The Anatomy of a Press Release

I write 1,000-page biology textbooks. Tomes. So when in early June a nice public information officer (PIO in media lingo) from Columbia University called to ask if I'd like to write a press release--a mere page or two--I jumped at the chance. Pay was minimal (and yet to arrive) and a byline nonexistent, but I would have a shot at actually making the news, to distill the essence of some exciting new research result in a way that might make it into the mouth of NBC's Tom Brokaw or the pen of The New York Times' Nicholas Wade. "Have you ever heard of something called 'gp120' binding to something called 'CD4,' or 'X-ray crystallography'?" asked my Columbia connection. I answered in the affirmative, explaining in a torrent why the news was exciting, that viewing the choreography of HIV's gp120 approaching a T cell's CD4 receptor would reveal new targets for vaccines and therapeutic interventions to fight AIDS. Papers were being published in Science and Nature (C.D.Rizzuto et al., Science, 280:1949-53, June 19, 1998; P.D. Kwong et al., Nature, 393:648-59, June 18, 1998; R. Wyatt et al., Nature, 393:705-11, June 18, 1998) and would be overnighted to me. The Columbia PIO had called me because exciting work was happening too quickly for the release-writers to keep up, and she had read some of my work. And so I had an assignment to write a press release. I didn't tell her it was my first. But, I had been consulting press releases for years as I conducted research for magazine articles, and I felt I knew the beast well enough to attempt to create one. The novelty and challenge of writing a short, snappy account of complex research intrigued me, after having experienced science at the bench and as a science writer for various magazines. Writing the press release would prove revealing in terms of which parts actually made it into the news, and which didn't. A press release announces important and/or interesting new work. At a minimum, it must include the researchers' names, titles, and affiliations; a clear statement early on of what the work shows and why it is important; where results are being published with citation information; and how a journalist can contact a source. The PIO who typically writes the release then runs interference, fielding questions, supplying photos, perhaps coordinating a press conference, and setting up interviews between journalists who take the bait and the otherwise hard-to-reach researchers. A terrific press release includes background and context, quotes from the researchers, and perhaps comments from other experts, all fair game for print or broadcast journalists to use. A mediocre press release might lapse into hype mode, extolling a "breakthrough" that will simultaneously cure baldness, yeast infections, African sleeping sickness, and global warming--and allow you to clone yourself (B. Palevitz and R. Lewis, The Scientist, 12[15]:7, July 20, 1998). Exaggerated, perhaps, but not a far cry from this real intro from a medical society's recent press release: "A breakthrough study on in vivo gene therapy may someday provide the medical know-how to eliminate inherited metabolic disorders." The "breakthrough" turned out to be a study of a very rare inborn error of metabolism in mice. Pediatricians were heavily quoted in the first half of the release, with nary a mouse mention till near the end. Scientists can control how their research is presented to the public by being accessible to PIOs, and by carefully checking drafts of press releases for a host of problems: errors, oversimplification, misinterpretation, insensitivity, or cutesy phrasing (see R. Lewis, The Scientist, 10[1]:15, Jan. 8, 1996, and 10[23]:11, Nov. 25, 1996). Although such errors are often not obvious to the nonscientist, they can cause a researcher to cringe. Consider a press release that ran several times during the development of Herceptin, the recently approved monoclonal antibody-based breast cancer drug. The narrative of the press release refers several times to the "overabundance of the HER-2/neu gene," whereas the quoted experts more correctly refer to the gene's overexpression. I envisioned biology students everywhere picturing a chromosome peppered with HER-2/neu genes. The narrative also includes, "Sentenced to a painful death by her former doctors more than 3 years ago, little evidence can be found today of the aggressive breast cancer that had spread to her liver." The PIO assured me that the statement was accurate, but as a cancer survivor, I feel it could have been phrased in a more sensitive manner--it frankly gave me the willies. Many newspapers that picked up the press release verbatim included both the genetic oversimplification and the "painful death" comment. I had great fun writing the press release. Instead of having to pursue scientists to coax an interview, the members of the research team awaited my call. I spent a pleasant Friday afternoon (June 19) chatting with Peter Kwong, the young researcher who did the bulk of this very elegant work over the course of a decade, then on the following Monday I spoke with his mentor, Wayne Hendrickson from Columbia, and Joseph Sodroski at the Dana-Farber Cancer Institute. Quotes from all three told the story in the press release. My greatest challenge was to explain the results of the experiments, but I also tried to capture how they were done--by X-ray crystallography. I opted for a straightforward description, mentioning the technique's pivotal role in revealing the structure of DNA in 1953, and more recently, in developing protease inhibitors. Then I sat back to see what would happen. A press release is distributed to journalists in various ways, often on the Web in password-protected sites, usually a few days before the study is published. Journalists pledge not to break this embargo. The gp120/CD4 story, which I zapped to Columbia on Monday, June 15, to be released later that day, was embargoed until Wednesday the 17th. All in all, I thought the press coverage was excellent, with reporters