Wednesday, December 26, 2012

Geriatricians are the GPS of the health care system

It is sometimes difficult to explain how geriatricians are different than other physicians.  Yes we deal with diseases and syndromes like Dementia, weight loss, transitions of care that other physicians don't like to manage.  But there is something more and I always had a hard time succinctly explaining it.  I decided to use the idea of a GPS to explain how qualitatively Geriatricians are different.

First, most physicians have very disease/organ focused view of health.  The right way to treat a heart that is failing is determined by studies and guidelines.  But for geriatricians there are a number of things we consider in addition that are very patient oriented.  On the continuum of therapeutic and palliative goals, what does the patient want us to accomplish?  How is health, quality of life, suffering and risk defined?  These are questions that patients can only answer for themselves.  Geriatricians call this goals of care.  In addition to efficacy questions that most physicians ask, geriatricians take into account life expectancy (will patients live long enough to see a benefit), what is the potential harm (risk) and is that tolerable to a patient (pain tolerance), what is the real likelihood of a treatment creating benefit (is it worth it) and is a treatment worth the pain (combining risk and benefit).  By thinking of goals first and using them as an anchor (or destination in my GPS analogy), fitting evidence to patient oriented goals and limits, interventions become more rational.

Here are some big picture stats:
1/3 of severely ill Medicare patients with palliative goals feel that their care is too aggressive.
13% of people with severe dementia with a feeding tube got a feeding tube with no discussion between the doctor and family about whether it was even wanted.
Less than 50% of severely ill patients have an advance directive in their medical record, only 1 in 8 receive input from their physician and the vast majority of physicians are unaware of the patient's advance directive.
88% of dialysis patients never knew that home dialysis is an option
Most cardiologists that recommend ICD's never discuss the risks with the patient but feel adhering to guidelines matters more.
Not only do patients get more aggressive care than they want, sometimes patients request more aggressive care than doctors feel is in the patient's best interest.  And the opposite happens as well-older patients are sometimes treated as if they are already dying and desired, effective, tolerable care gets denied due to age alone.
These stats could go on and on.

I will add four examples from my own patients.  After doing what I do, I kinda assume that everyone thinks like me (why wouldn't they?!).
The first patient was a 90+ year old patient with severe dementia, bed bound, contracted, oriented to self and family but not place or time.  She was at home surrounded by family.  She was able to enjoy her family, eat and was pain free.  She developed a sacral wound that then walled off leaving a cavity underneath healed skin.  She had no fever or signs of systemic illness (normal white count).  She was admitted to my inpatient unit.  We consulted surgery who recommended debridement under general anesthesia as this likely went to bone.  We met with the family, discussed goals, risks and time to benefit.  We discussed her current quality of life, her goal of comfort and tolerance of the risk.  After getting to know the family we decided to leave the wound alone unless she became systemically ill and even then would still probably leave it alone.  We discharged her home with just local wound care.  I received a phone call from her quite irate family doctor who after about a half hour of discussion basically said I was not even trying to help the patient, committing her to death.  When I informed him of the extensive family meetings and content, he said families don't know what is right and that physicians need to tell the families what to do.  It dawned on me that this physician would rather have his patient die in the operating room or after after a painful post op wound care course that would involve institutionalization in a nursing home to get more daily advance wound care treatment than to simply focus on comfort, quality of life and calling it a day.  In his mind, real physicians try to help their patients live as long as possible even if it results in catastrophic failure and fails to meet any goals.

The second patient was very similar.  90+ years old with severe dementia in a nursing home.  Pt had progressive kidney disease to the point where dialysis was the "right" treatment for her kidney function.  However she was asymptomatic, could even stand up and dance during music recreation time and had her friends she ate with 3 times a day.  The renal doctor decided to call the son to initiate dialysis over my objections saying that without dialysis she would die.  Despite my objections that dialysis could kill her sooner, make her feel worse, make her spend less time enjoying her life even if she lived longer, the nephrologist convinced the son to put in a dialysis catheter.  One week later, the patient got septic and died from a catheter infection.  Again, "at least she tried."

