The Upshot reflects: Patients overestimate benefit and underestimate harm of tests & treatments

In February, an excellent systematic review article appeared in JAMA as part of their "Less is More" series. Patients’ Expectations of the Benefits and Harms of Treatments, Screening, and Tests: A Systematic Review by Drs Tammy Hoffman and Chris Del Mar provides a comprehensive overview of patients views of the risks and harms of various tests, procedures, and treatments.

Their Conclusions and Relevance section explains the take-away it nicely:

The majority of participants overestimated intervention benefit and underestimated harm. Clinicians should discuss accurate and balanced information about intervention benefits and harms with patients, providing the opportunity to develop realistic expectations and make informed decisions.

This week, the NY Times Upshot - a source I'm finding myself reading more and more often - posted their reflection on the article, If Patients Only Knew How Often Treatments Could Harm Them. They beautifully wove together some of the data from the article to make it easier to digest and understand.

For example, they highlight the 2012 Annals of Family Medicine study that looks at patient estimates of the benefits of screening for bowel cancer. 

94% of patients overestimated the benefits of bowel cancer screening.

Simple and persuasive examples like that help explain the problem with our current care, and the article culminates in a summary written by Drs Frakt and Carroll that is completely in line with the principles that drive the Less is More in Medicine approach:

Many of the studies in the systematic review show that people report that they would opt for less care if they better understood benefits and harms. Improved communication could better serve patients and might improve the efficiency of our health system if patients focus on getting the types of care for which the benefit outweighs risk of harm.

Since they've done such a great job expounding the article, I feel no need to provide my own explanations or reflection.

 These kinds of articles come into my email and RSS reader and across my twitter landscape in droves; being overrun with articles and action in the field of overdiagnosis/testing/treatment is a delightful problem to have.


Demanding Patients? Not so in Oncology

Surveyed physicians tend to place responsibility for high medical costs more on “demanding patients” than themselves. However, there are few data about the frequency of demanding patients, clinical appropriateness of their demands, and clinicians’ compliance with them.

Exactly. This JAMA Oncology paper looked at 5050 patient-provider encounters in the oncology context and found that patients requested things in 8.7% of the encounters, and these demands were only considered inappropriate in 1% of encounters.

Number and Types of Patient Requests or Demands (JAMA Onc)

Number and Types of Patient Requests or Demands (JAMA Onc)

I think we need to be very careful about blaming patients. I do it... but I'm getting better at seeing the bigger picture. Yes, sometimes they are in the stage of denial and struggling to cope with their diagnosis. They may ask for completely inappropriate tests or treatments. Sometimes their expectations are absolutely ridiculous but most of the time this is not the case. The patient is not crazy or 'demanding.' A lot of the time it is we clinicians who put some of the more unrealistic expectations on people's radar.

Educating patients wouldn't change this, except if we can encourage them à la Choosing Wisely to initiate discussions with their physicians about unnecessary tests and treatments.  Educating the clinician, particularly encouraging transparency and openness in communication is really important. However, the biggest thing we can do is to change the overall culture of the health 'system' and our society to make it "okay" to talk about these issues frankly.

I work quite frequently with oncology patients, often in a supportive or palliative role. I find it shocking that many of them have never discussed dying, have not made advanced care plans, and do not understand the goals of their treatment. Often a patient is receiving palliative therapy and yet they believe it is a curative therapy. They may demand aggressive medical treatment, not realizing that they are very close to dying.

Sometimes, when I liaise with the oncologist, he or she explains that they had frank discussions about these things, and I can see it in the notes. It's just been hard to accept and people don't really hear what has been said. Other times, "it just never came up." I find that that hard to believe. It should come up. Shouldn't it?

Not talking about the end of life is doing a patient a disserviceHow can they make decisions about their care without knowing what is going on? It also suggests - as made clear by this study -  that inappropriate interventions might be coming from the clinician, not necessarily initiated at the behest of the patient.

What do you think?


Minimally Disruptive Medicine: Thinking differently about nonadherence

In a follow up to Disutility: Finding the balance between benefit and hassle, I present this video from the North American Primary Care Research Group  (NAPCRG) Annual Meeting.

The answer to healthcare is education. Nevermind the other aspects of their life, nevermind that they have multiple diseases, side effects of medications, and not enough time in the day to do all the health 'work' that we give them . Teach patients, yell at patients, scare them into doing what you (the doctor) says. And if they don't take responsibility and do it, then... fire them as your patient!

