Friday, November 22, 2013

Obamacare and Insurance Companies - the Oped

 
The Affordable Care Act (Obamacare) designers were compelled to make the health insurance companies part of the deal.  No industry so rich and powerful can just be abolished. 

But the risk was high.  The behavior of insurance companies was part of the health care problem.  They profited by denying policies to those it judged poor risks, by revoking policies when subscribers got sick, and by tailoring policies with exclusion after exclusion, among many other tactics.  Under the ACA those practices would be outlawed – all applicants would be accepted at one price, with standard provisions, and no cancellations allowed.  Plus, importantly, profit and overhead rates would now be capped; if exceeded, the companies would need to rebate premium dollars to subscribers.  Companies would retain the power to enter or exit markets at will and to price every product freely, but they couldn’t cherry-pick their subscribers anymore.

The ACA, however, left the insurance company-provider relationship untouched.  The companies for years kowtowed to powerful high-priced hospitals, pharma, and large physician organizations, while dividing and conquering smaller physician groups, particularly primary care.  Thus arrived the increasingly prevalent High-Deductible Health Plans (HDHPs), which will continue under the ACA.  HDHPs represent terrible health policy: higher patient payments, disincentives for relatively inexpensive primary care, but continued support for high priced hospitals, procedures, and end of life intensive care.

With the new rules of the road, economists predicted the insurance companies would compete on the basis of efficiencies, service, and lower prices.  But were the changes enough?  Corporate behavior experts observed that a company’s culture tends to be persistent.  Would the predicted beneficent behaviors indeed emerge, or would other unpredicted ones, sly explorations of design flaws perhaps, take their place?  And also importantly, would the government be able to adjust rules and regulations nimbly as needed? 

The eventual outcome is unpredictable, but the early transitional period is alarming.  The insurance companies at this stage need to avoid catastrophic mistakes, even at the cost of initial market share.  So they cancel grandfathered policies for their own convenience; enter only selected exchanges to decrease risk; and charge high prices for profit safety, even if they might need to rebate consumers at the end of the year.  Moreover, they have created “narrow provider networks,” paying the included providers miserably, even if that leaves out the best doctors and compromises doctor-patient continuity.  Needless to say, the high-priced areas remain untouched. 

Obama erred in giving up so easily on the Public Option, which could have provided a safe haven for many in this interim period with uncertain choices.  He also erred in not providing a transitional plan of incremental changes for those who would be facing higher prices.  That would have been better than ignoring the problem.  He also erred severely in not altering the insurance company-provider axis.  Above all, he erred in not assembling a united, high-quality health care executive team.

While the short-term challenge is coverage, the long-term challenge will be reducing costs in the high-priced areas.  Progress is difficult under the best of circumstances.  Incompetence and venality are constants of life.  The key for progress will be for the government to finally assemble a first-class team, to recognize facts on the ground unstintingly, and to develop a willingness to confront entrenched interests.


budd shenkin

Thursday, November 14, 2013

Obamacare -- What A Transition!

 
Early on, the Obama Administration made a decision that was probably correct, although they probably did not understand how compromising it was.  They decided to keep the vested health interests intact as they pursued reform.

How could they have done otherwise?  Who could be more powerful than the vested interests in health -- insurance companies, hospitals, academic centers, pharma?  It was bring them into the solution, or do nothing.  That’s just reality.

But the bargain could be Faustian.  How could the ACA bring in the health insurance industry and still accomplish reform?  The answer was, while they kept the players, they would change the rules of the game.  Up to that time, insurance companies made money by strict underwriting (declaring many people uninsurable, and others insurable only with very high premiums, and yet others insured with exclusions); by enforcing rescissions (revoking insurance when illness struck); by clever marketing and much small print that ended up denying coverage for care in many instances; by negotiating very low rates of payment with primary care physicians (who are consequently an endangered species;) and by many other nefarious tactics.  Many.  There was insurance company competition (inadequate in many markets, however), but collectively insurance companies had few incentives to reduce prices or utilization of care, because their profit was simply a percentage of premiums. 

