NNT can be TNT for blowing up pharma marketing claims
A new drug comes on the market that promises to improve people's eyesight. "Clarivue! Make your cloudy days sunny again!"
Your editor says, "This Clarivue is like Viagra for eyeballs. It's going to be flying off the shelves. Write up something for the Web in the next hour."
Your next move should be to find out the NNT: the number needed to treat. It will help you answer the most important question: How many people would need to take Clarivue in order for one person to actually see better?
This is the number that John Carey used as the springboard for his revelatory look at statins in BusinessWeek. It's the number that two University of Oxford professors have said "is simple to remember and directly supports efforts to work with patients to make the best possible clinical decisions for their care."
Put another way, the NNT is the number of patients that would need to undergo a particular treatment over a specific time period in order to see their health improve beyond what would have happened had they done nothing or had they undergone a different treatment.
A low number means a drug is highly effective with a broad range of people. Four people take it, and one sees better. For a vaccination, for example, the NNT might be one. A high number means the drug may be effective but only with a narrow set of people.
As Carey pointed out, Pfizer has advertised Lipitor as reducing the risk of heart attack by 36%. But the NNT is actually 100. That "crucial point is hiding in plain sight in Pfizer's own Lipitor newspaper ad," Carey wrote:
The dramatic 36% figure has an asterisk. Read the smaller type. It says: "That means in a large clinical study, 3% of patients taking a sugar pill or placebo had a heart attack compared to 2% of patients taking Lipitor."
Now do some simple math. The numbers in that sentence mean that for every 100 people in the trial, which lasted 3 1/3 years, three people on placebos and two people on Lipitor had heart attacks. The difference credited to the drug? One fewer heart attack per 100 people. So to spare one person a heart attack, 100 people had to take Lipitor for more than three years. The other 99 got no measurable benefit. Or to put it in terms of a little-known but useful statistic, the number needed to treat (or NNT) for one person to benefit is 100.
Compare that with, say, today's standard antibiotic therapy to eradicate ulcer-causing H. pylori stomach bacteria. The NNT is 1.1. Give the drugs to 11 people, and 10 will be cured.
Of course, very few drugs have an NNT near 1. The key – as with writing about absolute risk when writing about relative risk – is to provide readers the full picture of a drug's benefit. Is it worth it for 40 people to spend $100 a month for three years just so one of them can avoid a heart attack? Maybe so, especially if it is your spouse, child, parent or sibling who avoids the heart attack, but what if that number jumps to 400? And what if your employer (General Motors) is going out of business because of skyrocketing medical costs or your government is footing the bill?
Along with Carey's article, BusinessWeek ran a stunning graphic showing the NNTs for several drugs that illustrated how NNTs can be different depending on the type of people taking the drug. Statins work wonderfully in people who have had a heart attack or who have signs of heart disease. In that group, the NNT is between 16 and 23. If you've been to the ER with your life flashing before your eyes, you like those odds. For people without heart disease, the NNT shoots up to higher than 500. That should give pause to people who like to say that statins should be put in drinking water. (Carey's example of Lipitor having an NNT of 100 was based on a specific clinical trial.)
NNTs are not often called out in drug studies, especially the ones sponsored by drug companies, but if a company is pushing a new drug as a breakthrough, you should ask for the NNT. Chances are good that you will have to follow Carey's footsteps and calculate the NNTs for yourself. (You can find more great examples of NNTs here.) But the math is simple. Let's recap using another expert source, Dr. Leslie Citrome at New York University School of Medicine in her excellent Journal of Family Practice article on NNTs.
First determine the difference between the frequencies of the outcome of interest for two interventions.
Then calculate the reciprocal of this difference.
For example, let's say drugs A and B are used to treat depression, and they result in 6-week response rates of 55% and 75%, respectively. The NNT to see a difference between drug B versus drug A in terms of responders at 6 weeks can be calculated as follows:
Difference in response rates=0.75–0.55=0.20
In this example, you would need to treat 5 patients with drug B instead of drug A to see 1 extra responder. If the NNT had been 5.5, you would round up to the next whole number (6) because you can't treat a fraction of a person.
Interpreting the importance of NNT values is easy, too. The smaller the NNT, the larger the clinical difference between interventions; the larger the NNT, the smaller the difference.
An NNT of 100 or more usually means little difference exists between interventions for the outcome of interest.
An NNT of 2 would be hugely important and is rarely encountered.
So the next time an actual Clarivue pops into view, read the fine print and find your way to the NNT.