Or not.

Dr Victor Montori, champion of Minimally Disruptive Medicine, explains a radical new way to think about "nonadherence" and the work that we give our patients to do.

NAPCRG Plenary I: Minimally Disruptive Medicine; Victor Montori, MD


Must Watch VIDEO: (TEDMED) Are zombie doctors taking over America?

I don't know how I hadn't already seen this. It is perfect.

I've been feeling lately that it is impossible in this system to deliver good quality care and to be 'happy and well and fulfilled' myself, I can related to almost everything Dr Zubin Damania (aka ZDoggMD) says in his TedMed talk.

For years I've enjoyed ZDoggs hilarious videos but for some reason I did not see his TED talk until today.

Physician Zubin Damania, Director of Healthcare Development for Downtown Project Las Vegas, has a plan to fight back against a system that can dehumanize doctors and patients alike.

So much of what he says about his journey into medicine resonates with me, but this one section captures it all perfectly:

"A typical day in my life went something like this:

It begins with a mad scramble from room to room seeing complicated, sick and scared patients and their families, followed by three hours of paperwork using an electronic medical record that's two keystrokes away from becoming sentient and destroying humanity, all the while answering phone-calls from insurance companies, colleagues, and administrators each asking that I do more and more with less and less while improving quality and patient satisfaction. And speaking of which, that patient for whom I refused Vicadin because I was convinced it was destroying his life just ripped me a new one on Yelp! And I get through my 31st voice mail and I'm trying to delete it and it won't delete and I'm thinking I have to call I.T. now and I realized the reason it won't delete is that it's a live person I'm talking to. And it's my wife. And she's asking me to come home.

When I finally do get home, I can't be present with my two little girls because all I can think about is 'could I have missed something in this insane shuffle that could hurt or kill somebody?' Later, rinse, repeat for an entire career."

"Enough was enough. I realized I had a choice to make. I could either submit to this insane system and retire and then do my thing, but have my soul crushed and join the ranks of the swelling physician-zombie apocalypse. Or, I could surrender.

Just give up the external validation, drop the mask, and really surrender to who I actually was and have some hope of reconnecting to my purpose."

I need to do that. Dr Damania and Iora Health understand the fundamentals of providing the right kind of care. If we weren't wasting so much time and energy on all kinds of things that don't really matter, we could start to care for people again.

How can we do it?

- remember "medicine is about human relationships"
- "fix the reimbursement system . . make it an affordable, flat fee that incentivizes everyone to keep everybody well."
- "revolutionize the care model and the culture of the team; health coaches, nurse practitioners, doctors, and patients working together collaboratively non-hierarchically to change the health of the whole community"
- "glue it together with technology that helps rather than obstructs
- "do things FOR people instead of TO people"

I want to be a part of a clinic like that!


Lown Institute Road to #RightCare conference March 8-11, San Diego

I'm sharing this invitation to remind you to register for the Road to RightCare conference. It will be a fantastic look at how to operationalize solutions to overdiagnosis, undertreatment, and the gamut of associated problems. I'll be there too!

If that isn't reason enough, it is a great city! As I'm currently in San Diego, I can tell you that there's plenty to do in you spare time (eg. today we are renting bikes to explore Coronado Island and check out the beach; later we'll taste some great Mexican food and sample some local microbrew). 

With the holidays around the corner, we want to remind you to register for the Lown Institute third annual conference: 

Road to RightCare: Engage, Organize, Transform 
March 8-11, 2015 
Omni San Diego Hotel

Among the many great speakers featured at the conference, you don’t want to miss the opportunity to hear our dynamic keynote speaker, Diane Meier, MD, director for the Center to Advance Palliative Care, Mt. Sinai Health System. If you’ve never heard her speak on a topic of great importance to all of us, here’s a sneak peak

Dr. Meier is one of the many speakers, panelists and attendees who will be addressing our culture of medical overuse in the U.S. and proposing solutions to eliminating it. 

We’re finalizing our meeting agenda and will be sending you weekly updates. We promise that this will be the best conference you’ll attend all year. 

To learn more, see the agendaRegister now. 

. . .

Thank you for supporting and participating in our work. We look forward to seeing you in San Diego. 


Vikas Saini, MD
Lown Institute

Shannon Brownlee, MSc
Senior Vice President,
Lown Institute


Register Today! 
Click here to visit our website, review the agenda and register. 

