does not have diabetes as a primary disease of concern.  The more subtle manipulation would be to maintain a minimal number of podiatrists in the network, and keep the podiatrists that have nasty offices, rude receptionists and shitty hours in a bad neighborhood.  This type of network design will make the insurance product less attractive to a person with Type 2 diabetes than a product with a broad, unstructured network.

Conversely, network and benefit design can be used to build a slightly healthier risk pool than the general population.  Benefit design can be used to build a healthier network by offering to cover the health club dues of members.  Adding the regional sports medicine center is also a network decision that will attract younger and more active members than the general population.


52 replies
  1. 1
    jayackroyd says:

    Am I reading this right?

    3000 out of pocket before the deductible is exhausted, then a 50% copay on most services. Cap is 6350. Premium is 348 a month. Is that right?

    Who would buy this? It’s typical of the NY State Bronze plans.

  2. 2

    Shorter Obama: “If you had a shit plan when the law came into effect, you get to keep your shit plan. If your shit plan was changed or created since then, it isn’t grandfathered.”

    Now normally I would not care if people are dumb enough to desire insurance that provides them little preventive care or will drive them bankrupt, but I don’t want them raising our cost when they end up in the ER.

  3. 3
    aimai says:

    What bothers me the most is that there shouldn’t be any “Bronze Plans” unless you are mean’s tested into the 2 percent. Why should the state let you buy/be sold a plan that only works if you are independently wealthy? This is so backwards. Poor people should get the gold plans through medicaid, and rich people should have to buy the bronze plans. Both should get access to the same doctors and networks etc… but rich people should just pay more. I get that we need buy in from rich people but this is ridiculous.

  4. 4
    jayackroyd says:

    @aimai: Why should the state [require you to buy] a plan that only works if you are independently wealthy?


  5. 5
    Richard Mayhew says:

    @jayackroyd: yep, you’re reading it right — the question is what is the subsidy level because if you are buying bronze, you’re highly likely to be subsidized and what are the alternatives.

    If the subsidy level is fairly high, then it makes sense to buy that plan if for nothing else to get the discounts that insurers get from provider’s chargemaster

  6. 6
    jayackroyd says:

    @Richard Mayhew: Yeah, but I’ve been in a conversation with a couple of people. They say, correctly IMO, that the ACA will only succeed if the unsubsidized participate fully. Otherwise the death spiral looms.

  7. 7
    Baud says:


    The death spiral is about the mix of healthy v. unhealthy, not subsidized v. unsubsidized. Or so I thought.

  8. 8
    jayackroyd says:

    @Baud: the unsubsidized healthy who have been priced out of the market is what is supposed to make this go, make it self sustaining. Markets and all, you know.

    And what good do the premium subsidies do if you’re still out of pocket 3K before the insurance kicks in at 50 cents on the dollar? Why would I pay even 100 dollars a month for that? I’m better off setting that aside for fee for service.

  9. 9
    PJ says:

    @jayackroyd: The thing is, even these bronze plans are much much better than what could previously be obtained by individuals on the open market. There are a number of different bronze, silver, gold and platinum plans offered by various insurers under the NYS Exchange, all with different costs in each particular level. The New York COOP, Health Republic, had the best rates. Their bronze plan was not great either, but it is less than $100 more, $438/month, for the gold level, which has a $250 deductible, no or low co-pay on most doctor visits, and OOP cap of, if I remember correctly, $3500. Under current insurance rates offered by Blue Cross Blue Shield (through the end of 2013), I would have to pay over $1400/month for the equivalent of a bronze plan, so even the bronze plans being offered represent an incredible reduction in cost from the status quo.

  10. 10
    kc says:

    @The Other Bob:

    Yeah, those dumb moochers.

  11. 11
    jayackroyd says:

    @Richard Mayhew:

    “If the subsidy level is fairly high, then it makes sense to buy that plan if for nothing else to get the discounts that insurers get from provider’s chargemaster”

    Right. You don’t have to pay 1000 dollars out of pocket for the ER visit, just 300.

    I get that this is a real benefit. But it’s an awfully fucked up system that makes you pay a gatekeeper to get access to what are merely wicked steep rather than completely fantastical prices.

  12. 12
    maximiliano furtive, formerly known as dr. bloor says:

    When networks are working for the public good, they provide a means of cost control as an insurance company will include in a network a provider who will take 120% of Medicare while it may exclude a provider who wants 220% of Medicare

    Uh-huh. In RI, any willing provider who wants to offer mental health services to the working poor and impoverished should be ready to accept 30% below Medicare and routine violations of law for timely payment of clean claims. None of the commercial plans for the well-off pay 120%, and most barely clear 100%. Of course, if the providers ever tried to organize a collective action against these abuses, we’d be in violation of anti-trust laws.