The third patient was an office patient, also in his 90's.  He was diagnosed with severe heart failure and was hospitalized 3-4 times in 6 months.  He was also diagnosed with metastatic melanoma.  He had a fall and minor hip fracture not requiring surgery.  During one of the hospitalizations, his cardiologist walked into his room and recommended an implantable defibrillator saying that "it would help" him.  No further discussion.  So the next day, he had a $50,000, 10 year device placed.  When I saw him in the office I asked him what he hoped it would accomplish.  He had no idea.  I had a very frank discussion that he was going to die in the next year either from congestion in his lungs, metastatic cancer, continued frailty/immobility or an arrhythmia.  The purpose of the ICD would be to not allow him to die of the most peaceful way (the arrhythmia).  He said that he didn't want it, I emailed the cardiologist who agreed to turn it off and it became a $50,000 oops.  (I got paid $100 tops for the discussion).  The unfortunate thing was although he signed on to hospice, the cardiologist stopped seeing him and didn't turn off the defibrillator.  Fortunately, 6 months later he did die without the device ever firing.  But at least the cardiologist "tried."

The last patient was a 90+ year old bed bound patient with severe dementia at my nursing home.  She was unable to talk, did not recognize herself.  She had end stage renal disease.  She had severe daily arthritis pain and sedation from pain medications.  She had a heel wound as well.  She went to the hospital for fever or shortness of breath (I don't remember), coded (vfib arrest), was shocked back into sinus rhythm but had to be intubated and was sent to the ICU.  So now my 90+ year old patient has end stage dementia, end stage renal disease, severe daily arthritis, respiratory failure on a vent, s/p cardiac arrest and has a heel wound.  A surgeon came over and recommended a below the knee amputation under general anesthesia.  Infectious disease, vascular, the critical care doctor and the family agreed.  Despite my phone calls over 2 days, the surgery went on as scheduled.  The next day after the amputation she died on a ventilator in the ICU.   At no point did anyone else consider hitting the breaks.  At least the vascular surgeon "tried."

So what's my point?  Without an anchoring in goals, the amazing things we can do with medicine is like a fast exotic car with no brake pedal, steering wheel or destination in mind.  We do a lot, sometimes it is too much, too little and often ineffective for what matters to patients.

I leave you with my attempt at a cartoon.
Health care with geriatricians...

Health care without geriatricians...

Geriatrics isn't dead

As a relatively new geriatrician (finished my fellowship in 2009), I often hear about how geriatrics is dying as a field from those within academia and who have a larger view of geriatrics as a field (see this link for an example).  While geriatrics (100 US med grads, 275 total per year) may never be in the same ballpark as Internal Medicine (8800 new grads a year), Family Medicine (3300), Pediatrics (2900), Gen Surgery (2300), ER (1700) or Obstetrics (1300) in terms of volume, the field may persist quite healthily as a smaller specialty such as Plastic Surgery (116), Vascular Surgery (123), Thoracic Surgery (90), Allergy Immunology (131) or most pediatric subspecialties (ER, GI, Endo, Cardiology, Rheumatology, Nephrology etc).  After seeing numbers that JAMA publishes yearly I am more optimistic that geriatrics will at least survive as a specialty.  Ideally we would be the other end equivalent of pediatrics but that's not going to happen.

Wednesday, June 20, 2012

Bring up hospice is like proposing to one's girlfriend

1.  It shouldn't be a surprise
2.  You should have some idea of the answer before you ask the questions
3.  It is not how you start the conversation.
4.  Don't consult social work, ask yourself!

It's interesting to me how intimidating it is for doctors/residents/med students to bring up hospice (for a variety of reasons) and for patients/families to have it brought up or to bring it up.  So to make thing simpler, I've come up with this analogy that I think applies for at least these 4 reasons.

Tuesday, June 5, 2012

Why it is difficult to collaborate with your physician

Recently I read two articles in the NYTimes (1, 2), stating how doctor's and patients fail to communicate or collaborate meaningfully and why that is necessary.  I recently wrote about what questions I would ask a doctor when a doctor proposes a medical intervention whether that is a surgery, a medication or a test.
Here is the problem:  doctor's don't have the answers to these questions readily available and even if they did, they don't have the time to explain the answers to those questions.