Registration Deadline
 February 10, 2015

Good + Good + Good = Bad? Recognizing the Harms of Polypharmacy


This month, Dr Roger Ladouceur writes in the Canadian Family Physician (CFP) about a 65 year old patient of his with many co-morbidities. He treated her with guideline-based care, resulting in her taking 16 medications and perhaps, as he solemnly suggests, in someone's death.

Not hers.

We will never know, but polypharmacy may have been a factor in why her vehicle careened across the median and struck a pedestrian who was crossing the street, eventually killing him.

It is noble of Dr. Ladouceur to draw attention to this case; it takes a brave person to consider error, reflect meaningfully, and to move forward with purpose. Trying to help this woman achieve health is not a "mistake" by any means, however with the opportunity to take a step back, he saw that in trying to help her, he may have caused harm.

Good intentions are essential to providing good care, but with everything we do, we may cause harm. Polypharmacy is a perfect example of how evidence and guidelines can compound to create an untenable state. As he writes:

I am not the only physician to prescribe so many medications; most of us often prescribe a substantial number. This is not about assigning blame; we are following the recommendations. Each medication is justified for the indication for which it is prescribed. But what happens when they are all taken together, even when there are no drug interactions?
Could it be that, for a man who went for a walk after breakfast, prescribing 16 medications was fatal? Could it be that, sometimes, we do too much?

Read the very thoughtful article here.

Less is More in Healthcare (an evening with Dr Iona Heath)

Speed. Need. Greed.

Dr Iona Heath came to Vancouver, BC to talk about When Less is More in Healthcare, thanks to efforts by many, and particularly Dr Tracy Monk, of British Columbia's Center for Relationship Based Care.

I was humbled by Dr Heath's first slide: A screenshot of my website. That's right. Jaw-dropped, cheeks rosy, I stared at the front page of Less is More Medicine in all its glory on the screen at the front of the room.

And it got much better from there. Dr Heath wove clinical examples with philosophical perspectives to make the case for Less is More in Healthcare. Her essay along similar lines - The Art of Doing Nothing - can be read here.

To us she spoke about the barriers to achieving this kind of care and how or why things wound up as they have. Speed can be a problem in many ways. Rushing to cure and fix everything, we forget the patient and their wishes; an elderly woman goes to the hospital, gets poked and prodded and optimized, and dies two weeks later anyway. How efficient. ?. She would have stayed home and been cared for at home, if given the choice.

In the talk, as in the essay, Dr Heath makes reference to Christopher Rick's book Beckett's Dying Words. His reflections help illustrate the problem of our strange obsession with fighting death at all costs; later, she drew from Beckett himself:

Samuel Beckett understood more about futile doing than most. He is ­de­scribed by the literary critic Christopher Ricks as:

The great writer of an age which has created new possibilities and impossibilities even in the matter of death. Of an age which has dilated longevity, until it is as much a nightmare as a blessing.

In Malone dies, Beckett writes:

And when they cannot swallow any more someone rams a tube down their gullet, or up their rectum, and fills them full of vitaminized pap, so as not to be accused of murder.

(my German is not great but I believe the essay is a transcript from Dr Heath's keynote lecture at The Art & Science of General Practice and Family Medicine)

The most resonating part of the talk for me was the dissection of our persistent and inflated idea of "need."  Dr Heath found many ways to point out the ridiculousness of our quest to create patients, including thoughts from post-war Polish poet Zbigniew Herbert:

I invented a bed with the measurements of a perfect man
I compared the travellers I caught with this bed
It was hard to avoid - I admit - stretching limbs, cutting legs
The patients died, but the more there were who perished
the more I was certain my research was right.
The goal was noble.  Progress requires victims.

However, it was Iona's casual phrases, like "the contemporary distortion of need" that so eloquently stated the issue. By diagnosing "risk," like an increased risk of heart attack or stroke, we inflate the need. We create a responsibility to do something, do anything to prevent a potential thing that hasn't happened yet and might but probably won't.

We have done this in the case of mammography. For years we promoted self-exams and clinician breast exams. Then we realized those were a waste of people's time and encouraged worry about lumps that were nothings. So we said "just mammogram." In 2009 I read the New Zealand guidelines for screening for breast cancer. At that time, they were encouraging against clinician exams and even suggesting that mammography might not be recommended in future. I failed to convince one of my preceptors, a wise and thoughtful guy. For years he had helped women find "the lump" that ultimately was hacked off, irradiated, and chemo'd, and their lives had been saved. Patients, convinced by their doctors for years and years, still feel something must be done. "Early detection is key."

