This release summarizes an article in the Journal of the National Cancer Institute which reports that healthy lifestyle habits, including regular vigorous exercise, are associated with reduced risk of death from prostate cancer. The release does a fine job of describing a complex study that actually analyzed two studies and came up with a common scoring system, but could have explained the study limitations and articulated the benefits more clearly for readers. It also overstated the strength of the evidence by suggesting that these habits “save lives” — this study wasn’t capable of demonstrating that.
Prostate cancer is costly — in both lives and money. It is the most common cancer in U.S. men, according to the Centers for Disease Control and Prevention. In 2012, more than 177,000 men in the U.S. were diagnosed with prostate cancer, and 27,244 died from the disease. In 2006, it was estimated to cost the U.S. more than $9.86 billion. New information that can shed light on how lifestyle choices may limit deaths related to prostate cancer is both welcome and newsworthy. However, with so many people potentially affected (and likely to be affected), it is important to explain the benefits of those lifestyle choices in language that will be easily understood.
The study evaluated the combined effect of six lifestyle habits on prostate cancer mortality. Those six factors were, as described in the journal article: “not currently smoking or quit 10 or more years ago, body mass index under 30 kg/m2, high vigorous physical activity, high intake of tomatoes and fatty fish, and low intake of processed meat.” It’s tough to put a price tag on those lifestyle choices. How much does “high vigorous physical activity” cost? How much does it cost to avoid eating processed meat? It would be easy to say this category doesn’t apply. But the release does address the issue, saying the study findings underscore “the ongoing need for more effective prevention measures and policies to increase exercise, improve diet quality and reduce tobacco use in our population….It takes co-operation and effort from multiple areas, like insurance companies, employers, policy makers and city planners, to figure out how to creatively support and encourage more exercise into most busy adults’ working day.” Recognizing the need for policies that support these lifestyle changes is important — and addresses the cost issue in a meaningful way.
This is where things get a little tricky. The research described in this journal article looked at data from two different studies: the Health Professionals Follow-Up Study (HPFS) and the Physicians’ Health Study (PHS). The HPFS tracked 42,701 men from 1986-2010. The PHS tracked 20,324 men from 1982-2010. The release (and journal article) explain that the researchers used the HPFS data to develop a “healthy lifestyle score,” and then applied that score to the men in the PHS to see if it predicted whether someone would contract a lethal form of prostate cancer. The researchers identified the number of lethal cases of prostate cancer in each group to calculate benefits. The release states: “Participants with 5 to 6 points [the highest scores] in the HPFS had a 68 percent decreased risk of lethal prostate cancer[,] and a 38 percent decreased risk was observed in the [PHS] for the same comparison.” Several points to make here. First, the ‘healthy’ group appeared to have been compared to the ‘least healthy’ group (score 0-1) and not all the other groups combined. This was noted in the published results but not mentioned in the release. That’s important. Second, the release does not note (but the journal article does) that the 38 percent decreased risk seen in the PHS was not found to be statistically significant. That’s also an important distinction. Third, this is the relative reduced risk, and doesn’t tell readers what the actual, absolute reduction of risk is.
The release doesn’t address harms. But the lifestyle choices at issue here are (for the most part) so broadly defined that it would be difficult to say what the harms would be. What harms are associated with eating tomatoes, avoiding processed meats, or not smoking? We’ll rate this one not applicable.
The release does a good job of explaining what is a somewhat complicated study design. The study is basically an analysis of data drawn from two large, previous studies — and the release describes the overall sample size, age range, and length of those two studies.
But the language used to describe the results is inappropriate. Since this was an analysis of observational study data that’s unable to prove cause and effect, it’s misleading to use active verbs to describe the results. The release does this in several places:
It’s more appropriate to state that these habits were “associated with” reduced risk — not that the habits “save lives” or “cut” risk.
Another thing that’s not clearly explained to readers (as noted above) is what it means to develop a lifestyle scoring system based on one study’s data set, and then apply that scoring system to a second data set, when the results section is provided for both data sets. It’s confusing.
We’d encourage public affairs specialists preparing news releases and reporters using them as their initial sources to focus more on the limitations of this type of research. Sometimes referred to as “data dredging,” it’s basically finding statistically significant results only by chance. In this case it’s a statistic of a statistic (up to 47% of cancer cases could be prevented with these lifestyle interventions).
No disease mongering here.
Funding sources were clearly identified, and there are no clear conflicts of interest.
Things get tricky again here. As the release notes, most prostate cancers don’t metastasize and aren’t life threatening. But some do metastasize and can be fatal. Are there different types of prostate cancer, just like there are different types of flu? Are there genetic factors that affect risk? Readers would have benefited from that information.
And we suppose that the release could have also mentioned other dietary or lifestyle factors that might help reduce risk that were not considered in the paper. However, the range of healthy behaviors discussed here is pretty broad, and it seems excessive to expect the release to address an entire laundry list of possibilities. We’ll give this a Satisfactory rating.
The study was limited to a relatively high level income group and the interventions may not be attainable for all men, in all socioeconomic groups. Does partaking of vigorous physical exercise on a regular basis require access to a gym or just a high degree of motivation? Are tomatoes and fish attainable year round for all socioeconomic groups? The study addresses this to some extent when it says, “It takes co-operation and effort from multiple areas, like insurance companies, employers, policy makers and city planners, to figure out how to creatively support and encourage more exercise into most busy adults’ working day.” We’ll award a Satisfactory here.
There are a lot — a lot — of papers out there on lifestyle choices and prostate cancer. In fact, the opening sentences in the abstract for the journal article says as much, while articulating what sets this research apart: “Several lifestyle factors have been associated with risk of lethal prostate cancer, but little is known about their combined effect. Our objective was to develop and apply a lifestyle score for prevention of lethal prostate cancer.” However, the release does not acknowledge all of the existing work on lifestyle and prostate cancer, and therefore fails to highlight how this paper fits into that broad body of work.
Apart from the cause-and-effect statements addressed above, there is no unjustifiable language used here.