Reporting on Medical Screening Trials: Context is Everything
The recent debate over CT lung cancer screening for heavy smokers is a great example of how big the stakes can be when reporting on medical screening, and how important it is to get the story right. Remember last year's furious arguments over the merits of breast cancer screening mammograms for women under 50?
To recap: the National Lung Screening Trial, a large, federally-funded study of 53,000 people, showed that screening former and current heavy smokers with expensive CT scans can reduce the risk of dying from lung cancer by 20 percent compared to chest X-ray screening.
Just days after the lung cancer screening study made headlines, a steelworkers union demanded immediate screening for its ostensibly high-risk members. One hospital quickly aired a television ad trumpeting its own CT scanner.
Studies like these also play an important role in when, whether and how private health insurers, and Medicare and Medicaid, will cover health screening tests.
So providing context and the necessary cautions for the National Lung Screening Trial – indeed, any research about medical screening - is critically important. (Remember the controversy over mammogram screenings for women under 50?) Some media did just that, notably NPR's Richard Knox, according to Gary Schwitzer of Health News Review. Others did not.
Where did some media coverage fall down?
1. Not including the "number needed to treat" to save a life. In this study, 300 scans were needed to save one life. That's a lot of expensive scanning with not-insignificant radiation exposure and the danger of false positives to consider. Antidote blogger William Heisel provides a great overview of this concept and how to calculate it when reporting on medical studies.
2. Reporting relative risk but not absolute risk. A 20 percent reduction in the risk of death from lung cancer sounds great, but how many people in the lung cancer study screening actually died from lung cancer? Of the 53,000 study participants, writes Jaan Sidorov in his Disease Management Care Blog, only 354 CT scan patients died of lung cancer, compared to 442 of the chest x-ray patients. Heisel offers tips for reporting on absolute vs. relative risk here.
3. Cost: not every story provided the cost of CT screening. Sidorov estimated the cost of helical CT scans, the type of scan studied, to range from $300 to $1,000. Making a few reasonable assumptions, he calculated that it would cost $180,000 to save one life from cancer.
Is it worth it? That's a question for ethicists (and, more likely, your insurance company) to ponder. In the meantime, remember that when reporting on medical screenings, context is everything.