In addition, to clinicians like me, health insurance companies became a loathed and reviled industry, both for how they treated patients and how they treated us.

The ACAs calculated that a change in rules would motivate a change in behavior.  The ACA got rid of specific underwriting by mandating community rating and abolishing exclusion for prior conditions.  The ACA also regularized terms of insurance so that they were open and fair to patients and wouldn’t contain hidden small print exclusions.  The ACA also set a limit on insurance company overhead and profit.  The intent was to harness competition so that insurance companies would compete on service and price, rather than who could fool the public the most.

So this was the ACA’s great gamble – could a company and an industry be changed by new rules of the road?  Or, since the companies themselves and the personnel in the companies would remain unchanged, would the essential culture and characteristics of the industry prevail in the end?  Would they find a way to perpetuate their old behaviors even with new rules?

And, to fill out the picture, note this: insurance companies stand between two entities in the health care world, patients and providers.  The ACA would change the rules on the one axis, patient to insurance company.  But the rules on the other axis, insurance company to provider, would not be changed at all.  Sometimes the power in that axis rests on the side of the provider, as with powerful hospitals that would be mandatory to have in a network, or large specialist groups.  But sometimes the power rests with the insurance company, as when they deal with the typical small physician group.  Negotiations on the provider side would remain as before.

Given this bet, which way would you go?  Change the rules and change the behavior, or change the rules and see the ways that culture persists?  The economists typically go with rules and change, looking for “rational behavior under changed conditions.”  Me, I generally go with culture, and given my years of experience in the field, I go with my by now well ingrained mistrust of insurance companies. 

Also, think of this: what does change require?  Certainly, it requires a new set of incentives, which were delivered by the ACA.  But change also requires the use of new techniques, either invented or adopted from elsewhere.  What are the odds that any invention or adoption of new techniques by insurance companies will redound to the benefit of the public, or clinicians, or efficiency?  These companies have been imaginative, all right, but beneficently imaginative?  Not so much. 

Many in the field shared my skepticism.  This was the basis for the “public option,” a governmental plan that would compete against the private health insurance companies.  If the private companies sought to game the system, the public option would keep them honest by its competition. 

There is another argument for the public option that I didn’t see made at the time, but which in retrospect seems obvious: even if the privately competitive system would in the long term deliver an efficient result, what about the short term?  Under conditions of severe change, many insurance companies would be very conservative, seeing as their prime concern that they not make a catastrophic mistake.  Many companies would want to stay out of many exchanges, letting others take the risk and planning to ease in afterwards.  Many companies would also want to charge very high prices at first, even if they had to return some of it to subscribers at the end of the year, just to give themselves a cushion.  Companies would want to take the opportunity to divest themselves of policies that were of poor value to them, and thus would simply discontinue that line of business. 

In the short term, then, for transition, there was an argument for a public option as the transitional object for patients without employers.  But to my knowledge no one thunk it.


What We See Now

We won’t know for some time whether or not the ACA achieves its desired results.  We are now at transition, and what we see is the very short-term effects of the ACA. 

In transition, fear and stories of loss will predominate, and that’s what we are hearing.  Most obvious, of course is the catastrophic mishandling of the Exchanges produced by the incompetent Obama health team – they weren’t great in producing the law; they were horrid in standing up to Republican criticism of “Obamacare,” just horrid, which led to their avoidance of making timely progress in implementation for fear of further Republican talking points; and they have been at least as bad in implementing it (see the Cutler memo at http://www.washingtonpost.com/blogs/wonkblog/files/2013/11/Cutler-implementation-memo-1.pdf.)