That anecdotal experience is hard to trump, but as the data floods in, we are discovering that less is more. It is the responsibility of physicians and the medical community to undo the messaging that we touted for years.

It never boils down to just randomized controlled trials, NNTs, or confidence intervals. We are caring for humans, and that cannot be reduced to just fighting disease or perceived risk.

Powerful words bear repeating and so we were given some of Annemarie Mol's to chew on.

The Dutch philosopher Annemarie Mol, in her book The Logic of Care, writes about how ‘the logic of choice’ now undermines ‘the logic of care’.  She says:  ‘Even if good care strives after good results, the quality of care cannot be deduced from its results.  Instead, what characterises good care is a calm, persistent but forgiving effort to improve the situation of a patient, or to keep this from deteriorating’.  The richness of that aspiration compared with getting the numbers to the correct point is incredibly important.

Mol goes on:  ‘You do what you can, you try and try again.  You doctor, but you have no control.  And ultimately the result is not glorious:  stories about life with a disease do not end with everybody ‘living happily ever after’.  They end with death.’ Unless we as both a profession and as a society get over the idea of death as medical failure, we are doomed to torture our patients when we should be leaving them alone. (from The International Futures Forums)

Exploring need, the example of the overdiagnosis of ADHD and the medicalization of our children emerged. Dr Heath explained that labeling and medicating children teaches them 'that they are not normal, not responsible for their behaviour, and that answers come in pills.' It might be that our kids are not disordered - they just learn in a different way. See this snippet of Sir Ken Robinson's great Ted Talk on the subject:

Lastly, there is greed. Diseases are invented in order to market drugs. Normal behaviour is medicalized, or deemed pathological as we sterilize the definition of 'normal.' Those who have get more, and just because we can do something expensive and fancy and new doesn't mean we should. Christopher Ricks:

It is now almost impossible to die with dignity in USA unless one is poverty stricken

If you have the means, you'll be poked and prodded beyond your heart's desire. We've got it wrong most of the time. Doing more is not better, and fails to solve the underlying problem. "We seek technical solutions for existential problems" (Overdiagnosis: when good intentions meet vested interests—an essay by Iona Heath) and that leads us down a very slippery slope. Just one more test. Let's try this treatment.


Let's stop. Let's think of the patient and what matters to them. Or as the patient, what matters to us? Must we keep our numbers within normal parameters? Shouldn't we just strive to feel good, pursuing a "modified hedonism," in order that we might appreciate more the quality than the quantity of life we are given?

General practitioners would do better to encourage people to lead lives of modified hedonism,so that they may enjoy, in the full, the only life they are likely to have.

(J McCormick, Health promotion: the ethical dimension, Lancet, requires subscription)


I admire the strength of Dr Heath. She is an unapologetic champion for the right care, and explained to us with wisdom and compassion the reasons we've wound up in this state of aggressive care, and offered solutions for moving past it. The role of the Family Physician and the connection we share with patients is central to this.



Personal Genomics: Is knowing MORE better than LESS?

Like the total-body MRI, it raises my eyebrows.

Are you SURE you want that?

Are you SURE you want to know every detail of what might be wrong with you, even if it might not actually be something causing trouble, and might be something that you cannot change?

For some people, the answer is YES.

For others, NO.

Personalized medicine, the kind of medicine that suggests individuals undergo genetic and molecular tests in order to develop customized care plans, is in the news. The Vancouver Sun has recently hosted an interesting series of articles all about this technology. It could save lives. And it can be an expensive, slippery slope. (See: Genome sequencing: A costly way to save).

Dr Brad Popovich, chief scientific officer of Genome British Columbia, discussed his own experience with genetic testing with the Vancouver Sun (Personal genomics: The test for everything (with video)). It is illuminating to read about the experience of a researcher and his questions moving forward. However, as staff and CSO of the organization, of course he needs to promote his work and his company's mandate; I'm not sure his experience is reflective of the average person's when thinking about genetic testing.

Another perspective is offered in the Sun by Dr. Matthew Farrer, Professor of Medical Genetics at the University of British Columbia (No quick genetic fixes for complex diseases). As he points out, it's nowhere near as direct a process for diagnosis as some might think:

“The average human genome gives you something like four million variants from one person to another. So, to interpret what those mean and what the consequences are to a person’s future well-being or susceptibility to disease is a pretty grey box.”

