Helping Journalists Get It Right

JGIM

PERSPECTIVES

Helping Journalists Get It Right
A P h y s i c i a n ' s G u i d e t o I m p r o v i n g H e a l t h C a r e R e p o r ti n g
Karen Stamm, BA, John W. Williams, Jr., MD, MHSc, Polly Hitchcock Noel, PhD, Rita Rubin, BSJ
È

News reports are the way that most people, including many Physicians felt that coverage is too sensational; journalists
physicians and scientists, first learn about new developments said it is not. Journalists believed that they could be
in medicine. Because these reports can raise awareness, objective; physicians disagreed.
influence behavior, and confer credibility, physicians should
Instead of viewing journalists as antagonists, physi-
share responsibility with the media for accurate reporting.
cians are better served by viewing health care reporting as a
Physicians can work with reporters to avoid sensationalizing
collaborative process in which both parties bear respon-
tentative findings, overstating benefits, and making inap-
sibility for producing fair and accurate news stories.
propriate generalizations. This article includes pragmatic
suggestions for crafting effective news releases and explaining Physicians benefit when consumers of their research are
numerical data. It details ``rules of the road'' for interviews. well informed, whether they are people who receive health
Working collaboratively with news reporters to improve the care, such as patients and taxpayers, people who provide
quality of medical stories in the lay press benefits patients and health care, such as physicians and other medical profes-
physicians alike. sionals, or people who fund health care research, such as
KEY WORDS: journalism, medical; mass media; physicians; those in government, industry and philanthropy. The goal
communication; health education. of this article is to outline strategies that we think will
J GEN INTERN MED 2003;18:138 ± 145. improve the quality of medical stories in the lay press and
result in a better-informed public. The recommendations
are pragmatically focused and emphasize interactions with

I n surveying the literature on the quality of medical the print media. They are based on a review of pertinent
journalism, we discover no shortage of challenges that articles obtained in a search of the electronic databases
need remedy. News stories about medical research fail to MEDLINE, HealthStar, and PsycINFO that were identified
mention potential harm to patients, fail to report benefits using the following search terms: communication, com-
quantitatively, and fail to mention costs.1 They frequently munication media, health education, mass media, and
fail to put new research into a meaningful context and fail numeracy. The authors also based recommendations upon
to mention limitations of the research.2 Articles commit their personal experience in roles as a former television and
sins of omission and sensationalism.3,4 Two people bear radio journalist and a public relations officer; an internist
the ultimate responsibility for the quality of a medical news and clinician-researcher who has interacted with local and
story: the clinician or scientist explaining the research, and national media outlets; a clinical psychologist with exper-
the journalist who interprets it for the public. A nationwide tise in health psychology and mass media dissemination;
survey of 857 respondents explored the relationship and as a medical reporter for USA Today and former senior
between these 2 groups.5 It revealed that while reporters editor of U.S. News and World Report.
felt certain that they get the technical details of medical
reporting correct, physicians were certain that they do not. CONSTRUCTING YOUR STORY: THE NEWS RELEASE
One strategy for getting your message to the public is
to send a well-written news release to health care reporters
Received from the University of Texas Health Science Center at to spark their interest. You might assume that if your
San Antonio (KS, PHN), San Antonio, Tex; The Center for Health research is being published, the medical journal will issue
Services Research in Primary Care, HSR&D, Department of a news release. Ask the journal that publishes your article
Veterans Affairs Medical Center (JWW), and Duke University
if this is the case. Only a handful of journals prepare their
Medical Center (JWW), Durham, NC; VERDICT, an HSR&D
own press releases on newly published research. These
Center of Excellence, South Texas Veterans Health Care System
include Annals of Internal Medicine, British Medical
(PHN), San Antonio, Tex; and USA TODAY (RR).
Address correspondence and requests for reprints to Ms. Journal, Circulation, Journal of the American Medical
Stamm: Faculty Associate, Department of Medicine/Division of Association, Journal of the National Cancer Institute,
General Medicine, The University of Texas Health Science Center Lancet, and Pediatrics.6 The vast majority of journals do
at San Antonio, 7400 Merton Minter Blvd. (11C6), San Antonio, not have the staff or the inclination to prepare news
TX 78229 ± 4404 (e-mail: kstamm@verdict.uthscsa.edu). releases. The New England Journal of Medicine only in
138