Also prominent are the stories of losers, those who will pay more.  The ACA failed to account for these souls who would suffer – there should have been a plan for transition, so that they could have their policies stepped up year by year instead of all at once.  The policies they had and didn’t want to leave benefited them in many instances by not paying for services they would not use – obstetrical care, pediatric care.  They are not now persuaded by the need for a community to support these services much as the general public supports public education.  Chickens of incompetence in both design and messaging coming home to roost.  It is indeed so tragic that Obama never assembled a truly competent health team, and continued to stick it, but change of team is an executive skill that is hard-won.

In transition, we also see on the provider side a continuation of the tactics that have so embittered clinicians against insurance companies.  The insurers have been busy constructing “narrow networks.”  These are plans that enroll as providers on those clinicians who are willing to sign up for 70% or 80% of Medicare fees, which are themselves already less than adequate for primary care.  Thus, with only a few misguided practices signing up as providers for these plans, patients are often prevented from keeping their doctors.  Why?  Because this is what insurance companies do.  The independent practicing clinicians are the weakest political actors in the game – after patients – so they are the weak link that insurance companies go after.  Constructing such networks is the least-risky path for insurance companies, as they keep their costs down and premiums still high.

Less noticed at transition is the continuation of the basic insurance plans that we have gradually been accustomed to – high deductibles and high copayments, leading to high out of pocket costs.  These are very regressive plans that economize in the least inflated of all medical services, primary care and other outpatient care services, while leaving intact payment to the very inflated services of hospitals and procedures – see http://static.squarespace.com/static/518a3cfee4b0a77d03a62c98/t/51dfd9f9e4b0d1d8067dcde2/1373624825901/2012%20iFHP%20Price%20Report%20FINAL%20April%203.pdf.

As fear predominates, the absence of the public option is not mentioned.  In fact, however, that is just what is necessary.  In time it would be expected that private plans would be able to out-compete the public option; but right now, what we wouldn’t give for that safe haven.  If only the Obama team had been able to think that through.

In time, we don’t know what will happen.  It is possible that insurance companies will team up with provider groups, all of whom become more efficient together and are able to increase revenues through higher enrollments with reduced prices and higher quality.  Capitalism generally outperforms socialism.  One can hope, and I do hope.  I’m hoping that my pessimism about persistent culture in companies is proved wrong.  I’m hoping that the whole thing does not just land kerplunk.  I’m hoping that the Obama Administration’s having lain down with dogs does not wake up with too many fleas.  I’m hoping.

After all, there is always hope.


Budd Shenkin

Thursday, October 10, 2013

Administrators, Clinicians, and Professionalism

From “Not Running a Hospital” (runningahospital.com) by Paul Levy, today:
“We still find hospital administrators more interested in market dynamics, mergers, and the like than improving the quality and safety of care.  We still find doctors untrained in the use of the scientific method to achieve clinical processes.  We still see medical schools and residency training programs as woefully deficient in such matters.”
And from an article in Kaiser Health News today:
A recent study by researchers at the Yale School of Medicine found that only a third of 400 elderly patients were discharged with a follow-up doctor's appointment and 25 percent were handed instructions written in impenetrable medical jargon.

Right.  When I read about this, I am outraged.  It gets to the core of professionalism.
I learned this in internship.  On rounds at UCSF, I said to my attending pediatrician that the lab was very slow and unresponsive, and that as a result the patient would just have to suffer.  I was being passive aggressive, and I knew, provocative.
He turned to me with heat  He said, "You will not do this!  You will go down to the lab and get those results personally!"  (The attending was Henry Shinefield, a wonderful man and clinician, with whom I developed a very warm relationship.)
Structurally, he was wrong.  I would be enabling the lab's poor service.  A system correction was needed, and all the clinicians should have gone to administration and demand they get the damn lab fixed.
But his doctor's solution was also correct.  The patient, the individual patient, comes first, and the doctor's job is to get it fixed for the patient, right now.
In fact, to be successful, both views are needed, the clinician's impatience to fix the individual situation, and the the administrator's view to fix the system for everyone once and for all.  
In our 10-office practice, I taught our managers, "Patience is not a virtue!  Insist that things go right, right away.  If you don't get a fix on something, call me directly.  That will be my job."  I taught them not to have an "off my plate" mentality.  It's not off your plate if the problem is simply on some else's plate now -- it's only off your plate when the problem is fixed.
How successful was I at Bayside in teaching organizational professionalism?  I truly don’t know.  I think I changed some of the managers’ attitudes, maybe.  I know they admired my point of view, but how much could they adopt it themselves?  Some did.

