8.1- Talking with the media

For the final unit of this course, we’re going to talk about how to reach broader audiences either by talking with the media or by writing about science for the general public. After you have published some papers, you may be contacted by journalists who are interested in your work. I encourage you to share your work with the media so that it can reach a larger audience. In this module, I’ll give you some tips for talking with the media. I’m a science journalist so I interview a lot of scientists. I’m going to give you some tips from the interviewers’ perspective. First of all, what are journalists looking for? When I’m interviewing a scientist I obviously want to understand the science. So if there’s anything I need clarified about the science I’m going to ask that. But most times I already have a good concept of the science. What I’m after in the interview are things that help put the story in context, give it a human dimension, and make it relatable to non-scientists. I’m hoping that the scientists can tell me what they think is the big picture significance of the work, the take home message. I also want to know how the research affects people. For stories about basic laboratory research. It’s nice if you can tell the journalist how your work could potentially affect people even in the long term. Also journalists have to write about new things. They have to have a news hook. There’s some reason why the story is being written about now. If you can tell a journalist what’s different or new about your results what distinguishes it from previous studies that’s very helpful. Journalists are looking for descriptive engaging language. So if you give me colorful prose, metaphors, descriptions, something controversial or lively or something that has some bite to it, I’m likely going to quote you on that. Make sure that if you are going to say something controversial or biting you don’t mind being quoted on it. Interesting stories or anecdotes like how you came up with the idea for this study are often great. Anything that’s surprising or paradoxical or ironical journalists love that because that makes the story compelling. If you were trying to make a weight loss drug and it turned out that it worked better as a cancer drug, that’s an interesting story that you might want to share with the journalist. Journalists are also looking for people-focused stories. For longer feature stories I may ask you things like what motivates you and what keeps you up at night because I want to bring a human dimension into the story. Researchers are people too and readers want to read about people. It’s nice to give some historical facts or something about the development of the idea. Journalists also need a first quote for their story. If it’s a story focused on a single study a sweeping comment about the significance of the work makes a great first quote. So try to have a statement prepared that clearly and concisely says why the work is important. Finally, you may be interviewed as a commentator on a peer’s research. If they have space, journalists try to include an outside perspective on the work. If you point out the controversies or you get very bold criticisms or laudatory praise that’s likely to make it into the article. So just be careful about that. Remember, journalists are looking to be able to quote you directly. This means you need to speak clearly and concisely and in layperson terms. If your language is too complicated or full of jargon or boring they will have nothing to quote you on. So be prepared. It may help to craft a few take home messages ahead of time and write these down. Make sure they’re clear and easy to understand in the active voice without too much clutter or jargon. You can read these to the journalist and that’s what you may end up being quoted on. It helps to pretend that you are talking to an intelligent relative or friend who isn’t a scientist. You may want to picture your uncle or your grandmother or your grandfather somebody that you might be explaining your science to over the dinner table who’s not a scientist themselves. Make sure you give the journalist clear take home messages. Don’t be wishy-washy about what the findings mean. State clearly what you think the journalist should focus on in their piece. Try to anticipate any confusions or misinterpretations that the journalist or the lay public might make about your work especially if it’s something controversial. You can say to the journalist here’s something that people might misunderstand about my work and let me tell you why that’s wrong. Also make sure to give a clear statement of the key limitations of your work. The journalist might not ask for that but it helps the journalist to put your research in context. It keeps them from overblowing or overstating the findings. Finally, think carefully about how you present numbers to the journalist or any general audience. Not everybody is comfortable with numbers so it’s important for you to think about the most transparent way to present numbers. For example most people understand one out of 100 better than they understand one percent. And most people don’t have a good sense of what say a nanometer is. So if you use the term nanometer you need to say how small that is. If possible compare it to something people are familiar with. So you might say a nanometer is one billionth of a meter and the width of a single human hair is about 80,000 nanometers or the thickness of a sheet of paper is about 100,000 nanometers. It gives people something concrete to understand. Because I think this is so important. I’m going to go into more detail on explaining risk to journalists and the lay public. If you’re talking about risks make sure you present those risks in the most transparent easy to understand way that you can. People are bad at understanding risk. We tend to be very frightened of things that have a low probability of occurring whereas we are not adequately frightened