the risk of decision making

Monday, July 13, 2015

There is a joy to be found in numbers and many sports fans have translated their appreciation of statistics to the level of passion because of fantasy sport leagues. The concept is simple. Managers and coaches devise on-field strategy based on their expectation of how the past might predict the future, and bettors rely on the same thought process. As it turns out, medicine works the same way, but the stakes may be a little higher.

A good example might be the approach to the patient with chest pain. While appreciating that there are many serious diseases and illnesses that can cause chest pain, pressure or tightness, most people and doctors worry about heart disease. The heart is a muscle and like any other muscle in the body, requires arteries to deliver oxygen and nutrients. If plaque and calcium narrow those arteries, a patient might experience symptoms during exercise or work or even walking upstairs. This is called angina and it may not be pain, but could be pressure or tightness of indigestion or perhaps nothing more than n indigestion or mild shortness of breath. If a plaque ruptures and a blood clot forms to completely block the artery, the section of heart muscle that it supplies will die if the blood flow is not re-established quickly. This is called a myocardial infarction or heart attack.

When a patient presents with chest pain to the ER, as protocol, the nursing staff may do an EKG, even before a doctor sees the patient. That EKG may diagnose the acute heart attack and the patient will head off to the cath lab to have a cardiologist open the blocked artery. But if the EKG isn’t exciting, it’s up to the doctor to decide whether to proceed down the “is it the heart?” pathway. Sometimes, clinical judgment says that the pain is coming from the esophagus, or lung or chest wall and no further heart testing is needed; but if the symptoms are suspicious and risk factors are present (smoking, high blood pressure, high cholesterol, diabetes and family history), more needs to be done.

Heart muscle that is irritated can leak a chemical, called troponin that can be measured in the blood. It takes a couple of hours or more for that test to turn positive, so an initial normal troponin is heartening but not necessarily conclusive. It means that no heart damage has yet been uncovered but a repeat test may be worthwhile.

A normal EKG and normal blood test means looking back in time that no heart damage has occurred, but does that mean that the heart arteries aren’t partially blocked? Is there a possibility of a heart attack in the near future? What do the statistics say about predicting the future and how low does the risk have to be before the patient and the doctor are happy?

The American Heart Association recommends that patients with chest pain, who are found not to have heart damage, should have their heart imaged in some way, within 72 hours of their visit. That imaging could be a stress test while walking on a treadmill, a radioactive dye injected to show blood flow to the heart, an echocardiogram (ultrasound of the heart), a cardiac CT or the gold standard and most invasive test, a cardiac catheterization. Each has its indications and place in the risk stratification of the patient, but the decision as to which test to order has some art mixed in with the science. All these tests are not cheap and requires a fair amount of technology, and may not readily be available.

Or the patient can assume some risk. There have been research and studies that have tried to put a number to that risk. In one group  of patients who wee followed whose chest pain story was slightly or not suspicious for heart disease, who were younger and also had normal EKG and blood tests, the chance of having a major adverse heart issue within six weeks, was 1.7%. This number could drop to below 1%, if the patient were watched for a longer time frame and more blood tests done, but theirs never became zero.

Many patients and their families would accept a risk of less than 2%, meaning that more than 98 times out of 100, their heart was healthy. But the numbers also need to be looked at from the doctor and hospital perspective. A busy hospital ER might see a hundred or more chest pain patients each week. That less than 2% risk could add up to hundreds of missed heart attacks a year… not very acceptable numbers. While the Heart Association guidelines may be too strict, others may be too lenient and that is where clinical judgment, practical experience and common sense come into play.

Statistics may help drive strategy, but a manager’s gut feeling or game savvy may suggest a road less travelled when shifting fielder or changing pitchers. The art of medicine uses the same intuition to help guide the doctor and the patient to decide which plan of action is appropriate for one particular situation at one particular time. In medicine, statistics are fine for the general population, but when a complication happens, it happens 100% of the time to that patient. The goal of medicine is to maximize care, minimize risk and use resources wisely… and for that reason, doctors could learn a thing or two from managers, coaches and fantasy draft junkies.

