aspirin, statistics and super bowl coaching

Monday, February 2, 2015

In medicine, much is made of statistical analysis and its importance for determining which test might be appropriate to make a diagnosis and what treatment might follow. The rigors of statistics try to persuade even the most skeptical person about the benefit or risk of a situation.

Consider this situation. A decision needs to be made that in one analysis will be successful 20% of the time, will have no change in 40% and will result in harm the other 40%. Another study looking at similar data over a longer period of time and concluded that success would be achieved only 41% of the time. It would be difficult to recommend the routine use of such a treatment unless there was a compelling situation. With just a few seconds left in this year’s Super Bowl, the Seattle Seahawk coaches chose to pass the ball on the one yard line, instead of letting their “beast mode” running back Marchand Lynch carry the ball. Disaster befell the team as the pass was intercepted, New England wins and Pete Carroll is thought to have made the worst coaching blunder in the history of football.

Back to the numbers from statistics pros Nathan Jahnke, and Neil Greenberg. Jahnke writes for profootballfocus.com and reports that in the 2014 season, Lynch had 5 attempts from the one yard line and had scored once, had no gain twice and had lost yardage twice (20%, 40%, 40%). Greenberg, of the Washington Post, reviewed Lynch’s whole career and found that he scored from the one yard line in only 15 of 36 attempts (41%). Though much of statistics are mental gymnastics, this time it seems that the national outcry for the Seahawks to run the ball may not have been 100% warranted.

Monday morning quarterback is a tough position to play, even though most people feel well qualified and sometimes compelled to assume the role. It begins with knowing what happened and then presupposing that a positive outcome would occur had a different, “better” decision been made. The basis for most of these day after analyses often rely upon dogma and how can that ever be wrong. Dogma is a set of principles or facts that laid down by an authority that are incontrovertibly true. In medicine, changing dogma is an arduous process akin to changing the course of a large ship. It takes great time and effort to alter its direction since momentum can be a very powerful force to overcome.

Aspirin is a great drug and is recommended in many situations and its use to decrease the risk of heart attack and stroke is correct…but at what cost. According to researchers, regular aspirin use may decrease the risk of nonfatal heart attacks by 20% and all heart events by 10%. However, those aspirin users were 30% more likely to have a serious gastrointestinal bleed (bleeding from the stomach lining or ulcer). There is a concept of numbers needed to treat. In this case, it takes 162 people taking daily aspirin to prevent a nonfatal heart attack, but in that group, two people would develop serious bleeding episodes. Aspirin helps make platelets less sticky, decreasing the ability for blood to clot, but it also is irritating to the lining of the stomach and intestine. Remember that aspirin is really an acid, salicylic acid.

There are good studies that show that aspirin is very beneficial in patients who have already had a heart attack, decreasing the risk of a second heart attack by 20-30%. For people with atrial fibrillation, aspirin may be the drug of choice to thin the blood and prevent stroke. The key for doctors and patients is to individualize the use of aspirin to the patient’s clinical situation and balance risk and reward. The US Preventive Services Task Force recommends the use of aspirin on a case by case basis. The dogma that everybody should take aspirin is being tempered with the daylight that comes when researchers ask why a treatment is being recommended. “Because it has always been done that way” is no longer an acceptable answer.

It’s always good to hold dogma’s feet to the fire and quarterbacking on a Monday morning offers a prime opportunity to do so. The success of day after opinion is always 100% but that statistic is also open to interpretation.

PLEASE DO NOT STOP TAKING ASPIRIN IF IT HAS BEEN RECOMMENDED BY YOUR DOCTOR.

This entry was tagged , , , , , , ,


death and taxes

Monday, April 28, 2014

While the NFL may try to lay claim as America’s national sport, there is no doubt that baseball is king of statistics. The basics of hitting, batting and fielding averages have matured to an alphabet soup of performance indicators and averages, all to judge talent and project performance. Numbers matter in baseball from a single year and the .300 batting average and winning 20 games, to lifetime achievements, whether you believe in Hank Aaron’s 755 or Barry Bonds’ 762. Numbers matter in medicine but sometimes they frustrate. Welcome to the rant.

For all that can be done to improve the quality and quantity of life, it is what the patient brings to the table that may be most important. John F. Kennedy perhaps best summarized the ideal for medical care in this country when he asked his fellow Americans: “ask not what your country can do for you — ask what you can do for your country.” And in the case of lung cancer…not much on both sides.

Last year, an estimated 228, 190 people were diagnosed with lung cancer, a close third behind prostate and breast cancer, give or take a couple thousand patients. Last year 159,480 people died of their lung cancer, almost three times as many that died of prostate and breast cancer combined. Those are somber statistics, especially when $315 million were spent by the National Institutes of Health on cancer research.

More statistics. Smoking is the principal factor that causes lung cancer and is responsible for 90% of new cases per year. Radon is responsible for almost all the rest of ling cancers. It is an odorless gas that is released from elements found in earth and rock when they naturally decay. Exposure occurs in the workplace, especially in miners who work underground and in house basements, especially houses that are relatively airtight. Radon causes cancer more easily in smokers and the risk of cancer increases when smoking and radon exposure mix. Still of the 20,000 or so cancers caused each year by radon exposure, about 10% occur in nonsmokers.

It would seem that the best way to prevent lung cancer would be to get rid of smoking. The Surgeon General has had warning messages on cigarette packages since 1965 and every adult qualifying for Medicare has known since they were 18 years old that smoking kills and while tobacco has no redeeming medical benefit, about 20% of the US population smokes.

There is, however, a major incentive for government and your local politicians to be thankful for the 42 million Americans who smoke because they generate $17 billion in tax revenue at both the federal and state level. That is $17 billion with a B, is more than 53 times the amount spent on lung cancer research. Of course, smoking is also a major risk factor for the big killers, heart disease and stroke. Money is spent on research for those diseases and it costs money to care for those afflicted. But the numbers don’t lie. Income from smokers is on the plus side of balancing the government budget.

There are three types of lies according to Disraeli, Twain and others: Lies, damn lies and statistics. One can reach numerous conclusions from the same numbers. In baseball, I choose to believe that Hank Aaron holds the home run record, regardless of what the record books say and I get to interpret baseball stats that way. I also choose to believe that there is no joy in Mudville every time a new smoker is conscripted into the population. There is also no political will to take on the tobacco companies and the gas stations who are the prime sellers of tobacco products. A million here, a million there and pretty soon you’re talking real money…like $17 billion. In tight economic times what politician wants to raise taxes by $17 billion…unless they think there is a market to tax E-cigarettes? Stay tuned for more statistics.

 

This entry was tagged , , , , ,