Posts tagged Health care

The health insur­ance sys­tem in the United States is bro­ken. But you already know that. What you may not real­ize is why it is not only bro­ken, but in many ways is not fix­able. At least not with­out chang­ing some of our fun­da­men­tal beliefs about how health insur­ance and our health care sys­tem should work.

To fully grasp this, let’s start by look­ing at insur­ance itself. Insur­ance exists because there are rare but ruinously expen­sive events that hap­pen to peo­ple at unpre­dictable times. The pur­pose of insur­ance is to spread the risk of those rare but ruinously expen­sive events across a large enough group of peo­ple and a long enough time that the cost per per­son per unit of time becomes bear­able. For most of the insured, this is a los­ing propo­si­tion. That is, even exclud­ing admin­is­tra­tive and profit costs, you are likely to spend more on insur­ance than you will receive in claims. The more cat­a­strophic and rarer the insured event, the more likely this is true.

And yet we buy insur­ance any­way, mostly because the cost to us of the rare event is so great that we are unwill­ing to accept the con­se­quences. This is com­pletely ratio­nal and expected behav­ior. There’s an entire arti­cle that can be devoted to a dis­cus­sion of how insur­ance relates to the spec­trum of peo­ple from risk-​​averse to risk-​​seeking, but that’s far more involved than nec­es­sary for this arti­cle. The Wikipedia arti­cle on risk aver­sion is quite good, though.

Now let’s look at health care. We treat it as if it were this mono­lithic beast, when in fact it con­sists of sev­eral loosely related sep­a­rate parts. I’ll describe them here, and you’ll see what I mean.

• Cat­a­strophic exter­nal events: These are things like car crashes or earth­quakes caus­ing book­cases to fall on you. They’re essen­tially unpre­dictable, and the cases that mat­ter most require expen­sive med­ical attention.
• Cat­a­strophic inter­nal events: These are things like aggres­sive can­cer or degen­er­a­tive dis­eases. These are, to vary­ing degrees, more pre­dictable than the cat­a­strophic exter­nal events, and many of these ben­e­fit from early diag­no­sis and treatment.
• Chronic con­di­tions: Asthma and some types of dia­betes go here. Treat­ment of these is focused on the symp­toms, as the under­ly­ing dis­ease is not curable.
• Com­mu­ni­ca­ble dis­ease: I’m not count­ing things like the com­mon cold. This is more about dis­eases for which treat­ment addresses the under­ly­ing dis­ease and/​or pre­ven­tion of acquir­ing the dis­ease in the first place.
• Gen­eral main­te­nance: This includes both asymp­to­matic and symp­to­matic check­ups, where there is not a clear indi­ca­tion of a par­tic­u­lar dis­ease. Non-​​communicable infec­tions and minor com­mu­ni­ca­ble dis­eases like the com­mon cold also apply here. Basi­cally, this is the stuff for which you go to a Gen­eral Practitioner.
• Elec­tive ser­vices: Most cos­metic surgery falls under this cat­e­gory, but so do a few other things. There’s room for dif­fer­ences of opin­ion on many of these, such as gas­tric bypass.

One prob­lem with insur­ing health­care is the notion of a total loss. In tra­di­tional insur­ance, there is a known max­i­mum value for an event. In the case of auto­mo­bile insur­ance, there is a max­i­mum lia­bil­ity in the pol­icy, and col­li­sion will cover a max­i­mum of the cur­rent value of the vehi­cle, at which point it is con­sid­ered a “total loss.” Health insur­ance has attempted to accom­plish a sim­i­lar result with life­time caps (which are no longer legal). But peo­ple are under­stand­ably squea­mish about assign­ing a dol­lar value to their lives. Nobody wants to be told that they or their loved ones are going to die due to a lack of money. This imme­di­ately changes the insur­ance equa­tion, because pre­mi­ums need to be sig­nif­i­cantly higher if the poli­cies have no caps on claims.

Another prob­lem with insur­ing health care is most of the above cat­e­gories don’t all fit well with the tra­di­tional notions of insur­ance. Cat­a­strophic exter­nal events are the only sort of sit­u­a­tions for which clas­sic insur­ance is a per­fect match. Gen­eral main­te­nance would be silly to cover in the case of auto­mo­bile insur­ance (imag­ine insur­ance cov­er­ing oil changes), but for cat­a­strophic inter­nal events, the cost to the insurer is often sig­nif­i­cantly lower if dis­cov­ered early, which means that it’s less expen­sive to cover gen­eral main­te­nance if it results in early dis­cov­ery of cat­a­strophic inter­nal events. But gen­eral main­te­nance will rarely uncover these cat­a­strophic inter­nal events, so most of the money spent on gen­eral main­te­nance goes to areas not typ­i­cally asso­ci­ated with insurance.