    Can you take a break already? Insurance companies are a profit-making entity. Full stop.

  13. 13
    jayackroyd says:


    Yes, I know that. It’s why people chose to go naked, which they hate doing.

    But I don’t think this is materially better. It’s still unaffordable without a subsidy. And pretty fucking expensive with the subsidy. As Richard says, what you’re really doing is paying a substantial fee to get in network pricing.

    Pretty sucky.

  14. 14
    Soonergrunt says:

    Thanks, Richard. Great stuff, as usual. You keep posting, and I’ll keep learning.

  15. 15
    MomSense says:


    The thing is, even these bronze plans are much much better than what could previously be obtained by individuals on the open market.

    YES!!! And if you have a pre-existing condition you could not obtain any insurance in the individual market!

  16. 16
    jayackroyd says:


    Right Mom, but that’s not good news if the mandate gets postponed or people just take the penalty. The insurance companies will just say, sadly, we have to increase rates. And then more people opt out (NOT adverse selection, but just simple arithmetic. there’s no informational asymmetry here.) and here we go death spiral.

  17. 17
    Persia says:

    NH right now is facing a plan that excludes the major hospitals in the two most rural, poorest counties. What a coincidence.

  18. 18
    jayackroyd says:

    @Soonergrunt: Likewise. I very much appreciate these threads. They’re firmly grounded in reality. It’s a pleasure knowing I can ask a question and get a straight answer.

  19. 19
    Baud says:


    : the unsubsidized healthy who have been priced out of the market is what is supposed to make this go, make it self sustaining.

    Proof for the bolded claim?

    As far as the market goes, we’ll see how many people actually sign up in due time.

  20. 20
    MomSense says:


    Well I would buy the plan you highlighted because I was paying 1,600 per month with a 10,000 deductible, 50% copay and I can’t remember what the co-insurance was but that was bad too. I think there are a lot of people like me who would pay for that plan and feel grateful. It seems there are a lot of people just waking up to how treacherous the individual market has been for years. Ain’t no death spiral like the uninsured death spiral.

  21. 21
    jayackroyd says:

    One more dumb, obvious question I don’t wanna get wrong. This is after tax right? Premiums are not deductible?

  22. 22
    Richard Mayhew says:

    @jayackroyd: they are wrong — the population that needs to get into the risk pool is the relatively young and relatively healthy (Under-35 or Under-40 depending on who you are talking with). If roughly 28 to 32% of the risk pool is under one of those ages, the math works out nicely.

  23. 23
    jayackroyd says:

    @Baud: Wonks in email have been saying that. Let me find a public source. Not for “proof” of course but that it is fundamental to the plan design. (Although the mandate alone should be all the proof you need.)

  24. 24
    Richard Mayhew says:

    @maximiliano furtive, formerly known as dr. bloor: Mental and behavioral health is a very different market than physical health — and in this case, you’re right, MH private providers seldom pay even 100% of Medicare.

  25. 25
    Richard Mayhew says:

    @jayackroyd: The opt-out by rate increases is minimized by the subsidy design, there is a cap to income spent on health insurance. At 138% FPL, 5.25% of income is the cap, at 400% FPL, 9.5% is the cap — a rate increase spiral would impact some people but it would mainly impact subsidy levels.

  26. 26
    jayackroyd says:

    @Baud: From wiki:

    With universal healthcare as one of the stated goals of the Obama administration, congressional Democrats and health policy experts like Jonathan Gruber and David Cutler argued that guaranteed issue would require both a community rating and an individual mandate to prevent either adverse selection and/or free riding from creating an insurance death spiral; they convinced Obama that this was necessary, persuading him to accept congressional proposals including a mandate

  27. 27
    jayackroyd says:

    @Richard Mayhew:

    Thanks very much. Unsubsidized young and healthy? And, I’m sorry, I’m not sure what “risk pool” means or what the 32% is of. Can you explicate a little.

    I’d really appreciate it.

  28. 28
    Richard Mayhew says:

    @jayackroyd: Re: premiums before or after tax — speak with an accountant/tax lawyer to be 100% as I am only 95% sure on this….

    If an individual is buying on Exchange, it is after federal tax UNLESS that individual had money in a Health Savings Account, and uses that money for premiums/deductibles/co-pays etc. Most Bronze plans have either an Flexible Spending Account of Health Savings Account feature which allows for payment of deductibles/co-pays/co-insurance to be pre-tax income.

