We wished that some of the strong elements included in this story could have found their way into some of the other coverage of a study about the effects of spiral CT scans on reducing lung cancer deaths. Conversely, we wish this story had included more of the information necessary to give readers a complete picture of the benefits of this type of screening for cancer.
Smoking rates and the incidence of lung cancer have been falling in the US for several years. Despite these gratifying trends, about 225,000 Americans will be diagnosed with lung cancer in 2010 and approximately 157,000 will die from the disease. A major factor in survival after diagnosis is early detection. Chest X-rays are not sensitive enough to pick up lung cancer in its early stages. That is why more sophisticated diagnostic testing has been sought. Spiral computerized tomography (CT) has been touted as the best way to diagnose lung cancer early and save ives. This landmark study has demonstrated that spiral CT’s are indeed more senstive than chest X-rays and that early detection can save lives. The study is by no means a complete answer, and it raises many questions. Who should be screened, when, how often and who is going to pay the enormous bill are some of the obvious questions. Stories like this need to give the evidence an independent and critical analysis to help readers understand whether more screening is necessarily better for identifying lung cancer and saving lives.
The cost of diagnostic testing and the return on that investment in both clinical and financial terms should be routinely included in any story. Like many others, this story fails in this regard. Simply describing the cost of the diagnostic test does not provide the reader with an appreciation of the true cost of the testing. The number of people you need to scan to find one with an early stage lung cancer is 304. There are 33 million Americans who smoke daily. At $300 to $1,000 per scan, scanning everyone would cost $9.9 to $33 Billion annually. Any discussion on cost should also include comments on the false positive rate. In this case, 25% to 60% of subjects had a lesion that needed further workup. The financial costs of these false results should have been noted
The story presented some of the raw, absolute numbers, although we wished it had gone further. It says, “As of last month, 354 people who got CTs had died of lung cancer, compared to 442 who got ordinary X-rays. This worked out to a 20.3 percent lower risk of dying for the spiral CT group and the researchers stopped the study.” What it should have added was that, assuming 50 percent of all participants were put in each group, that would mean 1.32% died in the CT scan group and 1.65% died in the X-ray group. That’s an absolute difference of 0.33%, not quite as exciting as a 20% reduction in deaths. Also, the number of people needed to treat to save one life (or NNT) is 304. That may be a justifiable number for added investment in spiral CT scans, but it should be part of the discussion. Nonethless, we’ll give it the benefit of the doubt on this criterion.
The story misses the mark completely in its handling of the potential harms of the test. It does provide the important caveat that “Critics of the scans fear that smokers may not be motivated to quit if they believe screening can save their lives if they do get cancer.” It fails to note the relatively high false positive rate (25% to 60%) and the implications on unnecessary additional testing. This is especially disappointing because the NCI press release provided the information.
The story did a better job than some of the other coverage in evaluating the quality of the evidence, but it left readers confused by bringing up an older, potentially tainted study and not carefully comparing the two studies. It says, “In 2006, Dr. Claudia Henschke of New York Presbyterian Hospital-Weill Cornell Medical Center caused a stir when she published a study saying that 80 percent of lung-cancer deaths could be prevented through widespread use of spiral CT. Her ideas were controversial to start with and widely discredited when other researchers found her work had been paid for by a tobacco company.” But then it makes no attempt to talk about whether this new study is an improvement on Henschke’s work or what makes this study different. The press release from the National Cancer Institute provided those who bothered to read it with all of the important points of the study, both positive and negative. This story failed to take advantage of the information provided. The space dedicated to the 2006 study by Dr. Henschke could have better been used to more completely describe the study results and their implications.
The story avoided disease mongering and provided some great context. It says, for example, “the single best way to prevent lung cancer deaths is to never start smoking, and if already smoking, to quit permanently.” It also says, “About 10 percent of smokers develop lung cancer, but smoking causes other cancers as well as heart disease and stroke.”
The study raises many questions that will likely be debated for years by the experts. The failure to provide a single independent voice to the discussion is curious to say the least.
The study compared spiral CT scanning to chest X-rays and did mention smoking cessation.
This story did a better job than others in actually spelling out how commonly used these spiral CT scans are, noting that “Almost all advanced CT scanners can perform a spiral CT, and about 60 percent of U.S. hospitals have such a machine. Makers include General Electric Co’s GE Healthcare Siemens AG, Toshiba Corp, Hitachi and Philips.”
The story does note previous work in the early detection of lung cancer using spiral CT scanning.
The story did rely a little too heavily on press releases. For example, one quote from the study’s leader, Dr. Denise Aberle, is from this release. And her other quote and a quote from Dr. Christine Berg, are from this release. Only the Berg quote is attributed to “a statement.” To lift quotes from news releases without attribution is not good practice.