Of course, full cov­er­age of gen­eral main­te­nance with­out co-​​pays tends to lead to the insured overus­ing med­ical ser­vices. If it’s free, there’s lit­tle incen­tive to err on the side of fru­gal­ity. On the other hand, hav­ing co-​​pays tends to lead to the insured under­us­ing med­ical ser­vices. Since many dis­eases are much less expen­sive to treat when they’re either asymp­to­matic or mildly symp­to­matic, peo­ple with co-​​pays are more likely to remain undi­ag­nosed until the dis­ease has pro­gressed beyond the cheap-​​and-​​easy-​​to-​​treat stage. Since most Amer­i­cans aren’t doc­tors, and since med­i­cine is so com­plex today, few of us are able to ade­quately deter­mine the ideal time to see doc­tors for diag­no­sis and treat­ment. Basi­cally, gen­eral main­te­nance is simul­ta­ne­ously well-​​suited and poorly-​​suited to insurance.

Chronic con­di­tions and elec­tive ser­vices are badly matched to insur­ance. Chronic con­di­tions tend to have rel­a­tively pre­dictable costs, which are fairly con­stant over time (think asthma inhalers or birth con­trol pills). Elec­tive ser­vices are, by their very nature, entirely pre­dictable, since they are entirely con­trolled by the per­son receiv­ing the ser­vices. The nec­es­sary over­head of insur­ance makes cov­er­age of chronic con­di­tions and elec­tive ser­vices more expen­sive than sim­ply pay­ing out of pocket. But many types of elec­tive ser­vices are hard to dis­tin­guish from nec­es­sary treat­ments. At what point does recon­struc­tive surgery become elec­tive, for example?

Com­mu­ni­ca­ble dis­eases are worth­while to insure against, but work best when every­one is cov­ered. This is the same sort of sit­u­a­tion as we have with fire pro­tec­tion (not fire insur­ance). Just as it is less expen­sive to put everyone’s fires out as quickly as pos­si­ble than to have patch­work cov­er­age, where some peo­ple are cov­ered and oth­ers not, it is less expen­sive (and causes fewer deaths) to con­trol com­mu­ni­ca­ble dis­eases as quickly as pos­si­ble than to treat some peo­ple and leave oth­ers untreated. For com­mu­ni­ca­ble dis­eases, then, the best model is not insur­ance in the tra­di­tional sense, but rather uni­ver­sal cov­er­age. Often these dis­eases are diag­nosed at gen­eral main­te­nance appoint­ments, fur­ther cloud­ing the appro­pri­ate­ness of gen­eral main­te­nance coverage.

So, some health­care is well suited to indi­vid­ual poli­cies, some to uni­ver­sal cov­er­age, and some to no cov­er­age at all. And not all sit­u­a­tions fall neatly into a sin­gle cat­e­gory, as evi­denced by the blee­dover between cat­a­strophic inter­nal and main­te­nance. Already it’s a mess. But it gets worse.

By far, the major­ity of Amer­i­cans get their health insur­ance from their employ­ers. This means that they have lit­tle or no choice about the cov­er­age they receive, and from whom. Few of us have the lux­ury to choose our employer based on the specifics of the health insur­ance pro­vided, and even if we did we prob­a­bly would con­sider other fac­tors to be more impor­tant any­way. It’s silly to be in that posi­tion in the first place. Imag­ine if the food we were allowed to eat was deter­mined by the make of car we drive. Health cov­er­age and employ­ment are an equally absurd match, from a purely objec­tive per­spec­tive. I am aware of the his­tory behind it, but that hardly makes this model wor­thy of perpetuation.

Now, since most Amer­i­cans don’t get to choose their health insur­ance providers, this means the providers have lit­tle incen­tive to please the con­sumers of their prod­ucts. Instead, they need to please their con­sumers’ employ­ers, and these employ­ers’ needs may well be at odds with those of the insured. For exam­ple, an employer will typ­i­cally want employ­ees to be healthy, because healthy employ­ees are more pro­duc­tive than unhealthy ones. How­ever, once an employee has been diag­nosed with a ter­mi­nal degen­er­a­tive dis­ease, it is now in the best inter­est of the employer for the employee to not have the cov­er­age at all, and fur­ther­more to no longer be employed, because the employee has become a pure lia­bil­ity. Even an employee with, say, a child with mus­cu­lar dys­tro­phy would likely be a net lia­bil­ity to the employer.

In other words, we’re get­ting insur­ance for many things that aren’t well-​​suited to be insured, from com­pa­nies that don’t need to sat­isfy us, paid for by peo­ple whose inter­ests match ours, when they do, only by coin­ci­dence. Given that, it’s remark­able that our sys­tem works as well as it does!