1.  Guidelines and review articles often fail to present data in a clinically relevant way.  It is up to the individual physician to go through the studies in detail to translate the info into a clinically meaningful way.  Most docs don't have access to articles or time to do things like this.

2.  It takes me a good half hour to explain the proper use of Aricept for example or the role of cholesterol medications or when to do prostate cancer screening.  It takes time.  I don't have that time.

Good intentions are nice but the data as presented currently and the structure of the office visit as constrained by financial pressures makes collaborating more of a pipe dream.

Wednesday, May 23, 2012

Trusting your doctor

One of the main things I do is to work with patients and families in education and making decisions.  It can be the most rewarding thing I do and the most frustrating.

I've learned something about the frustrating times-at some point, people have to trust other people.  Life works better that way.  Many times, patients and families don't trust their health care team (which is understandable-really).  But the way they compensate for that is by thinking they can learn the expertise to oversee the health care team and direct the health care team.  I get that as a sentiment.  It is intuitive, but impossible to carry out.

I'll give you a simple example, whether or not to take aspirin to prevent a first heart attack or stroke.  There are a handful of studies over the last 20 years that have gone into making the USPSTF guidelines.  There are newer studies coming out (1, 2, 3).  All have methodological/epidemiological issues (internal validity-is the study well done).  They all have external validity issues (application to a particular patient/population in the real world).  They attempt to communicate risk and benefit by using things like relative risk, Number Needed to Treat, absolute risk reduction, 10 year risk, risk calculators etc.

It takes a lot of thought to digest this info and then it has to be balanced against the evidence for blood pressure reduction, cholesterol reduction, healthy lifestyle, diet, smoking, other blood thinners like plavix etc.  To fully appreciate whether taking aspirin is "worth it" or not, a person would have to understand study design, statistical analysis and have a pretty comprehensive view of the literature and then be able to make rational sense of outcomes.  This is not easy to do and in fact I would argue for a lay person (or even a physician of the wrong specialty), this would be impossible to do for the simple question of whether someone should take aspirin.

Other questions that come up: should I start a treatment, should I get a diagnostic test, how do I navigate the health system etc are all very difficult questions with very nuanced, subtle, yet vital issues involved.

At some point, patients/families have to trust that the recommendation they are getting from their physician is a good one.  If they don't feel that it is, then they need to switch physicians because the solution of trying to know as much as the physician isn't realistic (or any more realistic than me understanding fully a legal issue or real estate issue or a car mechanic issue).

Too many times I've seen people make decisions badly when they are trying to be their own expert.  I suppose this is in response to many patients experience that the advice they receive from physicians can be worthless.  I get that.  It is hard to find a good physician.  But the solution to a lack of trust is to find someone to trust or to rebuild trust, not to create the facade of being one's own expert.

Tuesday, May 1, 2012

What Questions Should I Ask My Doctor?

One thing I do on a regular basis is help patients and caregivers make medical decisions. I watch my colleagues in other specialties give information and guide patients in decision making and I am often left with a fairly dissatisfied feeling that the conversation didn't quite hit the spot when I see the decisions are made and the regret that many families have afterwards. Decisions to start dialysis, transition to hospice, place someone in a nursing home, proceed with an amputation, put in a feeding tube or a defibrillator, start chemo, send someone to the emergency room, put someone on a ventilator are all decisions that are very complicated. Even starting or stopping a statin for cholesterol is not an easy decision. When discussing code status (DNR/DNI), all simplicity goes out the window.  The way the medical community is supposed to help families is through a process called informed consent. This has been determined by the court system and ethicists as the minimal amount of information a doctor is obligated to give a patient and family before asking for a decision.