How successful was I with the clinicians?  I’m pretty sure they were professionally responsible to their patients, one by one.  How much were they responsible for seeing that a system got changed?  Some, but not so much.  This depends a lot on leadership.  I had quarterly meetings in each office with an agenda that featured systemic problems in the office from both the clinician’s and my own (leadership's) point of view.  After each meeting I made sure to effect improvements within a week, demonstrating to the clinicians that it was a responsible system.  But since I sold the practice I understand that the meetings have been sporadic at best and the Medical Director does not attend.  Leadership is key.

It is hard to overstate the importance of responsibility and persistence as ingredients of success.  As an executive, one thinks: who takes the job seriously?  Professionals take it seriously; that’s what is meant by being a professional.  Economics mixes us up by distinguishing an amateur and a professional by accepting money or not.  But the deeper meaning is the degree of responsibility accepted.  A true professional is someone who figures a problem is off his plate only when a problem is solved systemically.

It is so easy to talk about an "uncoordinated system," but within that system are real people, who are taking responsibility or not.  A system will hold the individuals accountable, but a real top-quality professional will hold not only him or herself accountable, but will hold the system accountable as well.  
As in the above hospital example with poor follow-up and undecipherable patient instructions, where are the professional members of the system headed to the offices of the administrator and the chief physician saying, how can you let this happen?  Where is your leadership in fixing this?  Do we only see individuals doing their jobs with their patients, heading down to the lab to get the results that should have been there at morning rounds, enabling a poor system?  Do we see ER docs readmitting patients, knowing they didn't have follow up appointments, knowing their instructions were indecipherable, and not insisting the leadership improve this performance?  I don't think so.  They might being a professional in the narrow sense, but it’s not enough.  
On the other hand, the profession called attention to the problem some 14 years ago (the Institute of Medicine report on safety), professionals conceived of this study and someone gave a grant for it, and a journal published it.  In my view it's too much time and the cumbersomeness reveals a lack of true interior professionalism -- if everyone were professional in the system the research results would have been different -- but at least we're on the right road.  It's just bumpy and meandering while people die and money is wasted, is all.

Budd Shenkin

Sunday, October 6, 2013

Kennedy

It is JFK season, but even apart from that, I’ve been reassessing him and the era in the last few years, reading accounts of general history of the era, Eisenhower histories, the Stephen King book, and most importantly the 2007 David Talbot book “Brothers.”  My life-long friend Bob Levin has also been immersing himself, principally in the civil rights literature, I think – Taylor Branch, and a book we both read about the Freedom Riders, which was great.

So here is Bob’s current take on the Kennedy Administration:

Those I know who believe most strongly that a vast conspiracy lay behind the assassination of President Kennedy place great emphasis on his commencement address at American university in June of 1963.  They believe these remarks revealed him to be committed to achieving global peace through agreements with Nikita Khruschev and certainly prefigured his intent to end our involvement in Vietnam and, hence, made the CIA, the military, and others decide to murder him.  Never a great believer in conspiracy theories myself, and not a greater admirer of JFK, I decided to see what went on between the time of this speech and his assassination five months later.  For my admittedly non-exhaustive research I turned to the Stanley Karnow book, my only Vietnam reference on hand, and my conclusion is i don't think Kennedy knew what the fuck he was doing with Vietnam.

Shortly after the speech he sent 3000 troops to Thailand because of unrest in Laos.  In early September, he told Walter Cronkite withdrawing from Vietnam would be a mistake.  He tried to get the New York Times to pull out David Halberstam because his reports were hurting the war effort.  He went back and forth about whether to support the coup against Diem, worrying mainly if it would work, not if it was the moral thing to do, eventually leaving it to his ambassador Henry Cabot Lodge to do as he felt best.  
             
