Death, Dying, Euthanasia and Physicians

shawnwhatley.comDying people show up each day in my emergency department.  Young and old take their lives, or have life taken from them.

Death is everywhere, if we look.

Recently, death and dying holds media attention:  legislation in Quebec, Dr. Low’s passing, and even Sue Rodriguez.

Aside from places like the Netherlands, physicians have almost no experience with euthanasia or assisted suicide.

Doctors know end of life care, not euthanasia; palliative care, not physician assisted suicide.

Dr. Wooder, president of the OMA, was wise to focus on what physicians know best, when he decided to focus on end of life care this year.

Emotional stories and tragic personal experiences fill most discussions about death.  We wade over our heads into content and debate unfamiliar beyond anecdotes.

When we don’t know, we tell stories.

We need to start with definitions.  As President Clinton said, “It depends on what the meaning of the word ‘is’ is…”

Six terms you need to know:

Euthanasia:  few agree on a definition, but most say that “euthanasia involves doctors making decisions which have the effect of shortening a patient’s life and that these decisions are based on the belief that the patient would be better off dead” (Keown, 2005).

Active, physician-assisted suicide (PAS):  you end your life.  You push the button, take the drug, or start the infusion.  Your doctor gets you started with supplies and instructions, but you end it.  If PAS doesn’t go far enough, your doctor might have to perform voluntary, active euthanasia.

Voluntary, active euthanasia (VAE): you let your doctor end your life.  You give consent; your doctor pushes the button…

Non-voluntary, active euthanasia (NVAE): your doctor ends your life when you are not competent to give consent due to severe disease, dementia, etc.

Involuntary, active euthanasia (IVAE):  your doctor ends your life against your wishes.

Passive euthanasia (PE):  PE involves withholding or withdrawing of medical treatment by a doctor with intent to kill.  Defining what includes “medical treatment” is tough and needs its own blog post.  Are tube feeds medical?

Finally, we need to grapple with: Intent – aiming to have something occur, and Foresight – being aware that something might occur.

Most of this post came from two great books:

Euthanasia, Ethics and Public Policy: An Argument Against Legislation by John Keown

Rethinking Life and Death: The Collapse of Our Traditional Ethics by Peter Singer

These books look from opposite ends of life.  Keown writes clear, careful prose and offers an in-depth review of the Danish experience of euthanasia.  Singers writes engaging, thoughtful philosophy about difficult cases in support of unpopular ideas like involuntary euthanasia.  On the back of Singer’s book, the Washington Post blurb says:

Far from pointing a way out of today’s moral dilemmas, Singer’s book is a road map for driving down the darkest of moral blind alleys…Read it to remind yourself of the enormities of which putatively civilized being are capable.

Keown reminds us that, “Hard cases make bad law“.  So far, most of the news serves up ‘hard cases’.  We need to move past these and start discussing how most people die and what we are doing to improve that process.  

What do you think?  Can we have an adult discussion about death?  Should we let everyone do whatever they want with their lives and how they end?

Physician Autonomy: an Historic Relic?

Physician autonomy was sacrosanct; even enshrined in the Warsaw declaration.  

Lately, it’s negotiable.

Ballooning costs of care and medico-legal risk make more doctors call for clinical practice guidelines (CPGs) to protect them, simplify complex decision making, and justify not ordering unnecessary tests and treatment when patients demand them.

Looking for something solid on which to make clinical decisions, physicians elevate evidence based medicine as a weapon against bureaucrats’ cost cutting and so-called expert opinions.  But, the weapon also slices off physician autonomy.

Battle-lines form, with cost on one side, autonomy on the other, and evidence held hostage by whoever can show it best supports them.  Patient centeredness, another hostage, usually hangs from physician banners.  National associations weigh in; things get nasty.

When physicians call for CPGs, are they calling for limits to their autonomy? 

Doctors usually say, “No.  We can ignore the CPGs when indicated.”

If so, it seems CPGs afford little protection from lawsuit.  If standard care is to ignore CPGs when indicated, how can doctors rely on CPGs to decrease medico-legal risk?

Having CPGs, but ignoring them at will, seems to be the worst of both worlds: loss of autonomy without decreased risk.

While the aroma of clinical autonomy lingers, any trace of operational autonomy in hospitals or large groups disappeared long ago.  Even so, some physicians discussing system decisions still say:

Every physician should be involved in every decision

Every change idea should be shared at the earliest possible moment with every MD

Consensus with all physicians should be reached before any change

Are we in a post-physician-autonomy age?  How does this impact professionalism?  Is autonomy something that’s earned or protected?  Is autonomy the wrong question?

Responsibility, accountability, and autonomy are inseparable.  Some want to remove physician responsibility  as a way to decrease physician influence; others want to remove responsibility as a way to decrease risk.  Some insist on keeping accountability, but work to remove responsibility and autonomy.

Should we forget about physician autonomy and only ask what’s best for patients?

Medical Error: System Solutions vs. Blame

Blame and Shame
Blame and Shame

“If you guys can’t think to order pregnancy tests, we’ve got big problems!” the consultant said.

He was a heart expert, but he remembered medical school:  women of childbearing age with abdominal pain and/or bleeding must have pregnancy tests.

We forget simple things, make simple cognitive errors, all the time.

Intelligence or experience will not protect you.

Every day brilliant people forget to:

  • close their zippers,
  • signal lane changes,
  • turn off the oven,
  • feed the cat,
  • lock the door

Medicine overflows with cognitive traps. It brims with safety-critical activity: you get only one chance to do something right or patients will suffer & die.

For example, a missed ectopic pregnancy can cause a woman to bleed to death.   An ectopic exists in 6-16% of women with pain and bleeding in the ED.  Every medical student learns early that ordering a pregnancy test can save lives.

But very rarely, blood-work gets ordered without a pregnancy test when providers are left to order lab tests separately.

The consultant believed emergency docs must be even more stupid than he suspected.  How could they miss something every medical student knows?

Uninformed individuals still blame people; experts tell us to blame the system.

Errors do not happen for lack of knowledge

We must improve the system and stop the ‘blame and shame’.

System solutions – 2 examples:

Behaviour-shaping constraints, or forcing functions.  For example, you cannot get your bank slip from a bank machine without collecting your bank card first; you cannot start your microwave without closing the door first; you can’t start your car without putting it into park. Constraints prevent medical mistakes.

Opt out vs. opt in: these terms get used in advertising and mailing lists.  To capture everyone for a list, advertisers put people on the list until they ask to be taken off.  Advertisers assume you want to be on the list, opt in is assumed, until you opt out.  (Opt out pops up in discussions about ways to increase organ donation)

Using opt out for lab panels for women of childbearing age with abdominal pain could decrease the chance of forgetting a pregnancy test.  Providers could opt out of performing the pregnancy test, but would be forced to think about it first.

Some still resist system solutions and try to blame people for medical errors.  What do you think?  Can we rely on system design to decrease error?