JGIM Volume 18, February 2003 139

2002 hired an outside public relations firm to work with the release is the first step in translating your study from the
press, although it still does not issue press releases. language of science to a more accessible description of your
Journal articles that are promoted with a news release work. Some journalists will base their report entirely on
appear to have a much greater chance of appearing in the your news release. Anecdotal evidence suggests that
popular media. One retrospective content analysis of spending time working with your institution's public affairs
newspaper stories and journal press releases found that office to craft an accurate, pithy news release is worth the
when newspaper stories referred to journal articles, 84% of effort. Dr. Meir Stampfer of the Harvard School of Public
these referred to articles that had been promoted in press Health found that ``the story was usually reported better if
releases while only 16% referred to articles not mentioned the journalist had a press release that we had a hand in
in press releases.7 The greater visibility that a physician preparing.''5
may enjoy by being featured in a news release requires Because medical reporters receive so many news
greater responsibility to ensure that research findings are releases, public relations professionals recommend lim-
reported accurately, completely, and in proper context. A iting a news release to 2 pages. If the release is spare and
recent examination of the news release process at several meaty, and still fills 3 or 4 pages, break it into a 2-page
high-profile medical journals found that the news releases release with a 1-page sidebar that addresses one important
frequently presented data in exaggerated formats, and issue. Or add a page of frequently asked questions to
failed to highlight study limitations or conflicts of interest.6 deliver the information.
Tell your journal editor that you'd like to review the draft Before you call the public relations folks, however,
news release. consider the questions they will ask you. It will make your
If the journal does not plan to issue a news release, initial appointment more productive. You probably will be
contact the public relations office at your university, asked some or all of the following questions: How will these
hospital, or any institution that has a stake in your findings help patients (the media's readers/viewers/
research. Even if reporters had easy access to all of the listeners) live longer or better? Will it reduce treatment
thousands of English-language medical journals (at most, time or save money? Is this truly new information or does it
news organizations subscribe to only a dozen or so medical confirm what other researchers have found? Do your
and scientific journals), they certainly do not have the time findings suggest a need to change standard treatment
to scour them for newsworthy studies. Journalists welcome now, or could they revolutionize treatment in the future?
press releases about brand-new, or, even better, about- Physicians have a responsibility to help the public
to-be-reported research, especially if the release comes apply research findings to their everyday lives, and to tell
from an academic center. people when research does or does not suggest a change in
That does not mean that every publication or pres- their lives now. They also have a responsibility to provide
entation at a scientific meeting merits a press release. fundamental information about the conduct of their
Medical reporters are bombarded with e-mails, faxes, and research so that the public, journalists, health care
snail mail from people trying to pitch them a story. professionals, and other potential stakeholders can evalu-
Pharmaceutical companies have internal and, often, ate its quality. In crafting a news release with a public
external public relations machines aggressively trying to relations consultant, consider the basic questions that a
promote new products. A number of physicians, hoping to complete medical research news story should answer
establish themselves as media-friendly experts in a par- (Table 1), adapted from The New York Times syndicated
ticular subject, have even hired their own public relations columnist Jane Brody.8 As you prepare to answer these
agents. The New York Times syndicated health columnist questions, think carefully about the language you'll use
Jane Brody says, ``I spend an entire workday each week and how you can best translate complex scientific and
going through my mail and journals. Typically on that day, medical concepts into lay language. It's likely that your
I fill three large wastebaskets with mailed and faxed research is newsworthy if your answers satisfy the public
material not worth keeping. Finding something worth relations practitioner, who generally has a good sense of
keeping has become a task akin to looking for a needle in what will interest reporters and editors.
a haystack. And finding something worth writing about Ð While most news releases are in print form, some
worth communication to a public that tends to swallow universities and journals produce video news releases
every bit of health news hook, line, and sinker Ð has (VNRs) that are transmitted to television stations nation-
become an even greater challenge.''8 wide via satellite. You could be videotaped and asked to
So what is most likely to attract journalists' attention? provide a patient who could also be interviewed on camera.
In general, reporters prefer to write about research involving (Print reporters might also want to interview and pho-
people, rather than petri dishes or animals. Usually, the tograph a patient.)
larger the study, the better. Research that challenges the Distribution of the news release is a strategic issue,
conventional wisdom is more appealing than the ninth or best discussed with a public relations professional. Most
nineteenth study that confirms earlier findings. university public affairs offices have developed media
A professionally written news release has a better mailing lists, and can advise you about whether national,
chance of being read by a reporter or editor. The news regional, or local distribution is most appropriate to your