The only support I found for the existence of an anti-war attitude in Kennedy was Larry O'Donnell's recollection of JFK's telling him he would pull out troops once he was re-elected, but couldn't do it before without being tarred as soft on Communism.  But O'Donnell was a Kennedy loyalist likely to paint him in the best light, which, in 1970, when he recounted this conversation, would have been to make him a peacenik.  And even if O'Donnell's memory and account were accurate, it still doesn't mean Kennedy would have acted in line with this sentiment.
   
So I see the Am U speech as just political talk.  I think Kennedy was a pol, playing things for maximum advantage, not out of principle.  (Certainly that's how he acted in the South on civil rights.)  Maybe that was enough to get people in the CIA pissed off enough to want to kill him but he was not the figure this other crowd is trying to make him out to be.

I believe Kennedy was essentially an unprincipled politician, telling audiences what he thought they wanted to hear, always seeking to manage events to his and his party's political advantage.  Certainly, that was how his administration conducted itself with respect to the civil rights movement then raging in the south.  And civil rights, remember, was more of an issue than Vietnam was in 1963.  Freedom Summer was about to launch, whereas hardly anyone knew where Vietnam was, and there was no anti-war movement to even speak of.

Then again, as my friend Richard Weber points out, it isn't necessary to burnish JFK's reputation in order to find motivation for the CIA, for instance, to take him out.  He had already pissed them off by firing Allen Dulles and not giving them carte blanche in Southeast Asia.

I myself have a different view:

I went around for years thinking that I liked Kennedy, his panache, the change from the 50's, but I knew the charges that he was essentially a cold warrior, not so progressive on many things.

Then I read the book by Talbot, who polishes everything to a high sheen, and my images of him reunited: he was indeed great, or at least was getting there!  He was fighting the militarists who predominated, but was severely limited in what he could do.  The Talbot book essentially says that he learned quickly, and was in ascent to the gods of right-thinking.  He takes Bobby's ascent in righteousness in the subsequent years as a surrogate for where JFK was going.

That is a speculative view, but an attractive one to reunite my psyche.  And it can't be disproved by his knowledge of political necessity; you can only do what you can do.  I think I'll stick with it.
Budd Shenkin

Wednesday, September 25, 2013

Disillusioned with government

 
There are many reasons and ways to distrust government, God knows.  As a card-carrying liberal, I have always thought of the good things government can do.  But the more I get involved in government programs, the less I trust.

Here is a beautiful example.  What could be more important to kids than vaccines?  Aside from clean water supplies, there have arguably been no measures more important for saving lives.  Public health demands that children be vaccinated.

In this country the great preponderance of vaccines are dispensed by private medical offices.  No problem there – in general, they do a good job.  Enter, government!

The government says, why should patients on Medicaid get vaccinations the way all the private patients do?  Let’s do it our way!  Instead of having local offices purchase the vaccines and charge for them (with overhead included in the charge), let’s buy the vaccines ourselves (it’s called the Vaccines For Children program, or VFC), ship them to the doctors, and have them give the vaccines to the Medicaid kids.  And since the offices don’t pay for these vaccines, let’s not pay them anything except for administering the vaccines – let them order, receive, inventory, refrigerate, etc. for free.  In other words, if an office agrees to see Medicaid children, let’s make sure they lose money on the deal, because they will have to pay for their overhead out of their own pockets.  That’s their reward.

But wait, you say you’re not satisfied, you say you want more for your money – tell ya’ what I’m gonna do!  Not only will you get no payment for overhead, but if you somehow drop a dose on the floor, you will have to pay us the cost of that vaccine.

And that’s not all.  In addition, we will dictate how you will store your vaccine; it can only be in approved refrigerators and freezers, which you have to buy out of your own pocket.