conveying the nature and importance of the work quickly and succinctly. But the hectic pace of newspaper coverage led to some watering down of exactly how the researchers got this new peek at HIV. The Associated Press merely wrote that the team "figured out the 3D shape," burying Peter Kwong's identity in an "and other scientists" phrase. The New York Times wrote about "X-ray pictures" and the molecular interaction being "extremely hard to X-ray," as if an HIV particle ambled into a hospital radiology department for a medical scan. The Times typically interviewed several top AIDS researchers who all said wonderful things early in the article, the "round up the usual suspects" approach that lent a gee-whiz ambiance to an otherwise excellent account. Some articles appearing soon after the embargo expired didn't mention at all how the work was done; they just referred to "molecular snapshots." As news reports began to accumulate with hardly a mention of the venerable X-ray crystallography, I slowly realized the reason--most people don't care how things work, and how discoveries are made. These are the questions that a scientist asks first--HOW?--but the answers may be what alienates regular nonscientific folk from science. And, as long as people are reading about what the researchers did and why it is important, I guess it is OK if the how-to is omitted. A few days later, descriptions of X-ray crystallography began to creep into magazine articles. Time magazine called the technique "an exacting imaging technique .... that shows the precise spatial relationship of atoms within a large molecule ..." and that it was able to reveal "structures ... at the bottom of crevices, or obscured by great forests of sugar molecules." Christine Gorman's description was so lyrical that she didn't need quotes or names or even a press release. Science magazine's news article on the work, by writer Michael Balter, seemed to borrow heavily from the battle-themed "lede" (journalese for opener) of my press release (M. Balter, Science, 280:1833-4, June 19, 1998). This was either a coincidence (likely), or a sign that I wasn't a total failure at this first press release endeavor. Balter, with more time than the newspaper writers, explored Kwong's contribution in depth in his detailed article. Nature's accompanying news article, by two scientists, unfortunately ran right after an article entitled "101 Uses for Fossilized Faeces," certainly a tough act to follow (J.P. Moore and J. Binley, Nature, 393:630-1, June 18, 1998). Guess which story made a bigger news splash? I interviewed the excrement expert for a Notebook item in The Scientist, and he admitted to not actually knowing 101 uses. Press releases undoubtedly ease a journalist's life, but they bother me because reliance on them eliminates much research from the running. Someone, at some point, has to decide which work to publicize. This task might fall to a department chairperson or committee, a public information officer, or simply politics or popularity, with recognized superstars easily attracting coverage. I can open any issue of Science or Nature and easily find at least three stories of import that have not been announced via press release or highlighted in blurbs in the front of the publication. A study reported in the July 15, 1998 issue of JAMA-The Journal of the American Medical Association, which is devoted to scientific publishing and peer review, reveals the impact of press releases in deciding the content of science news (V. de Semir et al., JAMA, 280[3]:294-5, July 15, 1998). Noting another study that found that 81 percent of articles in the British press that mentioned scientific journals were alerted to the information by press releases (V. Entwistle, British Medical Journal, 310:920-3, 1995), the researchers, from Pompeu Fabra University in Barcelona, followed science articles in seven major newspapers from December 1996 through February 1997, as well as tracking press releases and the table of contents from Nature, Science, Lancet, and the British Medical Journal for that time period. Of the 1,060 stories, 142 reported on information in one of the four journals. Of those 143 stories, 119, or 84 percent, covered articles linked to press releases. The researchers' conclusion: "Thus, journal articles that appeared in press releases were better represented in the general press." Given the power of the press release, a scientist doing interesting research should think about how that work might translate into a one- or two-page description, written in plain language, even before the PIO calls. This is the easiest way to increase the chances that research results are publicized accurately, understandably, and in context. Ricki Lewis is a contributing editor to The Scientist , and the author of several college life science textbooks for McGraw-Hill Publishers. In conclusion, I offer the following specifics: Get the credits right. Name the students and postdoctoral researchers who contributed significantly to the work. Cite granting agencies. Define jargon but don't oversimplify, and minimize use of acronyms. A sentence such as "The LOH studies using FISH on YAC and BAC clones revealed a small deletion between markers D9S1779 and D9S1858" won't mean much to anyone other than a geneticist. Describe how the experiments were done. Let the media choose whether or not to include this information. Place the work in context. Mention ideas and research that led to the reported work, and the possible implications of the results and future experiments. Don't be hesitant to speculate, but be certain to state that an idea is speculative or hypothetical. If necessary, remind the PIO of the difference between a theory and a hypothesis, a frequent source of confusion to the public. Place the work in the continuum that is scientific progress, countering the widespread media view of science as a series of "breakthroughs" geared solely to improving the human condition.

The Anatomy of a Press Release

Interested in reading more?

Become a Member of