Nonprofit, nonpartisan journalism. Supported by readers.


UCare generously supports MinnPost’s Second Opinion coverage; learn why.

Medicine’s ‘biggest secret’: number-needed-to-treat

The number-needed-to-treat (NNT) is the number of people who would need to receive a drug to ensure that one person gets better.

Over the past year, the Milwaukee Journal Sentinel and MedPage Today have jointly published an excellent series of investigative articles on the topic of “illness inflation” — a strategy used by pharmaceutical companies to take a well-established clinical syndrome and expand the diagnostic criteria to include a greater number of people.

Another name for this phenomenon is “disease-mongering.”

The eight conditions that reporters John Fauber, Kristina Fiore and Matthew Wynn have examined over the past 12 months are overactive bladder, female sexual interest/arousal disorder, attention deficit hyperactivity disorder (ADHD), low testosterone in men, intermittent explosive disorder, binge-eating disorder, pre-diabetes and premenstrual dysphoric disorder

As the reporters have pointed out, none of these inflated conditions is life threatening, and any benefits from the drugs recommended to treat them are marginal.

But the drugs have the potential to lead to unwanted and sometimes dangerous side effects.

“Since 2013, nearly 65,000 reports of serious side effects involving drugs used to treat five of the conditions have been reported to the FDA’s database,” report Fauber, Fiore and Wynn. “That includes 1,631 deaths.”

‘The biggest secret in medicine’

For a new article in the series, published Wednesday, the Milwaukee Journal Sentinel and MedPage Today asked Dr. F. Perry Wilson, an assistant professor of medicine at Yale University School of Medicine, to conduct a benefit-harm analysis for four drugs — Toviaz, Addyi, Vyvanse, and Androgel — that have been approved to treat five inflated conditions: overactive bladder, female sexual interest/arousal disorder, adult ADHD, binge-eating disorder and low testosterone in men. (Drugs used to treat the other three conditions highlighted in the Journal Sentinel/MedPage Today investigative series — intermittent explosive disorder, pre-diabetes and premenstrual dysphoric disorder — were not included in the analysis due to a lack of reliable data or because no drug has been approved to treat them.) 

Wilson began his analysis by determining what he calls “the biggest secret in medicine”: the number-needed-to-treat (NNT). This is the number of people who would need to receive a drug to ensure that one person gets better.

Wilson then balanced the NNT for each drug against the number-needed-to-harm (NNH) — the number of people who could be expected to be treated with a drug before one of them was harmed by it.

As Wilson explains in an online video that accompanies Wednesday’s article, few people are told either number by their physicians. When patients do ask their doctor about a drug’s effectiveness, “they usually get a response like, ‘Well, it seems to help most people,’ or something like that,” he says.

Benefits vs. harms

So what did Wilson find in his benefit-harm analyses?

In the case of Vyvanse (lisdexamfetamine), an amphetamine approved to treat both adult ADHD and binge eating disorder, the NNT is 2.9. That means that about three people have to take the drug in order for one person to notice some relief of symptoms.

That must be balanced against an NNH of 5. “For every five people who take it, one will get insomnia, after factoring out the placebo effect — those in the control group who saw improvement simply by taking a sugar pill,” write Fauber, Fione and Wynn.

Vyvanse, like other stimulants, also carries a high risk of dependence and abuse, and can raise blood pressure and heart rate, as well as the risk of a heart attack or stroke, they add.

For Toviaz (fesoterodine), which is used to treat overactive bladder condition (once known as incontinence, or the inability to stop urine leakage, but since expanded to include a sudden urge to urinate and having to get up at night even without incontinence), the NNTs are even larger: 3.6 for leakage (one out of every 3.6 people who use the drug) and 5.7 for reduce urgency.

The NNH for one of the drug’s side effects, dry mouth, is also 3.6, while that for constipation is 25.

But even those numbers don’t seem so problematic compared to the NNTs for Addyi (flibanserin), which is used to treat low libido in women, and for the testosterone supplement Androgel, which is used to increase sexual activity in men.

