To what end?

Those three words have become something of a mantra, a mission, a philosophy of care.

• To what end do I prescribe a medication?
• To what end do I make a diagnosis?
• To what end do I order tests?
• To what end am I documenting?
• To what end is there a patient record?
• To what end do I send a person to a specialist?
• To what end do patients need to come to see me in the office?
• To what end do my patients have me as a doctor?

This is tied closely to the thought process of a recent post in which I discussed the flawed idea of a problem-oriented health care system.   When we treat a problem as the center of a person’s care, we look at one end in their care: fixing the problem (usually using a procedure).

You may rightly ask the question: “To what end are you writing this blog post?”  Touché.  To answer that smart aleck question, let me go through several of the above bullet points for doctors who work for the usual employers: insurance companies and government bureaucracies, and those who work for my current employer: the patients.

1.  To what end do I prescribe a medication?

This is actually what started me thinking on this line, as I found myself spending much more time talking people out of medications than I had ever done in my past life.  The goal of medication in the old system is often muddled and confused:

• Q:  Why give an antibiotic for an upper respiratory infection, or for “bronchitis” lasting less than a week?
• A:   Because it takes too long to convince people they are not useful, and they have already waited for 2 hours to be seen and will be angry to leave the office without an antibiotic.

I find I am far more likely to hold off on medications now, as people know I won’t force them to come back in if they get worse.  The cure for antibiotic overuse is “watchful waiting,” in which the body generally gets better without medication assistance.  The problem with the old system is that pesky first word, “watchful,” which implies paying attention (aka communication).

• Q:  Why use a cholesterol medication?
• A:  To lower cholesterol.  (Duh)

This, of course, is not the right answer for the sake of the patient.  Cholesterol is a risk factor, not a disease, and there are plenty of circumstances where the risk heart disease or stroke is low enough that the of the medication becomes significant.   I am now turning my patients’ attention away from cholesterol or other risk factors, and toward their risk.  My goal is not to fix a little problem, but avoid a big one.

2. To what end do I make a diagnosis?

It’s a really complicated topic, actually, made much worse by the need for diagnosis (ICD code) to get paid in many/most cases.

• Q: Is obesity a disease?
• A: It is as long as the insurers pay for the diagnosis.  They never have in the past, but recently have been doing so more and more.

I don’t actually care that much if it’s called a disease or not.  To me, it’s something that increases the risk of certain things and reduces the quality of people’s lives.  I don’t think it’s something that people should be ostracized for, but I also know that my weight struggles have been directly related to bad choices I make.  The only reason to call something a disease is if doing so opens opportunities to reduce people’s risk or improve their quality of life.

• Q: What about symptoms like fatigue, irritability, or foggy-headedness?
• A:  They go on the problem list as long as there’s an ICD code for it.

Symptoms are not diseases, nor are they risk factors.  They are a report of the person’s experience in life.  Symptoms are significant only in context of a person’s risk.  If a person has chest pressure and shortness of breath, is that significant?  The answer to that question is different for a 20 year old female and a 56 year old male diabetic smoker.  In the former, even classic angina is not likely to be heart related, while in the latter even atypical pain is taken as possible equivalents to angina.  Again, it all comes back to risk.

3.  To what end do I order tests?

There is only one reason to order tests: to gather more information to make a decision.  The end goal (as stated in #2) is not to make a diagnosis, but to decide what the path of action should be.  I recently had a patient come to me (as is often the case) asking for me to order an MRI scan.  In this case it was because of knee pain.  When I suggested that perhaps a visit to orthopedics the patient resisted, feeling that I was somehow offering a lesser alternative.  I countered with the following options:

1.  We get the MRI and it is normal – in which case you still have pain in the knee that has not been addressed.   I would probably refer them to surgery as I had suggested.

2.  We get the MRI is abnormal – This would cause me to refer as well.

3.  We don’t get an MRI and the orthopedic doctor decides to order one.  This won’t cost you any more

4.  We don’t get one and the orthopedic doctor decides to go straight to arthroscopy.  This would save you hundreds, perhaps more.

For lab tests, it’s really important to understand the situation you are in, and there is no better statistical tool than the pretest probability (chance it is present) of the condition you are trying to diagnose or rule-out.  I usually think of pretest probability in three main categories:

• The problem is unlikely to be present
• The problem is likely to be present
• I am not sure.

It helps me to ask two questions:

• Would I believe a positive or negative result?
• What would I do with a positive or negative result?

So if I swab someone’s throat for strep and they have a high pretest probability (red throat, fever of 102, no runny nose or cough, exposure to strep) a positive test would simply confirm what I know, while I’d wonder about the accuracy of a negative test .  On the other hand, if the pretest probability was low (the person felt normal), I’d mistrust the positive result.  It’s the exact same test, but the result is interpreted entirely on the pretest probability.

One more scenario in which a test is run is in the case of something that is low risk, but needs to be ruled out.  This is generally done when the problem is too dangerous to miss (cancer in an adult, meningitis in an infant).  It’s extremely important to have a very sensitive test in this case, that has a very low rate of false negatives.

4.  To what end do people come in to see me as a doctor?

Back in the day when I was paid by insurance companies, the answer often was, unfortunately: “because it’s the only way I get paid.”  There are many cases where the care could have been given over the phone, but that would shoot my business in the foot.  Unfortunately, you can’t take care of people if you can’t pay your bills.  Even though the business was not my primary goal in terms of priority, it was the highest priority in terms of order.  I first had to have enough money to pay my staff, our office lease, my salary, and all the other expenses, then I am free to give care.

In my new world I am paid just the same if the office is empty (something that would have worried my in my previous life) than if it’s full.  So why come in to be seen?  There are really two reasons:

1.  The patient wants to.  Some people just would rather come in and talk to me.  I am not sure if this is due to my sweet personality or some chemical imbalance on their part (just kidding, of course).   That’s perfectly fine with me.

2.  I can give better care by having them here.  The physical exam is sometimes important.  Sometimes I need to be able to look them in the eye when I talk to them.

5.  To what end am I people’s doctor?

In my old practice the answer was often: to help them in case they are sick.  The patient would probably say the same, with the addition of: so I can get medications and tests when I need them.

My new practice has two main goals:

1. To get my patients as old as possible.
2. To keep them as healthy and happy as possible while they do the old-growing.

The goal is no longer medical intervention, it is avoidance of care.  The goal is not to treat problems, but to avoid them.  The goal is not to order tests, but to not need them in the first place.  The patient is my employer, and so my I do what keeps them happiest with my care.

And that’s a nice world to live in.

Rob Lamberts is an internal medicine-pediatrics physician who blogs at Musings of a Distractible Mind.