Saturday, April 4, 2015

Physician-Assisted-Suicide and Pay-for-Performance: an Unholy Union

Recently, a young woman named Brittney Maynard became a symbol for the Physician-Assisted Suicide (PAS) movement in the United States.  Brittney had been diagnosed with an incurable form of brain cancer- glioblastoma- which had recurred after an initial brain surgery.  She then moved to Oregon, one of three states where Physician-Assisted Suicide is legal by statute, and purchased the drugs which she could use to end her life.  She kept them with her until November 1st, 2014, when she took her own life.  But not before recording testimony to be played posthumously in California legislature where a bill was introduced which would legalize the practice of PAS.  The bill looks increasingly likely to pass, as the California Medical Association which killed previous versions of the bill now looks to stay neutral and sit out the fight due to changing attitudes among its member physicians.

I have serious concerns about the way PAS comes to be implemented, and how it will interact with the statistical measures that are increasingly being adapted throughout the healthcare system to "measure quality".



Let's take two family physicians. Both carry 3000 patients, of which say 30 are terminally ill. Physician A is a great advocate of PAS.  Physician A's standard spiel in these situations is something like this:

"You have stage 4 state breast cancer that has metastasized to the bones in both of your legs and several other organs. If we continue to treat you, that may mean amputation of your legs, not to mention significant intractable nausea and vomiting that is difficult to control. If you would like, we can sign you up for more poison and more cutting. But I suggest that we make you comfortable, treat your pain, and allow you to decide when it's time to pass on peacefully while life is still worth living."

He or she gets all 30 of their patients to choose PAS- and reject further chemotherapy or surgery. To an insurance company (or Kaiser, or Medi-Cal, or Medicaid) these patients have been treated for the cost of a prescription of morphine- $50 each if that.  
Now let's take Physician B, who presents it as an option, but advocates against adapting it.  Physician B's standard spiel is something like this:

"You have stage 4 breast cancer that has gone to the bones in your legs. The odds of survival past a year or so are extremely low. But you know what? People do survive. They say 4% survives to 5 years. Well, what if you're one of those 4 in 100 people? What if we discovered an experimental drug that cures your kind of breast cancer next month? I have a few more chemotherapy drugs left to try- they can cause vomiting and pain, but we are typically able to treat that with opiates and anti-emetics. Now, if you really want, I can give you some drugs that you can use to end your life- but why give up now? Why not try to make one more of your grandkids' birthdays?"

(Disclaimer- the scenario is hypothetical and the numbers are made up, but realistic in some scenarios)

10 patients opt for PAS regardless. The remaining 20 patients receive expensive chemotherapy and surgery. According to the latest figure I remember, Medicare spends an average of $40,000 per patient in the last year of their life. In this case, that comes out to $800,000 spent on Physician B's patients. This does not count the 1 or 2 patients who end up achieving remission for 2-3 more years, requiring yet more sub-specialty care.

Neither physician lied.  Neither physician exerted "undue influence" upon the patient to choose PAS (which is illegal under the bill).  But from the perspective of the payer, which physician do they like better? Which are they going to give a bonus and send as many of their patients as possible to? Rather obviously, Physician A.

This is the awesome and dangerous power physicians hold. Physicians' knowledge enables us to talk 90% of our patients into whatever we want them to do- with the patient believing they made an informed and free choice. Ironically, it is far harder to convince patients to give up smoking or change their diet. But tell a terrified mother-to-be that she needs a c-section or a terrified middle-aged manager that he will likely die without angioplasty, and they will sign anything. I wrote a post on this topic- informed consent- a few months back, and it may be interesting to some.

Insurance companies know this. This was the impetus behind the great push towards health maintenance organizations in the 1990s and capitation payment structures for primary care: these insurance companies wanted doctors to give more speeches like Physician A- not Physician B, and used the power of the wallet to get them to do it. They explicitly paid primary care physicians (PCP) based upon how few advanced procedures their patients got- and required patients to get the okay from their PCP before undergoing any procedure. It worked- HMOs do control costs significantly better. But, in the process they developed a terrible name for themselves. Patients hated the idea that their doctors had a financial incentive to deny them care. Provable negative outcomes like increased incidence of mastoiditis in children due to decreased rates of referrals by pediatricians to otolaryngologists to surgically treat recurrent otitis media led to a backlash and the resurgence of fee-for-service models of care.

Many, many others are trying to co-opt the incredible and subtle power that doctors have to talk our patients into and out of various treatments. Physician decisions collectively determine ~80%(a frequently used number that I've seen) of ~$3 Trillion healthcare spending- something that is not likely to change anytime soon. Corporations, governments, everyone has a large and vested interest in what doctors say to their patients- why do you think so many people want to buy doctors free food all the time? This possible result of PAS legalization that we are discussing is but one perverse consequence of the drive to influence physician decision-making, but it is hardly the only one.

(Ironically, for all the talk of Obamacare death panels, this incentive is not actually present in all ACA insurance plans which face a mandate to spend 80% of their funds on care. These plans have an incentive to get doctors to spend more so that they can charge more and make more in profits, particularly in any area with a paucity of plans offered. But nonetheless, in a competitive environment such as LA or SF, it still somewhat holds)

To summarize, what is the point of this hypothetical? The legalization of PAS may lead to a perverse incentive where doctors are paid more money or otherwise rewarded to encourage their patients to die earlier. If you think it can't happen, consider the experience of the UK NHS and it's Liverpool Care Pathway for the Dying Patient. NHS hospitals were basically ordered put a target number of their patients on this regimen. If they didn't, the hospital was penalized. This led to situations where doctors would put elderly patients on this pathway without talking to either the patient or the patient's family, and all kinds of similar, terrible abuses- some of which were overblown by the media, and some of which were not overblown nearly enough.  The danger is very real in the United States that measures labeled "cost efficiency at the end of life", or "providing patients with all options for end-of-life-care" may end up effectively encouraging doctors to put as many patients on PAS as possible.

We are far away from that scenario. Instead, the opposite happens- physicians often inappropriately encourage further treatment when palliative care is the best option. But nonetheless- I am quite concerned that the legalization of PAS and our system's insane incentives will combine to create a horrific outcome. The public forgives doctors who fight for their patients lives- even when they should not. Our society and culture is wired to respect the one who strives on against the odds- no matter the collateral damage. I do not think the public will be as quick to accept doctors who are perceived to play executioner of the ill.  Nor should they.

What can be done to protect against this?  Simple.  All states which legalize PAS should adapt an iron-clad rule: no organization may contract with, pay, publicly grade, or otherwise measure physicians in any way which can be influenced by the number of patients that physician prescribes PAS to.  Will this be an ironclad stop against abuses of the system?  I think it will be close enough, and will allow patients such as Brittney to take control of the manner of their own death while preventing many more patients from being talked into accepting PAS by someone they trust to have their best interests at heart.