    If an individual is getting their insurance through the SHOP Exchange (small business exchange), the employer contribution is definately pre-tax and I think the employee premium portion is also pre-tax (I am not 100% on this, more like 90% sure on this)

    This is asinine — the tax status should be the same no matter where the money originates from, and going forward the kludge of the Cadillac tax will start equalizing some of the tax status discrepency, but this has been the case since the 50’s so it was not a new feature of Obamacare.

  29. 29
    jayackroyd says:

    @Richard Mayhew: Got it. And the medical loss ratio limitation is supposed to prevent insurers from just going to town on the subsidies?

  30. 30
    Baud says:


    Nothing in there about subsidy v. nonsubsidy.

  31. 31
    jayackroyd says:

    @Richard Mayhew: One of the PR problems is that people on employer based plans report their post tax out of pocket contribution to their friends and relations, not the fraction of total comp that goes to insurance coverage. They’re not even aware of what it is really costing them. So the private market premiums look bigger than they really are.

  32. 32
    Richard Mayhew says:

    @jayackroyd: Okay, risk pool is the entire population being served by a plan or group of plans. In the most general sense it is the people who are buying Exchange policies as there is some serious back end plumbing between plans within a state that should create a quasi-de facto single risk pool instead of the current fifteen or twenty fragmented risk pools in my state alone. The bigger the risk pool, the closer actual experience will reflect projected/statistical experiences. The smaller the variance, the less risk premium has to be charged.

    28% to 32% of the entire population buying on the Individual Exchange that is under the age of 35 or 40 is about what needs to happen to make the math work out. And given that the median age of the uninsured is roughly 35 or 36, this is very plausible (although a decent chunk of this population is now Medicaid eligible).

  33. 33
    Richard Mayhew says:

    @jayackroyd: Agreed — although this too is changing. Look at your W-2 from last year and look at box 12DD — if you received employee sponsored health insurance, that number is what it cost you and your company in premiums.

  34. 34
    Richard Mayhew says:

    @jayackroyd: Medical loss ratios are subsidy independent. They are intended to make sure the health insurance company is spending most of their money on medical care or medical case management and not on hookers/blow or more prosaically back-end bureaucrats whose job it is to deny care.

    Medical loss ratios would have made sense in a no PPACA passed world either — it is just regulation of the economy.

  35. 35
    jayackroyd says:

    @Richard Mayhew:


    1) to Baud’s question, how much of this population would need to be unsubsidized, or does that not matter?

    2) What prevents insurers from simply raising premiums? The mandate keeps the participants in the pool, and the subsidies minimize hardship.

  36. 36
    Richard Mayhew says:

    1) Subsidy status is risk pool independent in that if we have a pair of identical twins, Jane and Judy Doe both buying the same Exchange product, it does not matter that Jane is buying with a $300 a month subsidy and Judy is paying for the entire premium herself. Subsidies are needed to get the young, healthy and typically poorer individuals into the risk pools, but once they are in, a federal dollar spends just as well as a private dollar (if anything the federal dollar is slightly better as it is guaranteed to show up on time)

    2) Good question — there are a couple of tools in play here to keep premiums under control. The first is public shaming in that all significant rate increases have to be reviewed by state regulators. This will be more effective in states with a culture of active and competent government. More importantly, Exchange products are reasonably substitutable products so competititon should keep prices in reasonable control (yes, this fails in New Hampshire, West Virginia and other states with only a single current Exchange entity), and subsidy design comes into play. I think subsidy design is the key here as subsidies are keyed off the 2nd lowest priced Silver plan in a market. Pricing a product significantly above the 2nd Silver means the out of pocket costs for a subsidy eligible individual are very high when a near identical product is either no cost or significantly cheaper. This should lead to fairly intense competition to cluster a wide number of Silver products right around the 2nd lowest (see California for an example)

  37. 37
    pseudonymous in nc says:

    The opposite reaction to this can be seen provider space, where hospital groups start taking ownership of specialist practices and become the entity that negotiates the contract with insurers.

    At least what this does is expose the most egregious bullshit — for instance, network pricing vs rack rate — that can be taken on in the next phase. After all, if the mandate works, then rack rate pricing should go away, and it’s possible to have a conversation about bullshit billing within the network model.

  38. 38
    PopeRatzo says:

    I am reminded of a certain Asian gentleman trying to make a dollar out of something less.

    This is a horrible plan. We were so enthusiastic about the President that we never noticed. Oh wait, we couldn’t have noticed because they were making it up as they go.