140 Stamm et al., Improving Health Care Reporting JGIM

Table 1. Questions to Address in a News Release though the information may be preliminary and may have
undergone limited scientific review. They also found that
1. What question was the study designed to answer? 3 years after the meeting, one fourth of meeting abstracts
2. Who funded the research? Have all potential conflicts of remain unpublished in medical journals.9
interest been disclosed?
3. Were the findings published in a peer-reviewed author-
itative journal? EXPLAIN YOUR DATA CLEARLY AND ACCURATELY
4. How many subjects participated in the study? How were
Your numerical results will be presented through the
they selected and (if applicable) how were they assigned to
the various study groups? media to an audience that often is ill prepared to
5. Were the subjects humans or animals? If they were understand numbers. Problems with numeracy are com-
animals, are there limits on the applicability of the findings mon, and low numeracy is associated with poor under-
to people? If the subjects were humans, what was their standing of risk information and quantitative data.10,11 For
composition in terms of gender, age, ethnicity, and other
example, low levels of numeracy are strongly related to
important sociodemographic characteristics? Were the
human subjects patients or from the general population? difficulty in making use of quantitative data about the risk
Or were they physicians or other health care professionals? reduction from screening mammography.10 To illustrate
6. If outcomes data are presented, how was the information the magnitude of the problem, 47% of adult Americans
obtained (e.g., administrative data, chart abstractions, could not calculate the difference between the regular and
observational data, patient surveys)?
sales price from an advertisement.12 Poor numeracy may
7. Are there any potential threats to the validity of the findings
(e.g., large percentage of drop-outs in a clinical trial or not be limited to the general population. In a study of first-
nonresponders in survey research, misclassification in year medical students, only 77% answered all 3 relatively
outcomes research)? simple numeracy questions correctly, and only 61%
8. Which variables were controlled for? Are there any other accurately interpreted standard presentations about risk
variables that could have influenced the findings that were
reduction.11 Students were less likely to interpret risk
not controlled for?
9. If the study describes a clinical finding, are there basic reduction data accurately when the data were presented
science studies that support it? (The intent is to have a solely as number needed to treat as opposed to relative
plausible mechanism for the observed effect. For example, risk, absolute risk, or the combination of all three.
antibiotics help heal peptic ulcers. Basic science has Given the problems with numeracy, it is important for
shown that the bacteria Helicobactor pylori are found in
researchers to communicate their results clearly, perhaps
more than 90% of ulcers. So it's logical that treating the
H. pylori may heal ulcers Ð and it does!) by using multiple methods. One possible solution is to
10. What has previous research shown and how do the new substitute verbal qualifiers (e.g., ``likely'' or ``rarely'') for
findings advance our knowledge? numbers, but risk communication research has shown
11. What are the practical applications of the findings? Are that the numerical equivalents for these terms vary greatly
there any important caveats?
by context (e.g., ``likely'' to rain versus ``likely'' to have a
Adapted from health columnist Jane Brody.8 painful adverse effect) and by individual.13,14 On the basis
of a case-study of reporting errors,15 research on cancer
risk communication,16 and our own experience, we offer in
message. The most ambitious dissemination strategy is
Table 2 guidelines to presenting numerical results. For
comprehensive and national, which requires current
readers wanting a more comprehensive guide to reporting
addresses of the medical reporters at major newspapers,
statistics, we recommend the text How to Report Statistics
relevant magazines, major radio and TV stations, syndi-
in Medicine.17
cated wire services (such as Associated Press), and Web-
based publications.
WHAT TO DO WHEN A REPORTER CALLS
Increasingly, journalists prefer to receive press
releases via e-mail. Fax machines break down, and the It's important to be accessible to journalists, who,
U.S. Postal Service is not always timely. With e-mailed facing deadline pressures, will simply call another
press releases, it is easy to attach charts or photos that researcher for comment if you are not available. There
could heighten reporters' interest in your story. are commonly 2 instances in which you can plan on the
Timing of the news release depends on whether you're need to be accessible: after a news release has been sent
disseminating data that is also being published in a or a news conference has been held. In these cases,
medical journal or presented at a medical meeting. A reserve some time over several days to respond to
number of journals and scientific organizations forbid reporters' requests. It can damage your credibility and
press coverage of new research prior to publication or your good relationships with reporters to make an
presentation (see section on embargoes). A new study announcement and then fail to respond to the resulting
about press coverage of preliminary data at scientific media interest.
meetings underscores the need for peer review before If you answer the phone and unexpectedly find a
releasing research results to the public.9 Authors found reporter on the other end of the line, feel free to tell the
that information from abstracts presented at scientific reporter you would prefer to call back in a few minutes,
meetings are widely reported in major media outlets, even after you've had a chance to collect your thoughts.


