Narrow Networks Found to Save Money
National health expenditures are rising faster than the economy. Most economists believe this is due to perverse incentives that encourage unnecessary medical spending. More than 30 years ago the RAND Health Insurance Experiment showed that patients spend nearly one-third less when exposed to significant cost-sharing. Yet, health plans continually look for new ways to save money in the absence of a health care market where patients act like consumers.
One such cost-saving method is the increasing use of so-called narrow networks, where health plans restrict the choice of providers in return for lower premiums. Narrow networks are sort of like when you were in high school and your mother took you to Walmart to buy a pair of jeans. She may have refused to shop for jeans at the mall where she knew they would be more expensive. Obama Administration advisor, Jonathan Gruber, wrote about the use of narrow networks in Massachusetts health plans. He found that patients used fewer resources when required to see general practitioners rather than specialty care. According to Gruber:
We found that those who moved into a narrow network plan due to this financial incentive used significantly less care — and paid less for the care that they did use. Overall, spending on health care fell by about 35 percent. Most importantly, this fall in spending was not uniform: we saw an increase in spending on, and utilization of, physician primary care, with large falls in spending on specialists, hospital inpatient and outpatient visits, and emergency room use. This suggests, in contrast to the common arguments that are made against narrow network plans, that they did not lead to a shift away from needed care — rather, they appear to have focused care in the primary care setting, perhaps through more binding restrictions on specialists/hospitals than on primary care physicians. This was true even for the chronically ill patients who were most in need of ongoing medical care.
Contrast this to reference pricing, another cost-saving experiment we wrote about recently here and here. If narrow networks are like restricting the stores where your kids are allowed to shop, reference pricing is like telling your kids they can shop anywhere. But you will only pay, say, $30 towards the cost of a pair of jeans. If they choose to spend $100 on Buffalo jeans at the mall, they must pay the $70 difference.
Several years ago insurer WellPoint partnered with the California Public Employee Retirement System (CalPERS) on a reference-pricing program to encourage enrollees to patronize lower-cost hospitals. It worked — enrollees quickly shifted to hospitals that charged less for joint replacement surgery. Another positive result was that in the process, many of the hospitals that had been charging higher prices lowered prices to remain competitive.
Experimentation with health care payment mechanisms is beneficial. For instance, we have also written about package prices for oncology rather than reimburse physician on a fee-for-service basis.) However, policymakers should not forget that incentives matter. It’s not enough that patients merely know that prices vary from one hospital to another. Consumers must have an incentive to select the lower price facility. Positive incentives are what is usually missing in health care.
It is important to control the costs of health care, for they are directly reflected in premiums.
Unfortunately, with a family group premium around $16,000 a year, once you add in co-pays and deductibles, a family can easily spend $20,000 a year for health care.
That is like having a second mortgage, or first, if you are among the increasing crowd who have to rent.
First and foremost, we have to get a handle on premiums. If premiums are unaffordable, the price of health care is irrelevant.
We need an insurer to devise a plan to lower premiums thru monthly increasing deductibles which are pre-funded.
A pre-funded deductible of $30,000 reduces the premium 60%. A $60,000 pre-funded deductible reduces the premium 80%.
Don Levit
Many years ago, in the 70’s, group plans were sold that paid specific amounts for services. Granted this is Fee for Service, but in this case the fee was specified up front. When you went for care you had an interest in telling your provider that, “My plan pays $60 for an office visit.”
Something like this might serve to lower todays premiums if non-catastrophic care paid up front fixed amounts. Sort of reference pricing in reverse. Schedules expressed as fixed amounts, yet determined as a function of Medicare prices would involve consumers up front. Fee schedules would be set to reimburse perhaps 50% of UCR fees with the balance pouring over to coinsurance after an annual deductible.
Catastrophic medical expenses would be covered in full after a certain out of pocket limit had been reached.
This idea is unrefined, but an attempt to involve consumers of care in the decision about what care is administered.
Thoughts?
As I learned recently, the Australian system works something like this for physician services. If a primary care doctor charges, say, $75 for an office visit, Australian Medicare may pay only $42. The patient is liable for the difference of $33 which they refer to as “the gap.” About 20%-30% of the population buys a separate supplemental insurance policy that not only covers the amount of the gap but also gives members access to private hospitals as compared to the public hospitals which are the only hospitals basic Medicare covers there. The additional access provided by the gap plan can also significantly shrink or eliminate waiting times for non-life threatening surgical procedures and other services that patients have to wait for if basic Medicare is the only coverage they have.
Separately, I couldn’t agree more about the importance of getting the incentives right in healthcare.
In general, do not use any of Professor Gruber’s research about the Massachusetts healthcare system as a reference point. It is always politicized and almost always totally misleading. (My recommendation might apply to all of his research but I only know about how Massachusetts works.)
In this specific case, the research you are referencing is not about “Massachusetts health plans.” It is a study on a sliver of a sliver of a sliver of health plans available to a sliver of a sliver of people in Massachusetts (some Massachusetts state employees choosing among the two or three options they had during their annual open enrollment a few years ago). Starting from this unrepresentative sample of highly paid Massachusetts residents, the study found that only 10% of those eligible chose the limited network option. That’s neither a big endorsement of the limited-network idea or a useful sample of what was already a very small and unrepresentative group of people on which to base findings about PCP use, ERs and all the other blather in his study.
Then this latest example of politicized Gruber research about Massachusetts bases its conclusion on a “distance” factor that it calls crucial. But Massachusetts’ many world-class healthcare facilities (especially the ones that would be used by Massachusetts state employees who are heavily grouped around Boston, the capital city) are all within a few miles of each other. Other states to which Professor Gruber suggests this bad research might apply measure the distance between healthcare facilities in hours of driving time.
(If the political angle is not clear, he is trying to persuade the political class that all the limited networks in PPACA-based insurance is not going to hurt PPACA-based insurance beneficiaries, as many of the beneficiaries themselves have claimed.)
But do narrow networks discourage utilization and thereby reduce costs? Patients used their general practitioners less probably because the doctors were always full and it was hard to get an appointment. That’s how my “narrow network” HMO plan worked out for me. My doctor was a clinic who saw anybody and everybody and those of us who worked couldn’t spend 3 or more hours waiting for a ten minute doctor visit. So I stopped going unless it was very acute situation that I couldn’t self diagnose/treat.