Twelve women must take Addyi before one experiences a modest improvement in her sex life (after ruling out the placebo effect), Wilson reports. For Androgel, it’s one in nine.

But one in seven women taking Addyi will experience drowsiness, and one in 14 men taking Androgel will develop acne. One in 33 of men using Androgel will also experience disturbing mood swings, such as impatience or anger.

“In short, the drugs won’t help most of the people who take them,” write Fauber, Fiore and Wynn. “And, in some cases, they are almost as likely to produce a side effect as a benefit.”

The reporters contacted the manufacturers of all the drugs analyzed by Wilson. Spokespeople for each said their company stands behind the safety and effectiveness of their product.

Know the numbers

As Fauber, Fiore and Wynn explain, NNTs are likely to overstate the effectiveness of a drug, as they are usually based on industry-sponsored clinical trials that do not reflect “real-world” situations.

“It’s Disney World versus the real world,” Dr. David Juurlink, a professor of medicine at the University of Toronto who studies drug safety, told the reporters. “The point is that, by design, industry-sponsored [trials] tend to make drugs appear more effective and safer than they are eventually found to be in practice because the [trials] milieu is often an artificial one.”

NNHs may also not reflect reality, as not all adverse side effects from drugs get reported. Many patients are unaware that symptoms they develop after starting on a drug are related to the medication.

But NNTs and NNHs are among the best tools we have to make informed decisions about the medications we are advised to take.

Ask your doctor about those numbers. Then decide if the potential benefits outweigh the potential risks.

FMI: You can read the latest article in the Milwaukee Journal Sentinel/MedPage Today series on inflated conditions on either publication’s website (here or here). I also recommend spending some time watching Wilson’s video explanation of why he believes NNT is “one of the biggest secrets in medicine.” To obtain NNTs and NNHs for more drugs, go to This website was started a few years ago by a group of physicians who wanted to develop “a framework and rating system to evaluate therapies based on their patient-important benefits and harms.” The site accepts no outside funding or advertisements.

Comments (3)

  1. Submitted by Paul Brandon on 12/15/2016 - 09:41 am.

    Good points!

    As long as American medicine is a private industry it will be driven by profits.
    In that case market size becomes a dominant measure.
    Another point that might be made is individual vs. population decisions.
    If a drug yields a 5% absolute risk reduction, that might not justify its use for any given individual. However, if one is making decisions for millions of people, that 5% rate could translate to millions of people helped.

  2. Submitted by Ray Schoch on 12/15/2016 - 09:45 am.

    Yet another example

    …in a very, very long list, of why our alleged health-care system turns out to be more of a profit-taking system than anything else, with actual patient health somewhere down the list, sometimes far down the list, of priorities for health insurance and, especially, pharmaceutical companies. Note that, with rare exceptions, really effective drugs, for genuinely serious medical conditions, are generally NOT advertised on television.

  3. Submitted by Alan Straka on 12/15/2016 - 04:11 pm.

    Androgel (testosterone) is not primarily used to increase libido. An increase in libido is generally considered a side effect, whether beneficial or detrimental depends on your point of view. To say it is the main reason to prescribe it is to set up a straw man to discredit testosterone therapy. Testosterone has a great many effects besides increasing libido such as reducing fat, increasing muscle, anti-depressant, heart health, treating ed and so on. As for side effects, all drugs have them and a few zits isn’t a bad trade-off for the potential benefits and the acne usually clears up over time. The problem is that many doctors don’t test testosterone levels before writing a prescription. If your testosterone levels are normal, you won’t reap any benefits from taking more. If you actually do have low levels as well as symptoms of low testosterone, a prescription might very well be what you need. If your symptoms persist, you move on to something else, if they go away then you have a valid therapy. No drug works for everyone but that doesn’t mean it is worthless for everyone.
    It is very true that drug companies have over hyped testosterone. Wouldn’t you if you could charge hundreds of dollars for a few cents worth of drug? It is hard to blame a business for trying to make as much money as they can. The real villains are the FDA and Congress who do everything in their power to make sure that the drug companies can go on making obscene profits.

Leave a Reply