And even that’s not all.  Some states (hello, Connecticut) will also supply “free” vaccines for even those kids covered by private insurance, so the practices can lose overhead on those kids also!

And then, let’s regulate some more.  Let’s make sure that if a practice has a shortage of a certain vaccine for the VFC program, you can’t borrow from your private stock to vaccinate a kid and replace it later, or vice-versa, no matter how good your record keeping.

And there’s even more – let’s mandate that all vaccines have to be stored separately according to who purchased it.  Here is what Graham Barden, terrific public spirited pediatrician from North Carolina has to say:

Just when you thought the CDC's requests could not get crazier.  The CDC thinks it is reasonable to have separate VFC, Private, S-CHIP, 317, State supplied vaccine. Five complete sets of vaccines. The bean counters want to make sure no one can say they are not counting beans!  Paperwork is now the measure of our success.”

As I said, I have always believed in government.  But when I get up close, government doesn’t seem so nice to deal with.

Budd Shenkin

Restrictive Medical Networks

 
We know that insurance companies are cutting costs by narrowing networks for physicians and for hospitals, or “tiered networks.”  Our problem is this conjunction that flows off the tongue, "physicians and hospitals."
To be more concrete, here is how it works for a primary care practice.  You as a primary care practice receive a contract proposal from Blue Shield, say.  You can sign up for network, A, B, C, or D.  If you sign up for Network A, all your charges will be paid for at 70% of the Medicare base price.  Patients who sign up for Network A, which costs the insurance company somewhat less because the fees paid are lower, will pay a lower premium than for other networks.  If you sign up for Network B, you will be paid say 80% of Medicare – a somewhat higher fee – but the only patients who can come to you will be those who sign up for Network B and pay the higher premium.  And so on.
From the patient’s point of view, it will be a juggling act – sign up for a Network and figure out where you can go.  If you already have a primary care doctor, see which Network you can sign up for and not change doctors.  We have seen by experience that many, many patients will sign up for a cheaper plan and make the doctor switch.
I’m not quite sure how it will work for hospitals.  It might be that patients in Network A will need to go to a hospital that signed up for that network, and thus will receive a lower payment from the insurance company.  Or it may be that the plan will be so-called “reference pricing.”  In this approach the patient can go to any hospital, but the insurance will pay the charge given by the lowest (or maybe second lowest) hospital in the wide area, and the patient needs to pony up the rest if he or she chooses a more expensive hospital. 
Now, there is every reason to restrict choice of hospitals.  This is where the money is, this is where the savings will be, and this is exactly where the excessive charges are.  There is plenty of evidence of great variability of charges with no correlation with quality (as far as quality can be measured, but that measurement is pretty good for hospitals as compared to outpatient quality.)  It's hard to narrow this network because of hospital consolidations and local monopolies that have resulted, but it must be done, somehow. 
On the other hand, there would appear to be no reason at all to restrict access to primary care.  Primary care impacts only minimally on the health care budget.  Good primary care, high quality primary care, is generally cost-saving rather than cost-enhancing.  There is no evidence whatsoever of primary care being excessively priced.
In fact, what might happen is the reemergence of so-called Medicaid mills.  One can save money in primary care by having clinics staffed by midlevel professionals – nurse practitioners and physicians assistants – and restricting access to patients by a variety of means.  Decreased quality of care in Medicaid mills is inevitable.  But because of the limited impact of primary care onto the health care budget, virtually nothing will be gained by this approach and much will be lost.  It is well established in health policy that the United States has too little primary care, not too much, and the restricted network approach will only further exacerbate that deficit.
[Note that I have not discussed specialists here, only primary care and hospitals.  Probably the best approach would be to split specialists into proceduralists and non-proceduralists, and treat the proceduralists like the hospitals, and the non-proceduralists like the primary care docs.  There are a lot more imaginative approaches that could be followed, but this might be a simple first step.]
Anyway, that’s my view.  There will doubtless be lots of confusion and some conflict, but if the insurance companies are in charge, as the ACA puts them, there is every reason to expect difficulties and missteps. 
Budd Shenkin

Wednesday, September 18, 2013

Second Opinions

Paul Levy has another interesting post on the value of second opinions.  He says:

Simon Schurr at Collaborative Medical Technology Corporation suggests in this blog post that current levels of overtreatment and inappropriate care could be reduced by more widespread and judicious use of second opinions.  He points to unnecessary surgeries, overtreatment of back pain, mistreating ovarian cancer, and outdated procedures.