This entry was tagged , , , ,


taking time to care

Monday, April 13, 2015

After the first week of the season, the rules to speed up the game of baseball may be having some effect. Games are taking a little less than 3 hours to play, about 6 minutes shorter than last year’s average. It seems that what was once America’s past time, is suffering because the combination of a shortened attention span and the inability to be patient has caused its popularity to wane. People don’t tolerate waiting very well, even if the reason for that wait is compelling. And that brings us to why people wait in the ER.

If one were to watch Grey’s Anatomy ER or Nurse Jackie, it would be quite reasonable to assume that all medical problems could be solved in 44 minutes; the rest of the hour is filled with commercials and next week’s preview. But technology has yet to catch up to what is portrayed on television and movies. The idea that technology cannot be rushed is anathema, not only for the patient and family who are actively being cared for, but also those who are waiting fro the next ER bed to be open or nurse who can provide can provide care.

Patient flow in the ER is at the whim of available technology and even though the calendar says it’s the 21st century, high tech does not necessarily equal high speed. Consider the child with abdominal pain. Parents often are concerned about appendicitis, an inflammation that if unrecognized, can lead to bowel perforation, significant infection and major complications. Patients often don’t read the textbook and forget to present with the classic symptoms of umbilical pain moving to the right lower quadrant, accompanied by loss of appetite and vomiting. Physical examination can be compelling and an elevated white blood cell count, that often accompanies inflammation, seals the diagnosis and the patient is taken to the OR for an appendectomy. That was the standard of care forever, or until the availability CT scans became the norm, and that standard had surgeons removing normal appendixes 15-25% of the time. It was a reminder that other things could cause right lower quadrant abdominal pain.

The good news about using technology to look into the body is that it can help make the diagnosis. The bad news is that it takes time and radiation. And here is where technology can’t be rushed. In many hospitals, if appendicitis is a consideration, blood tests might be ordered to help give direction. Depending upon the tests ordered, it might take 30-60 minutes having the blood drawn, delivering the tubes to the lab and getting back the results. Instead of rushing to CT, an ultrasound might help make the diagnosis, but that is labor intensive and is helpful only if the appendix can be identified. If not, a CT scan might be required. Often the images can be taken with just intravenous contrast injected through a vein, but some radiologists prefer having the patient drink oral contrast to help outline the intestine. If that’s the case, the drinking time takes an hour, the scan itself 10-15 minutes and then another 30 minutes or more for the images to be reconstructed and interpreted by the radiologist. Even new math can’t make the answer add up to 44 minutes.

In chest pain patients, blood tests can help decide whether a heart attack has occurred. Troponin, a chemical contained within heart muscle cells, can leak out and be measured by a blood test. An elevated blood test equals a heart attack, but it may take 6-8 hours for the test to turn positive. That means a patient who presents within a few minutes after experiencing chest pain and has a normal EKG (electrocardiogram) might have to wait many hours to find out if his heart was damaged.

The worry for women who have vaginal bleeding in very early pregnancy is not only that they may be miscarrying, but instead might be experiencing a tubal or ectopic pregnancy. One test that might help guide diagnosis would be a quantitative beta HCG, a chemical that is produced by the placenta. Its value should double every couple of days in early pregnancy and in stable patients, they may need to return in two days to help sort out the diagnosis. Meanwhile, the evaluation and care of the patient involves significant amount counseling and that take time.

Medicine and baseball share the concept that the game is not a prisoner of the clock. It’s done when it’s done, when the final out is made and when the diagnosis and treatment are complete. It is fair to try to make efficient use of time but not at the expense of the integrity of the game or the care provided. While the late Ernie Banks was quoted as saying “it’s a beautiful day fro a ballgame…let’s play two”, most people would rather get the diagnosis right the first time, no matter how long it takes, and avoid a second trip to the ER.

This entry was tagged , , , , , , , ,