Table 2. Helping Reporters Understand Numerical Data telephone interview, which tends to produce more conver-
sational quotes than written correspondence via e-mail.
Report the absolute event rates. 0.6 percent vs 1.0 percent and
60 percent vs 100 percent represent very different absolute WHEN REPORTERS ASK YOU TO PROVIDE NAMES OF
event rates, yet yield the same 60 percent relative risk.
PATIENTS TO ENHANCE YOUR STORY
Report benefit and harm results symmetrically. When reporting
the results of an intervention, present absolute event rates A well-described patient example can often convey a
for both benefits and harms. message where numbers fail. Journalists understand this,
Put the results in context. Help the reader calibrate the so no matter how well you explain your numbers, they'll
magnitude of the risk or intervention effect by placing the often ask if they can interview one of your patients. If you
results in context. Give the risk of events generally accept that a patient's story would enhance public
acknowledged to be common (e.g., common cold) or rare (e.g.,
understanding of your message, and you can identify a
hit by lightening). Compare your intervention effects to other
accepted treatments using outcomes common to both (e.g., patient, consider the ethics of asking a patient to consent
mortality). to an interview. The American Psychological Association
Convert odds ratios to risk ratios if the event rate exceeds 10%. has studied the issue of whether patient confidentiality
When the outcome of interest is uncommon (e.g., <10%), the
odds ratio is appropriate because it approximates the risk
ratio; odds ratios increasingly overstate the risk ratio as
outcomes become more common. Odds ratios are hard to Table 3. What to Do When a Reporter Calls
comprehend directly and are usually interpreted as being
equivalent to the relative risk. Many people have no intuitive When a reporter calls, have your media liaison ask for the
feel for odds or odds ratios and it is unrealistic to expect following information. Answers to these questions will help
reporters or the public to understand the distinction. There you decide whether you want to do the interview.
are simple methods of conversion for both crude and . What is your deadline?
adjusted data.18,19 If odds ratios cannot be avoided one . How much time do you need with the doctor?
should remind the reader that the higher the base rate the . Tell me what this interview is for Ð which publication,
more the odds ratio will overstate the relative risk. broadcast show, online news site, etc. Tell me how many
potential readers, viewers it has. Who is the audience Ð
Report the precision of the result. The confidence interval is a
physicians or lay public?
range of values consistent with the data, that is believed to
. Can you give me a general idea of what topics the interview
encompass the ``true'' population value.17 Help the reporter
will cover?
understand that the ``true'' effect in the broader population to
. Who else is being interviewed for this story? Reporters may
which your study applies lies within this confidence interval.
volunteer this information, or may choose not to divulge
Ensure that comparisons are appropriate. In reporting the their other sources.
results of comparisons, one should be clear about the
Tell your media liaison to provide you with the information
reference group. If an important interaction is identified,
above quickly, so you can provide times that you're available
aggregate data should not be used in reporting the results.
for an interview. Returning reporters' calls late is the best
guarantee that you won't be interviewed.
Determine who will call the reporter back to confirm a time for
the interview. Be clear on who's calling whom if it's a
Occasionally, journalists will call for your opinion of a telephone interview, and confirm the phone number. Give the
study you might not have yet seen. Ask the reporter to fax reporter a name and number to call in case they have to
you the study so you can comment intelligently. Another reschedule the interview.
way to conserve your time and avoid being caught off guard Ask your media liaison to make a practice of contacting your
is to train someone on your staff to assist. If you are the institution's public affairs office once an interview is sched-
uled. Time is of the essence, as reporters usually need an
author of a study expected to attract a considerable interview in the next day or two, and institutional clearance
amount of press attention, consider delegating the respon- can take that long. It's in your best interest to notify Public
sibility for screening reporters' calls and scheduling inter- Affairs as soon as possible because they can tip you to
views to a member of your institution's public affairs office. potentially sensitive topics for the institution that could come
An effective media liaison is a person who is usually up during an interview. Give the public affairs specialist all
the information you have gathered, including the reporter's
available to handle phone calls and who can take precise name and phone number. If the interview will take place on
telephone messages. If other support personnel handle campus, the Public Affairs office may arrange for the
your incoming calls, let them know where to direct your reporter's security clearance, convenient parking, and an
media calls. See Table 3 for guidelines you can give staff escort to your office. Contact your institution's public
members. relations office to learn its protocol. NOTE: This procedure
changes slightly if the request is from a local or national TV
Increasingly, reporters e-mail interview requests to reporter. Their requests are generally more complicated; they
physicians and are even willing to conduct interviews via usually want us to identify a patient to be interviewed or
e-mail. This approach benefits busy physicians who can arrange a procedure to be videotaped. When you get a TV
answer at their convenience and feel less at risk of request, you may want to discuss the story with the reporter
being misquoted. However, many journalists still prefer yourself, and ask the institution's public relations office to
get involved.
the immediate back-and-forth exchange possible with a