This article points out there may be an important role in mental health screening and lending clarity to diagnosis. Other articles document the benefits in assessing familial risk & individualizing cancer treatment or testing prenatally for conditions that are preventable or treatable if caught early on.

Unfortunately, there are lots of issues beyond the fact that this technology is in its infancy. Some things may improve as the body of research grows: the cost, the specificity, the availability. However, significant ethical issues may (rightfully) be a barrier to further advancement and application of this kind of testing.

Most diseases result from a combination of genetic and environmental factors. Some are predictable, and others less so. Some genetic predispositions never manifest as disease, or the disease can be more extreme than the tests ever suggest. That's the beauty of life - we cannot know the future. However, sometimes knowing what to expect can be helpful.

For example, in prenatal testing for Down Syndrome risk, some parents prefer to have an idea if their child might be affected so that they can read and prepare. Some parents may want to know about risk for the disease so they might choose to abort the pregnancy.  Other parents decline the testing because they will love and work with the child however they are born. Unfortunately, the testing is not perfect, there are many false positives and some false negatives.

The vagueness of genetic testing is a real challenge and may be confusing to patients as it may lead to greater uncertainty. As mentioned in the Personal genomics: The test for everything (with video) article, it's not only the patient who has to deal with the results:

Popovich’s decision, made with his wife Nicola, was made simpler because the two have no children. The presence of hereditary disease in his genome has no downstream impact . . .

Having your genome sequenced may have implications for your children, and the potential for this will have to be weighed before embarking on any such testing. Less importantly, there may be implications for life insurance and medical insurance eligibility, where universal healthcare is not available.

At The Conversation, Jacqueline Savard's essay, How genetic testing is swelling the ranks of the ‘worried well’ highlights some of the negative issues, particularly those under the umbrella of Overdiagnosis.

... genetic testing has the potential to give [patients] a label, adding them to a growing class of citizens, the “worried well”. These people are not ill as they don’t display symptoms of the condition with which they are labelled. Nor are they clinically diagnosed as having the disease in question. Instead, they are at a pre-phase, they are now genetically at risk for a particular disease. But the date, time, and list of symptoms they’re likely to experience or the severity of the illness are yet to be determined.

There are societal implications as well.

For society, the implications of genetic testing and over-diagnosis is the opportunity cost of treating the genetically at risk rather than those who are symptomatically ill. In the struggle for scarce resources, decisions are made between who and what diseases will receive the funds they desperately need.

Ms. Savard generously searches for a silver lining, but finds it to be false:

One potential (and limited) benefit of these tests is the creation of new markets for drugs or preventative treatments that can flourish as more people are labelled with conditions. But with health-care costs increasing and the need for rationing a central concern, there’s also the danger that resources will be diverted from public health measures that have a chance of preventing genuine illness in favour of treating “pseudodiseases”.

We don't need more disease. And we certainly don't need more diseases invented specifically in order to sell medications. But we do want people to live as well as is possible, at the same time allow the our natural course of existence to unfold.

As with everything, the answer probably lies somewhere in the balance. Perhaps a few very specific genetic tests will offer a cost-effective way to diagnose and even prevent serious ailments that are otherwise difficult to assess. The murky, inefficient, and ineffective tests can be abandoned so that we might focus on the other things that matter. 

If you don't feel sick and you are living healthfully (on your own terms), is that good enough? Isn't it?

What’s in a name? Why we need to reconsider the word cancer

doctor: "You have cancer."
our brains: Panic! Cancer! What!? Cancer!
our mouths:  "Oh . . . Uh oh! Well, is it a BAD kind of Cancer?"

It's the "Big C." It's a scary word. And fortunately there is a lot we can do with screening, advanced tools for diagnosis, and treatment in many modalities to combat the Big C.

These days, everyone gets cancer, unless something else gets them first. Cancer is part of the natural process of cells replicating, and it's all explained elegantly by the New York Times Sunday Review: Why Everyone Seems to Have Cancer.

But not all cancers will end our lives. There are "pre-cancers" and there are "early cancers," and these are often not going to hurt us, so they are okay to manage with a 'watch and wait' strategy. People can be scared into seeking aggressive treatment because they lump these "pre-cancers" in with the "very much cancers," which usually warrant a good battle in the right patient.

Sometimes it does matter if we can catch things early and treat them before they progress, like when we see changes of the cervix during pap screening; these can be treated to prevent progression to invasive cancers. That's really important, and it saves lives, but it only applies to a certain population, under certain circumstances.