His diagnosis: "The causes of inappropriate care are complex, but often the root is simply lack of knowledge, an honest mistake, or a healthcare provider who simply wants to help a patient when treatment isn’t working. Sometimes, profit-driven decision-making or fear of malpractice claims lead to over-testing and overtreatment."

His solution:  "The best approach may be a combination of well-informed patients asking the right questions and seeking top doctors who stay abreast of the latest research, and rigor in using second opinions."


My response:

Here, here!
Years ago I advocated (in the first ever Sounding Board section in the NEJM) for patient access to medical records, partly reasoning that if medical records got a wider circulation, their quality would improve, because clinicians would know that what they wrote could be seen by others.  Now, EMR serves the same purpose, as does the growth of group practices generally, where you know that your colleagues will sometimes see your patients and read your notes and judge.
Likewise, it's possible that even the knowledge that your patient would likely be seeking a second opinion would be very bracing.  When our pediatric practice seemed to have an excessive number of referrals to specialists for one HMO. I announced that I would be reviewing all referral requests from our group before they went out.  Instantly, the number of requests dropped, just from the knowledge that they would be reviewed.  I then had to send back the referral request for reconsideration by the clinician only rarely; just the existence of the review process did the job -- as expected.
Moreover, if you consider Groopman's How Doctors Think, you immediately are impressed with the desirability of second opinions,  to avoid the cognitive error of anchoring, if nothing else.
Here's another example.  A doctor friend of mine had a father-in-law, I think it was, who was sick for years and years without a diagnosis.  By chance, my friend and his father-in-law were walking around the halls of Dartmouth Medical Center and a doctor friend of my friend saw them, walked up to them, and said to the father-in-law, "You have hemochromatosis!"  He could just tell by the look.  After years without a diagnosis.  A single doctor just can't know everything.
Or, consider this.  Just two days ago I was at our health club and a 76 year old friend came up to me.  He had pneumonia almost a year ago and has never really recovered, he thinks.  For the last few weeks he has felt really, really tired.  He's worried -- as he should be.  He asked me about a second opinion -- from an internist, he specified -- because his regular doctor (who is in fact an internist), who he likes, hasn't pinned down anything.  Is his doctor saying subconsciously to himself that my friend Arthur has just run out his string?  Is he out of ideas?  What?  I don't know.  But I urged Arther to in fact get a set of fresh eyes on him -- take up the case from the beginning!  Why not?  Visits are cheap.
Finally, while we have many specialists in many fields, what we really don't have is a super specialist in internal medicine.  When I did my medical school internal medicine rotation at the Beth Israel we had an attending who was revered for his differential diagnosis acumen (Manny something).  His visits were prized, as various residents sought help, and to stump him.  The Chief Residents sought to emulate him.  But what happened to these Chief Residents when they moved on out of training?  Chief Residency was the best training possible for a job that didn't, and doesn't, exist.  You either go to a recognized specialty, or you go into primary care, which really presents a different set of circumstances than Chief Residency.  (For one thing, primary care puts a premium on prevention, which internists generally do not excel at, nor are they well trained for, etc.)
What we need is an institutional setting for the master internist, someone to turn to, or a set of doctors to turn to, who are really smart.  They wouldn't have to give primary care, and their field would feature breadth, rather than the depth of the recognized specialties.  These would be great for second opinions.
I guess you could do the same thing for surgery, but that's a more complicated issue.

And to reiterate: visits are cheap - it's the procedures that are expensive.

Budd Shenkin