Nurse Practitioners and the “Art of Medicine”

May 11, 2010 in Health Policy, Medicine by RangelMD

By the “art” of medicine I mean the ability to reason beyond rote memorization, strict standards of care, and “cook book” style medical practice. By “art” I do not mean the tendency to apply diagnostic and therapeutic medical practice in a freewheeling way that is not based on reasonable evidence (this is quackery). The art of medicine is something that is very difficult to teach in a didactic course. It’s something that is often learned in a practical setting from a mentor who pours out their years of experience onto the pupil. This goes on for 3 or more years and hundreds of 36-48 hour shifts during post-doctoral training for physicians. Yet this type of practical training is usually squeezed into a few months for nurse practitioners. One hopes that this does not make a difference in patient care. Experience tells me otherwise.

Recently I saw a patient who had been seen by a nurse practitioner (NP) a few months prior. The patient was an advanced octogenarian who had not been feeling well with recent loose stools, nausea, and weakness. I reviewed his medication list and found that the NP had prescribed metformin for this patient’s diabetes. Metformin is an excellent medication for diabetics since it is cost effective and does not have as much risk of weight gain and hypoglycemia (low blood sugar) as other medications. But it does have side effects, the most common being nausea, vomiting, and diarrhea, and a rare risk of lactic acidosis which has a very high fatality rate.

Although not an absolute contraindication, advanced age is associated with increased risk of metformin associated lactic acidosis because of age and diabetes related decreases in kidney function and the increased incidence of other co-morbidities including heart failure. Sudden illnesses in the elderly such as infection can quickly lead to dehydration which can lead in turn to kidney failure and induce metformin associated lactic acidosis.

I asked the NP about this issue and the apparent solution was to be argumentative. First the NP suggested that we periodically check a lactic acid level. Not only is this of no practical value since lactic acidosis can set in and lead to death in a period of hours but there is no evidence that periodic  monitoring of renal function and lactate levels reduces this risk. Then the NP’s plan B was to claim that metformin is safer in the elderly than other drugs called sulfonylureas because it won’t cause dangerously low blood sugar. But studies (Arch Intern Med 1997 Aug 11-25 and JAMA 1998 Jan 14)  have found the risk of hypoglycemia from these medications in the elderly to be very low (16.6 episodes/1000 person-years) and have concluded that elderly patients can be safely treated with these medications.

Then there are the practical aspects of avoiding harmful effects. Though the incidence and hence risk of hypoglycemia with sulfonylureas is higher than the risk of lactic acidosis with metformin, the symptoms of low blood sugar are much more apparent and can result in quick and effective treatment while lactic acidosis often presents as vague and non-specific symptoms leading to treatment delay. And there is no evidence that early intervention in metformin associated lactic acidosis can reduce the risk of death. And while lactic acidosis is mostly an idiosyncratic and difficult to predict condition, hypoglycemic risk can be effectively ameliorated by starting treatment with low doses and increasing the dose slowly while educating the patient and the caregivers about the signs and symptoms of hypoglycemia and ways to avoid it and treat it.

But the NP didn’t understand the biggest issue of all. Does the patient’s diabetes even need to be treated?

As it turns out, the patient’s measure of diabetic severity, the Hba1c test, was very low at 6.3% (normal being less than 6%) and was checked several months prior when the patient was not taking any diabetic medication. So the question remains, why should we treat this patient with medication?

There are two reasons why we treat diabetes. The first is to reduce the near term effects of hyperglycemia (high blood glucose) that can lead to increased urination, increased thirst, dehydration, weakness, blurred vision, ketoacidosis, etc.  The second reason is to reduce the long term risks from untreated diabetes such as heart disease, stroke, blindness, peripheral vascular disease, and kidney damage. Not only did this patient NOT have any symptoms of diabetes in his untreated state but given his advanced age and other co-morbid conditions, his chances of developing a complication or symptom of diabetes in its untreated state are minuscule compared to his risk of developing complications from a side effect of diabetic medications!

It is this subtle point that is the “art” of medical practice. It’s the ability to compile all the data and treatment standards and throw in rationality in understanding risk and benefit to determine the importance of knowing when NOT to treat the patient as much as knowing when and how to treat.

Had this issue been about a rare side effect of a very uncommonly used medication, I would have understood but this issue involved one of the MOST commonly prescribed medications with a very well known risk for an extremely common medical condition.

In the very least, this NP is a malpractice lawyer’s dream. If you are going to treat a patient with a potentially deadly medication when there are safer alternatives around then prudence would dictate that you do so for a good reason and document this reason as well as the fact that you informed the patient of the risks and what to look out for and how to avoid the risks (if possible). This NP didn’t document any of this and lawyers like nothing more than an uninformed patient.

So far, studies have not shown any difference in the quality of care or outcomes of care between NPs and MDs/DOs. But many of these studies were done at academic centers where the training and preparation of the NPs and physicians is generally excellent rather then in the usual outpatient setting where the quality of NP training and delivery of care can be much more variable and less controlled. And though the odds of a serious reaction in this patient are low, multiply this by hundreds of patients and the odds that there will be a serious medication reaction increase dramatically.

I don’t know how to teach the “art” of medicine but it seems reasonable to assume that this “art” will be significantly lacking for health care providers who are either inadequately trained and experienced and/or don’t have the time to apply anything more than a “cookie cutter approach” to medical practice.  As primary care dies a slow death in this country, both of these factors are going to become more common. NPs will have an ever increasing role in primary care since very few physicians are entering this field after training and most NPs will be employed by corporations and private clinics and under ever increasing pressure to see as many patients as possible.

Please share.