Share this article on:

News: It's All About the Numbers for a Treatment's Benefit or Harm

Coleman, Matthew

doi: 10.1097/01.EEM.0000425817.12006.1e


There are countless medicinal and procedural interventions in emergency medicine, but are they all valuable? David Newman, MD, the director of clinical research at Mt. Sinai School of Medicine, pointed out several effective — and ineffective — treatment methods in his lecture, “Number Needed to Treat,” at the American College of Emergency Physicians Scientific Assembly in October.

The number needed to treat (NNT), Dr. Newman explained, is the number of people who need to be treated with one intervention to benefit. Conversely, the number needed to harm (NNH) is the number of those who needs to be treated for one to be harmed by the intervention in a specific way.

Constituting the NNT is the absolute risk reduction (ARR), the literal reduction in the risk of the bad outcome. The relative risk reduction (RRR), on the other hand, is the risk of bad outcome compared with the baseline risk of the outcome.

Understanding these definitions is important because shared decision-making depends on these results, Dr. Newman said. “It is great that there is one of 10 people helped by an intervention, but if one of 10 people is harmed at an equal level, then you have skipped the most important part.”

Dr. Newman provided several examples of popular therapies in his presentation, including using thrombolytic therapy for acute myocardial infarction, which proved to have an ARR of 2.6 percent. “This means that one out of 38 people having a ST-elevation myocardial infarction will have their life saved if you give them thrombolytic, but half of these people are going to have an intracranial hemorrhage, so there is going to be a couple of people who are harmed,” he said.

Dr. Newman said some interventions have been shown to be positive:

  • Aspirin for stroke (saves one of 100 lives).
  • Reperfusion for MI (saves one to 30–40 lives).
  • Steroid for meningitis (saves one of 20 lives).
  • Hypothermia after cardiac arrest (saves one of six lives).
  • Rapid defibrillation (saves one of two lives).

Dr. Newman said some therapies, however, are unnecessary and even harmful.

“Stress tests for low-risk chest pain don't seem to help at all,” Dr. Newman said. “Sterile gloves for laceration repair don't do anything. Antibiotic use for bronchitis, strep throat, otitis media, and sinusitis had more harm than benefit.”

One of the more surprising mentions was proton pump inhibitors (PPIs) for gastrointestinal bleeding, which, was found to have no statistically significant difference in mortality, rebleeding, or surgery by a Cochrane review.

“This blows people away because we have been using it for so long,” Dr. Newman said. “If there is zero benefit, though, there will be no more harm, and, for no other reason than that, PPI may not be worth it.”

Copyright © 2012 Wolters Kluwer Health, Inc. All rights reserved.