Sometimes it doesn't much matter when we catch things, as we are learning with most cases of prostate cancer, for example. In recent years we have started to discover that the treatment might be worse than the disease, and maybe we shouldn't be screening for prostate cancer the way we have been.

Perhaps we should be more careful then with what we label "CANCER." That word tends to lead to a cascade of troubled thoughts, and we wind up worried about writing a will, rather than understanding what those dastardly – but potentially harmless – cells in our body are doing.

Alexandra Barratt, Professor in the Department of Public Health at University of Sydney writes:

Early cancers and pre-cancers (abnormal cells that could turn cancerous) found by screening tests, such as mammograms and PSA tests, should be renamed without (scary) words such as carcinoma or neoplasia in their title. They suggested they could be renamed IDLEs – indolent lesions of epithelial origin.

She goes on to explain the risks of overdiagnosis, the fallacy of lead-time-bias, and what we can do about all this as patients and providers, with the guidance of the National Cancer Institute.

Read more on The Conversation.


The Naturalness of Dying

I love my job most days. Whether solving a diagnostic puzzle, hearing a patient's story about an important moment, seeing an elderly person regain their pre-hospital fitness, or sending someone home in good health, the joy is unstoppable. Even in sharing bad news, there is sometimes laughter and relief, a thankfulness for a connection between doctor and patient.

There are very bad moments too. You'd think the worst would be losing a patient I am close with, telling a young person that their life will be radically different (or shorter) than expected, or finding out I've made an error despite my best efforts. These are low moments – but they are not the worst.

The most unsatisfying, defeating moments are those I face when struggling with the ethics around the time of death and dying. Specifically, these usually occur when I think I'm dealing with "end of life care" and a patient, or more often their family, is having a hard time acknowledging that's where we are. 'Denial' or 'refusal to accept' are terms that we throw around, and while part of the natural grieving stages, these states are not meant to be permanent. There also may be an element of impatience on the part of we, the healthcare providers, who have a hard time letting grief happen in its own time.

I have a very hard time being asked to artificially prolong the life of a patient who is suffering. Take for example a 93 year old woman with end-stage dementia who is no longer interested in food, too weak to swallow without choking, and cannot communicate much - especially not her wishes or preferences. When we reach this point, most doctors believe that it is time to let nature take its course, and to treat any symptoms that make the woman uncomfortable. This woman's current state is the unfortunate natural progression of her disease. However, the family does not want to let her go; they promised they would never "give up" and so they tell me they would like their grandmother to have a feeding tube. I emphasize that focusing on her comfort is not "giving up." They hear me, but they don't really hear what I say. Try though I might, I am unable to persuade them that this artificial means of prolonging her life is uncomfortable, inappropriate, and may be inconsistent with her wishes.

It is against my morals and against my oath to force nutrition into a patient who has a non-reversible condition which prevents her from eating. Yet legally, and some would argue ethically, the family has recourse to request what they feel is in the best interests of their loved one. Often, the physician is obliged to do it, no matter how wrong or like torture it seems. Just because something is medically possible doesn't mean it is right.

Obviously there is a minefield of tension around end of life care issues. There are grey areas, physicians are not always right, and everything has to be discussed with the best interests of the patient in mind.

It's our job as physicians to work together with our patients and the healthcare team, to shepherd people to good health. I also believe it is our duty to care for the sick and dying, to be by their side and help them achieve the best quality of life for their remaining time with us. When patients and families do not acknowledge that death is a part of life, that is when I struggle most.

There is a lot of fear around death, including fear of facing the unknown. They say it takes a very strong person to know when to leave the fight. I think it takes a stronger one still to know that even in accepting death and dying, the fight is not over. One should still fight: for happiness, comfort, spiritual peace, closure with family. But fighting death, when it is clear that doing so will only harm you, is hard for me to understand.

There comes a time when "good" health may not be possible. This is not unnatural, this is not a great injustice - this is dying. And this is a normal part of life.

Back in 1995, Dr. Jack McCue wrote a paper for JAMA that is timeless. He considers our attitudes around death and dying, suggesting that the medicalization of the process diminishes the autonomy of dying patients and results in bad and wasteful care.

When I'm having a bad day, struggling with this issue, I commiserate and quip to my colleagues: "Death is not the enemy." I have a feeling Dr McCue might agree.

Read The Naturalness of Dying, from JAMA.