    So now we get to talk about patching up this horrible monstrosity instead of actually dealing with health care costs. So somebody want to tell me how the ACA moved us toward single-payer?

    Sold out, we were. completely took.

  39. 39

    Do you take requests?

    Now what I need to know is, 20% of what.

    Let me expand: At the gym, the audiologist, the oncologist, everywhere my insurance card entitles me to a discounted rate during the period I’m spending my deductible–discounted rates vary.

    Is there anything in ACA that requires my health plan to give me, the individual member applying a subsidy of tax dollars to my premium, a particular rate? Or do they still get to run my plan through their billing software and sadly inform me that my co-pay is $734, because that’s 25% of the rate for Suckers Who We Hate?

    TIA if you can address this. Think it might be helpful shopping into for folks who have been bare or on HDHPs and have been lucky enough not to use services.

  40. 40
    TopClimber says:

    @jayackroyd:Perhaps you have factored in this number, or your income level disqualifies you, but subsidies also reduce copays and deductibles–not just premiums– for folks below 250% of poverty level. http://www.kaiserhealthnews.or.....idies.aspx

  41. 41
    Richard Mayhew says:

    @PhoenixRising: I’m not sure I’m understanding your question.

    If it is the co-insurance amount, it is 20% of the contracted rate above deductible until you hit your out of pocket max.

    So let’s work through an example:
    Deductible 2,000
    Coinsurance: 20%
    OOP Max: 5,000

    Dr. Joe bills your insurance company for $57,000 because your cat tried and barely failed in killing you.
    The insurance company sees that Dr. Joe is in your network and for the services rendered, there is a contracted amount that adds up to $22,000

    You are responsible for the first $2,000 for your deductible.
    On the next $20,000, you would be responsible for either 20% or $3,000 whichever is less. You are responsible for $3,000 in co-insurance on the next $15,000 in contracted amount. The insurance company is responsible for $12,000 of that 15K.
    The insurance company then pays out the last $5,000 to Dr. Joe.

    Overall, the insurance company pays $17,000, and you are responsible for $5,000. Dr. Joe is legally allowed to chase you down for that $5,000 but can not go after you for the difference between the billed amount and the contracted amount ($35,000)

    If the question is what is your rate — it has to be uniform for members of the same plan and it has to be disclosed to the member, or at least what the member is responsible for with relavant caps.

  42. 42

    @Richard Mayhew: Having used my individual plan for some unusual problems, I’m pretty familiar with the math above.

    Restating the question: Does ACA address the incredibly lousy contracted rates that are typically what I’m paying 20% of, as a customer with an individual plan (ie no HR benefits dept beating on the health plan to get my 954 employees a lower contracted rate)?

    The answer appears to be No.

    The incentives, of course, run the other way. The providers who will accept the group contracted rate have every reason to make the exchange plan ‘contracted rate’ = 90% of rack vs the 40% of rack that the group pays. That way the providers are getting paid more from the insured’s pocket, AND the insurer’s portion drives up their spending ratios to comply with ACA requirements.

    That is, the money they use to cover their 75%, or 50%, or 80% of my contracted rate is money that’s going to providers who are getting skinned by the large group rate. So the insurer gets to point to that and say, See, we’re not hurting you SO badly.

    This may represent an improvement for shoppers like me, too. Because I’m moving from a plan they sold 435 of to an exchange plan with 7500 members, there may be some negotiating by the insurer to bring that group’s rate in line with the employer groups. Not sure that’s going to happen, but it might.

  43. 43
    Richard Mayhew says:

    @PhoenixRising: Actually no, most of the Exchange incentives (mainly narrower networks and some insurers who are leaping from Medicaid only contracts to Exchange contracts) push towards fairly low rates as insurers are using Exchange as a means of saying no, or at least saying no in high provider density areas.

    I know for my area, most of the major insurers are offering at least one Exchange product where the pricing to providers is between Medicare +3 to Medicare + 10 instead of commercial pricing which is typically Medicare +25 or more. One insurer has decided to exclude the major academic medical center for a variety of reasons including the fact that the major academic medical center’s base Medicare rates are significantly higher than other hospitals.

    If you are out in the boonies, then the dynamic changes, but again, the subsidy design of basing subsidies on 2nd Silver encourages insurers to say no to high priced providers.

  44. 44
    Original Lee says:

    Thanks again, Richard, for this informative post. Now I understand why my plan only has one pediatric Ear, Nose, and Throat specialist for the entire metropolitan area, and why it took a year to get Original Daughter her ear tubes.

  45. 45
    Stan Gable says:

    This type of network design will make the insurance product less attractive to a person with Type 2 diabetes than a product with a broad, unstructured network.