And then we have some sce­nar­ios for which the cur­rent sys­tem is com­pletely use­less. For exam­ple, it has often been sug­gested from the right that young peo­ple (say, in their 20s) don’t need to be insured at all, or need only cat­a­strophic cov­er­age cou­pled with a Med­ical Sav­ings Account (MSA). This pre­sumes they will not be sub­ject to cat­a­strophic inter­nal events. Granted, the like­li­hood of a cat­a­strophic inter­nal event is rel­a­tively low, but what is our plan for those cases where they occur? As I noted above, some­one with catastrophic-​​only cov­er­age + MSA is unlikely to go for main­te­nance, which makes the prog­no­sis much worse and the cost much higher upon diag­no­sis of a cat­a­strophic ill­ness. Is a 22 year old going to be suf­fi­ciently well informed to decide the appro­pri­ate level of insur­ance? I’m not con­vinced any­one is, since humans are noto­ri­ously bad at gaug­ing risk. But those in their 20s are par­tic­u­larly unlikely to have the finan­cial where­withal to sur­vive even one such instance, fis­cally speaking.

So what do we need to do in order to fix this? Any­one who says they know the answer is either mis­in­formed or lying to you. There is no per­fect solu­tion, because med­i­cine today is more com­plex and expen­sive than ever before, and we must make some hard choices. No mat­ter what, we have to give up at least one of the following:

• Emer­gency care for all, regard­less of abil­ity to pay
• Max­i­miz­ing the finan­cial effi­ciency of med­ical care (i.e., most bang for the buck)
• Uni­ver­sal cov­er­age against com­mu­ni­ca­ble diseases
• Afford­able coverage
• Cov­er­age of all treat­ments, regard­less of cost

What we’re left with is a choice of which of those bul­lets are most impor­tant. This is where the dis­cus­sions should start, because which fea­tures you choose will quickly lead to the opti­mal solu­tion for that set of features.

One issue we have as a nation is an inabil­ity to have hon­est trade­off dis­cus­sions. Often, the con­ver­sa­tions devolve based on an assump­tion that there is an easy way out, that there is so much money being wasted that we can get every­thing we want if we just focused on improved effi­ciency. From per­sonal expe­ri­ence, I have never seen some­one present that argu­ment with­out being intel­lec­tu­ally lazy about it. That is, when pressed to explain pre­cisely where the inef­fi­ciency resides, the amount of waste described is always orders of mag­ni­tude smaller than the amount needed in order to close the fis­cal gap between where we are and where they want us to be.

For exam­ple, tort reform> is often, espe­cially from the right, described as a way of elim­i­nat­ing much waste from health care. But mal­prac­tice insur­ance costs have not risen over the past cou­ple of decades, and elim­i­na­tion of all mal­prac­tice judg­ments (which I can’t imag­ine any­one endors­ing) would reduce our med­ical expenses by only 2%. Sim­i­lar issues arise when the left points the fin­ger at insur­ance com­pany prof­its. I’m not argu­ing that these aren’t worth look­ing at, but they are very, very small com­pared to the rise in health care costs over the past decade.

Sim­i­larly, what is com­monly referred to as “defen­sive med­i­cine” is far less tied to mal­prac­tice costs than it is to the “Cov­er­age of all treat­ments, regard­less of cost” bul­let point. We want to make sure noth­ing is missed, and our doc­tors know this. Defen­sive med­i­cine is the nat­ural result when the direct cost to us, the con­sumers of med­ical ser­vices, is low rel­a­tive to the actual cost of the test­ing (because insur­ance picks up most or all of the tab), and doc­tors want to keep us happy as patients, and doc­tors don’t lose any money if patients have more tests. The “sys­tem” (usu­ally our employ­ers, ulti­mately) bears the bulk of the costs, and we as patients get all of the ben­e­fits. Is this some­thing we want to change? I know a cou­ple of peo­ple who, due to defen­sive med­i­cine, were diag­nosed with can­cer. Nei­ther would have been cor­rectly diag­nosed with­out the defen­sive test­ing. Both of them would have died if they weren’t diag­nosed until they were suf­fi­ciently symp­to­matic. While I rec­og­nize that this is anec­do­tal, I can’t help but won­der if any of us are so firm in our beliefs about cost reduc­tion in health­care that we are will­ing to die for those beliefs.

I’ll dis­cuss pos­si­ble solu­tions, and the Patient Pro­tec­tion and Afford­able Care Act (com­monly referred to as the “Health Care Reform Bill” or pejo­ra­tively as “Oba­maCare”) in par­tic­u­lar, in a future arti­cle. In the mean­time, I’ll leave you with a few questions:

• Which of the five bul­let points are you will­ing to do with­out, and why?
• Are you will­ing to accept fewer tests in exchange for lower insur­ance pre­mi­ums, but at a cost of higher risk of death?
• If you see inef­fi­cien­cies in the sys­tem today, which ones are big enough to make a seri­ous dent in the cost of med­ical care in the US if we addressed them?
• Which of the cat­e­gories of med­ical care should be pro­vided by health insur­ance, and why?
• Do you think health insur­ance should be pro­vided by employ­ers at all? Why or why not?