about things that occur more commonly. So we fret about terrorist attacks which occur with an extremely low probability but we ignore the risk of heart attacks which are way more likely to kill you even if you’re young and healthy. Presenting risks in a way that’s easy to understand is a great public service. A couple of principles to keep in mind wherever possible prefer whole numbers to fractions or percents say 10 in 100 rather than 10 percent. Also wherever possible opt to present absolute risks rather than relative risks relative risk can be misleading. For example if an exposure increases your risk of cancer from one in 10,000 to two in 10,000 you can call this a doubling of risk but you better also give the absolute values as well so that the reader understands that the absolute increase in risk is just one in 10,000. I want to share a case study with you. This is a case where the researchers of the study and the people who put out the press release did an excellent job of describing the risks. I was working as a journalist at the time and when the press release came across my desk. I was so impressed that they had thought carefully about the way to present the numbers. They used absolute risks and whole numbers. The study was the Women’s Health Initiative. This was a large randomized, double- blind study where they were treating post-menopausal women either with post-menopausal hormones or with a placebo pill. At the time the study was started in 1990s it was taken for granted that post-menopausal hormones reduce heart disease in women. That was the dogma of the time but it had never been proven in a randomized trial. In 2002, the researchers of the trial did an interim data analysis where it became clear that in fact the dogma was wrong. In this study, the women who had been put on hormones had a significantly increased risk of both breast cancer and heart disease. It became clear partway through this study that the risks of hormones were outweighing the benefits so they stopped the trial early. At the time they stopped the trial 14 million post-menopausal women in the U.S. were taking hormones. The results were a bombshell. So the researchers had to be very careful about not scaring 14 million women. The researchers published the results in JAMA at the same time that they halted the trial and it was widely covered by the media. In the JAMA paper the researchers presented the results in multiple ways. Okay? One of the numbers in the JAMA paper was the relative risks. The relative risk comparing hormones to placebo for breast cancer was 1.26 and for heart disease was 1.29. And the way you interpret that, what those mean, is that women who take hormones have a 26 percent increased risk of breast cancer compared with women who take placebo and a 29 percent increase risk of heart disease. But is that the best way to present the numbers for the lay public? Those numbers sound very dramatic they sound frightening. That sounds like a big increase in risk. If the press release had focused on those numbers I’m sure that the media would have focused on those numbers as well. But the paper also looked at the numbers in a different way. They presented the numbers in terms of absolute risks. The rate of invasive breast cancer was .30 percent in the placebo group versus .38 percent in the hormones group. The risk increased there is only .08 percent per year. And similarly for heart disease the risk increase was .07 percent per year. I’ll just note for those of you who are paying attention to the math here that if you divide .38 percent by .30 percent that gives you roughly the relative risk but it’s not exactly 1.26 because slightly different models were used to generate the relative risks than the absolute risk. So if you notice that they don’t exactly line up that’s why. When you hear the numbers presented this way it’s clear that the risk to any one individual woman is actually quite small. When 14 million women are taking hormones an increase in risk of .07 percent or .08 percent translates to lots of cases of heart disease and breast cancer. But for any individual woman within that 14 million her personal risk is not increased by very much. As I’ve mentioned most people are not adept at thinking in percents and fractions. So we can do even better by translating those percents into whole numbers. In the press release from the Women’s Health Initiative they stated that the risk of breast cancer was increased from 30 cases per 100,000 per year to 38 per 10,000. That was an eight case per 10,000 increase. Those are numbers that most people can understand. Similarly for heart disease there was a seven case increase per 10,000 women per year. That’s the way the numbers were presented in the press release and that’s the way the numbers were presented in the media coverage of this study. So which is the best way to present the numbers for the public I’d argue that the absolute risks and whole numbers are much more transparent. The relative risks obscure the fact that the base line risk here is quite small. The relative risks are needlessly scary and shocking. I was impressed that whoever drafted the press release from the Women’s Health Initiative was thinking very carefully about numbers. Again they presented this the second way here the eight more and seven more cases per 10,000. And guess what all the major media outlets that covered this story presented those numbers. Most did not even report the relative risks. This case study goes to show you that as a researcher you have a lot of control over the numbers that the press focuses on if you give journalists the right numbers. The most transparent numbers those are the numbers that will likely appear in their stories. If you don’t help them with the numbers they’re likely to just report the most dramatic numbers from your paper. So think carefully about the presentation of risk and try to put things in absolute risks and whole numbers when talking to journalists.