    This strategy is just blindingly obvious though. If I’m able to weed out diabetics in year 1, then I can guarantee that 1) all of my non-diabetes endocrinologists will rapidly become diabetes specialists after they get bombarded with diabetes patients and 2) all of my competitors will quickly copy whatever I’ve been doing to minimize my exposure.

    I don’t see how this doesn’t hit an equilibrium point within a couple of years and the focus shifts to who is better at cost containment. Disclosure – I’m a long term type 1 diabetic.

  46. 46
    Richard Mayhew says:

    @Original Lee: ENT is a fairly “cheap” specialty to include in a network as there are not many super expensive diseases/conditions that require an ENT. The problem may be the “pediatric” portion of the ENT as there just are not too many of those — and as to why only one was in the network, I don’t know — network manipulation is a possibility but I would not put it in my top three suspicians.

  47. 47
    Richard Mayhew says:

    @Stan Gable: Yeah, these strategies are blindingly obvious, and I picked diabetes as an easy example. The changes in risk pool are minor tilts when dealing with large numbers, and these steps won’t drive out all of the diabetics, just some of them. For any specific disease, equilibriums are roughly reached after a few years, but there is a first mover advantage so these types of tilts will happen for different diseases/conditions on a fairly regular basis.

  48. 48
    Stan Gable says:

    The changes in risk pool are minor tilts when dealing with large numbers, and these steps won’t drive out all of the diabetics, just some of them.

    It would drive them out for a specific period of time but the diabetics aren’t going anywhere. I’d think that if you are good at reducing overall costs by eliminating diabetics, then your price advantage will translate into picking up an inflated number of diabetics during the next open enrollment period, regardless of how unattractive you make your offering to diabetics since cost is such a major driver.

    I realize you used diabetes as an example of a wider trend but the problem is that Type 2 diabetes is just a massive element of costs overall that it’s probably not a great example of anything other than itself. Put another way, you want to weed out your stroke/organ transplant/heart attack candidates? Then find the diabetics first.

  49. 49

    @Richard Mayhew: “…Medical loss ratios…”

    Surely the MLR rule is an incentive to fund overtreatment? Isn’t this, in fact, a problem we already have, and part of the reason why US health care is so expensive?

    I think I’m starting to recognize some elements of old-style public utility economics, here. Basically, the health insurance companies are allowed to collect 15-20% of gross revenues as gross profit and, if their actuaries are any good, they will succeed in this. This is similar to a utility allowed a guaranteed rate of return on investment. So there is a huge incentive to overinvest.

    Do I have this right?

  50. 50
    Richard Mayhew says:

    @The Raven on the Hill: Maybe —and public utilities are probably a good comparison for where health insurance companies are going in the next fifteen to twenty years, but we’re nowhere near that point yet.

    The big difference between health insurance and say the local integrated (generator/distributor) electric company as a public utility is that the electric company was a reasonably natural monopoly within a regional context or at least an extremely difficult nut to crack because of the amazingly high fixed costs needed to build both a set of power plants and the first mile to last mile distribution network. Health insurance in a regional market has barriers to entry (state regulators, network build-out etc) but those barriers are much easier to break through with new entries to a region over a three year span than building a parrallel electric grid.

    MLR might be an incentive to fund treatment in a single year, but long run, it is an incentive to lower premiums if there is a reasonable competetive market.

  51. 51

    @Richard Mayhew: As it happens I am out in the boonies. In fact, we can only see the light from the boonies on cloudless nights out where I live.

    So the key thing I misunderstood was that the subsidies are based on premium amounts, which create the base number from which the spending ratios are figured. And insurers will want to keep rates to providers low to meet those ratios, and therefore structure their networks to be profitable by screwing providers.

    Got it. I’m starting to think we may have to move, on a related note. There was a dearth of providers here, at all levels from PA/NP primary to subspecialists, last year. The insurers, all 4 of them, are going to rachet down reimbursement to the point where some of those docs are going to have to leave NM if they want to pay their student loans off. My family is already in the care of 3 pre-tired (55+, decades of practice in big Eastern cities) providers who work PT now that they can afford to.

  52. 52
    Original Lee says:

    @Richard Mayhew: According to DocFinder, there are 15 pediatric ENTs in my metropolitan area. Only one was in network. There was a six-week wait for appointments and a three-month wait for the surgery. The one we ended up using was not in network, we were able to get an appointment the next day, and were scheduled for surgery within one week. Naturally, we paid out of pocket but were surprised to discover that the anesthesiologist was in network, so we paid less than the estimate.

Comments are closed.