Reviewed and rewritten 30 August 2026. The original 2011 version of this post contained advice about leaving an advance directive where paramedics could find it. That advice was wrong, and the correction is the main reason this rewrite exists.

Most people who mean to sort out their end of life wishes get as far as thinking they should, and stop there. Of the ones who do something, a good number fill in the wrong document and believe they are covered.

So this is about which piece of paper does what, and about the one that paramedics can actually act on.

Three documents, and they are not interchangeable

An advance directive is a legal document. You write down what you would want if you could not speak for yourself. Anyone can complete one at any age, healthy or not, and it usually contains a living will.

A healthcare power of attorney, sometimes called a healthcare proxy or agent, names a person. That person makes decisions for you when you cannot.

A POLST is different in kind. Depending on your state it might be called a POST, MOST or MOLST. It is a signed medical order, written by a clinician after a conversation with you, and it is meant for people who are already seriously ill or frail. Not for healthy adults.

That distinction between a legal document and a medical order sounds like paperwork pedantry. It is the entire point.

The thing the old version of this post got wrong

The 2011 version suggested filling in a worksheet and keeping it somewhere accessible, on the refrigerator, so that arriving emergency responders could find it easily.

The refrigerator part is right. The document was wrong.

Paramedics cannot act on an advance directive. It is not a medical order and it is not addressed to them. A crew arriving to find someone in cardiac arrest, holding a beautifully completed living will that says no heroic measures, will in most circumstances still start CPR, because what binds them is a physician’s order and that is not what they are holding.

A POLST is a physician’s order. That is what EMS can honor.

So if the goal is specifically that responders in your home do the right thing, the advance directive is not the document that achieves it. Someone with a serious illness needs a POLST, completed with their clinician, and the refrigerator is genuinely the right place for it.

I am fairly annoyed that the original post got this backwards, because a person could have read it, done exactly what it said, and been resuscitated against their wishes.

Almost nobody has done any of this

A systematic review pooled 150 studies covering 795,909 Americans and found that 36.7 percent had completed any advance directive, and 29.3 percent had a living will (1).

The number that stopped me was the comparison inside it. Among people with a chronic illness, completion was 38.2 percent. Among healthy adults, 32.7 percent. Being sick barely moves it.

That tells you the barrier is not information and not risk perception. People with heart failure know they have heart failure. The conversation is just difficult and nobody makes time for it.

The part where I disagree with how this is usually taught

Here is the uncomfortable bit.

In 2021, three of the most senior palliative care physicians in the country published a viewpoint in JAMA titled, plainly, “What’s Wrong With Advance Care Planning?” (2). Their argument is that after decades of effort, the evidence does not show that advance care planning reliably produces care matching what patients wanted, and that completing the paperwork has been treated as a quality measure without having earned that status.

It is a viewpoint rather than a trial, and plenty of people in the field disagree with it. I find it persuasive, and it changed how I do this.

The problem with documenting preferences in advance is that real decisions arrive with specifics attached. Nobody writes a living will covering “pneumonia, on a ventilator, eight days in, kidneys failing, neurologist uncertain.” What a family faces is that. What the form says is something more general, written years earlier by someone imagining a situation they had never been in.

What survives contact with reality is the person you named. A surrogate who understands how you think can reason from that in a situation you never anticipated. A form cannot.

So my order of priority runs: name the person, talk to the person properly, and then complete the documents. Most public campaigns run that backwards and treat the form as the achievement.

What to say, and to whom

The conversation people find impossible is easier if you stop trying to specify treatments and talk about thresholds instead.

Useful things to be able to answer: what makes a day worth having for you, what state would be unacceptable to you rather than merely unpleasant, and whether you would want your surrogate to follow your instructions exactly or to use their judgment. That last question is worth asking out loud, because a lot of people assume the answer is obvious and families frequently guess wrong.

Tell the person you named that you have named them. It happens more than it should that someone discovers their role in a hospital corridor.

Then put the documents somewhere findable and give a copy to the surrogate and to your physician. A directive nobody can locate at two in the morning has done nothing.

A conversation that works well by video

I practice entirely by telemedicine, and advance care planning is one of the things that translates better to video than people expect.

Part of it is that the patient is at home, in their own chair, not gowned in an exam room. Part of it is that the family member who needs to hear this can be in the room, or dialed in from another state, without anyone taking a day off work. That is genuinely hard to arrange in person and easy to arrange by video.

What I cannot do remotely is sign your state’s POLST in every jurisdiction, and the rules on that vary. If you are seriously ill and need one, ask whoever manages that illness. It belongs with them anyway.

For patients

If you are healthy, the highest-value thing you can do is name a healthcare agent and have a real conversation with them. The form matters less than the briefing.

If you have a serious illness, ask your clinician specifically about a POLST. An advance directive alone will not direct paramedics in your living room.

Tell your family what you would find unacceptable, not which procedures you would decline. They will face a situation you did not predict.

For colleagues

Completion rates are a poor quality measure and we should stop treating a signed form as the endpoint.

Ask patients whether they want their surrogate to follow instructions literally or to exercise judgment. It takes one question and it changes what that surrogate does under pressure.

And check whether the seriously ill patients on your panel have a portable medical order rather than only a directive. That gap is common and it is the one that produces resuscitations nobody wanted.

The Bottom Line

An advance directive says what you want and names who decides. A POLST is a medical order for people who are already seriously ill, and it is the only one of these that paramedics can act on, which is the correction this post needed. Roughly a third of American adults have completed anything at all, and having a chronic illness barely changes that. The evidence that documentation alone produces the care people wanted is weaker than the campaigns suggest. So name your person, tell them how you think rather than which procedures you would refuse, and then fill in the paperwork.

Related Reading

Can Doctors Help Seniors Live Independently Longer?

Tips to Manage and Organize Your Elderly Parent’s Medications

Emergency Room vs Urgent Care: Which Should You Choose?

When to Use Telemedicine vs. Urgent Care: How I Spot the Sick One

Scott Rennie, D.O.
Board Certified in Obesity Medicine and Family Medicine

This blog is for educational purposes only and does not constitute individual medical advice. Always consult your own physician before making changes to your health, medications, or treatment plan.

Sources

1. Yadav KN, Gabler NB, Cooney E, et al. Approximately one in three US adults completes any type of advance directive for end-of-life care. Health Aff (Millwood). 2017;36(7):1244-1251. PMID 28679811. https://pubmed.ncbi.nlm.nih.gov/28679811/

2. Morrison RS, Meier DE, Arnold RM. What’s wrong with advance care planning? JAMA. 2021;326(16):1575-1576. PMID 34623373. https://pubmed.ncbi.nlm.nih.gov/34623373/

3. National POLST. Learn about POLST forms. https://polst.org/for-patients/about-polst-form/