From the AMA's website on informed consent
In the communications process, you, as the physician providing or performing the treatment and/or procedure (not a delegated representative), should disclose and discuss with your patient:
1. The patient's diagnosis, if known;
2. The nature and purpose of a proposed treatment or procedure;
3. The risks and benefits of a proposed treatment or procedure;
4. Alternatives (regardless of their cost or the extent to which the treatment options are covered by health insurance);
5. The risks and benefits of the alternative treatment or procedure; and
6. The risks and benefits of not receiving or undergoing a treatment or procedure.
 In turn, your patient should have an opportunity to ask questions to elicit a better understanding of the treatment or procedure, so that he or she can make an informed decision to proceed or to refuse a particular course of medical intervention.

 It has taken me a while to figure out what I don't like about the informed consent process. Informed consent has it all wrong!  The problem with this model is that it spends all the time talking about the procedure/alternatives and not enough time emphasizing how it is contextualized to the patient.  I suppose under risks and benefits it could be assumed that a doctor would let a patient know how a given intervention will impact their life, but the way informed consent is done, it often stops with the immediate physiological benefit (you will not have a blocked bowel, your heart will restart, it will fight the cancer, your blood pressure will be lower).

There are two problems
1.  No matter how detailed the information, there is simply no way for a patient and family to truly understand the above 6 points without understanding the full medical literature combined with personal clinical experience.  It is one thing to say a patient may suffer or that a procedure may succeed, but what does that look like exactly?  Not being an oncologist, no matter how much education I get on chemotherapy, I will never really understand how one regiment is better than another without looking at the literature and seeing how patients are affected in real life.  Informed consent is a facade that makes doctors, patients and families think that the patient/family is making the decision autonomously where in reality it is like picking a dish at a new restaurant: you can read the description but you don't really know how it will turn out.  To put it another way, I have a very low illiteracy when it comes to cars.  No matter how many times someone explains to me different types of catalytic converters, a-I don't really care and b-I don't get it, just fix it or put a new one in and make my car go...which brings me to my second point

2.  The most important part of the decision making is figuring out how the options fit in with a patient's goals of care, not the ins and outs of the medical procedure, the literature and the clinical experience and the science.  For example, patients need to have some sense of whether a procedure will help them live longer, cause pain, be more independent, enjoy life, suffer less.  I think giving patients/families the best information we can about goal directed outcomes will allow them to make much more relevant choices.

Here is my schematic
These are the four areas I think about when helping a family make decisions.  Too many times the conversation takes place completely in the top right circle of the medical science and technicalities of the intervention itself.  Too often how it would impact a patient's goals gets ignored.  I would argue that if a conversation regarding a medical decision took place completely in the top left circle, bottom left circle, it would be a much more complete and accurate way to make a decision that will reflect the patient's best interest and will produce the least regret.

One thing I worry about is when doctor's tell patients what their goal should be and patients tell doctors what procedure/treatments they want.  That gets things backwards.  Ideally a patient would state a goal and the doctor would tell them the best way to achieve that goal.  So for example, a physician may say, I know your goal is to live as long as possible with your wife at home, let's do hormonal therapy for prostate cancer because it will maximize your chances of surviving while minimizing the toxicity of treatment.  Or he may say, I know your goal is to live as long as possible and you are okay with taking risks, let's do surgery and chemotherapy as your best chance for living as long as possible even though it is riskier.
When it comes to informed consent, I usually don't like how it is done anyway even putting the goal issue aside.  Here is what I would want discussed:

1. Is it even effective in an older adult? 
2. Is it effective for a clinical endpoint that the patient cares about?
3. What is the time frame for efficacy (over long term/short term)
A.  Will the patient outlive the usefulness of the intervention?
B.  Will they achieve it within their life expectancy?
4. What is the likelihood of achieving benefit (i.e. NNT)?  
Is that worth it to the person?
5. Is it worth the risk of side effects?
Is it worth the risk of failure
6.  Will it achieve their QOL goal?  
7.  Is it a priority among all the person’s medical problems?  

At the end of the day, contextualizing a decision in a patient's goal is far more important to me than discussing the ins and outs of the science of a medical intervention.  Unless it is translated to a goal, the information is useless.  And I would argue that a translated decision into a goal without the discussion of the ins and outs of the intervention would still lead to a pretty accurate choice that reflects the patient and minimizes the chances of making a wrong decision or guilt.