requires that health care professionals rule out giving vidual reader (e.g., `Ìf you are a woman 45 years or older
reporters the names of patients, either current or past. with x, you may be at risk for y. Ask your doctor about.'').
Although the types of stories psychologists tell the media Avoid medical jargon, acronyms, and statistics. Talk
may differ from those of physicians, their suggestions offer to the reporter the way you would talk to a neighbor who
some valid points for consideration20: The patient's knows little about your field or to an intelligent 8th
welfare always comes first. Is the patient deciding to grader.
participate to please the clinician? Is the value of the Create ``word pictures'' to explain technical concepts.
interview for public education purposes? Normal informed For example, ``The stent works a bit like a Chinese finger
consent procedures apply here. Tell the patient the puzzle. Just as the puzzle tightens its grip on your fingers
purpose of the interview, where it will take place, how as you twist it, the stent pushes more tightly against the
long it will last, and where the information will be arterial wall as blood pushes through it.''
presented. Reassure the patient that agreeing or declining Anticipate questions and prepare short answers rather
to participate will not have any impact on their sub- than long ones. Use full sentences. Short answers are less
sequent health care. The clinician or a staff member likely to be edited or misunderstood.
should contact the patient first and describe the process Prepare for the possibility that the interviewer's ques-
to see if the patient is willing to participate. tions won't give you a chance to convey your message.
Think about ways to ``bridge'' to your message with a
phrase like, `Ì'd also like to point out. . .,'' or `Ànother point I
NOW THAT YOU KNOW WHAT TO SAY, might mention here. . ..''
HOW DO YOU SAY IT? Prepare a list of other experts who could comment
knowledgeably on your study findings. Reporters may
Dr. Sam Nixon of Houston, former president of the
interview other experts to put your findings into context.
American Academy of Family Physicians and the Texas
This is your opportunity to direct them to research leaders
Medical Association, says he gives 2 different types of
who are likely to offer balanced views, or to clinical experts
answers when he's being interviewed on tape.5 If he wants
who are particularly knowledgeable about the implications
to make sure he is quoted, he gives pithy answers. If he
for clinical practice.
wants to ensure he isn't quoted, he'll give rambling
Expect that tight deadlines will not permit a chance to
answers.
review the news article. News organizations rarely will
Approach the reporter as you would a student, advises
allow interviewees to review the entire text before publica-
R. K. Farberman in the book, Psychology and the Media.
tion or broadcast, although journalists are usually willing
``Think of the interview as a teaching opportunity, and as
to read back or e-mail quotes so interview subjects can
such, communicate to the student (the reporter) in language
check them for accuracy. Unless you have an established
he or she can understand. . .. By taking the time to educate
working relationship with a reporter, it's best not to speak
the reporter and provide him or her with all appropriate
off the record. Anything you say could end up in print or
background material, you are making an investment in
on the air.
more thorough, more accurate news coverage.''5
Table 4 contains a list of general and specific tips to
help you when giving interviews. Remember that inter-
Preparing for the Interview
viewing is a learned skill. Physicians who expect to have
Expect your interviewer to have had little or no regular or ongoing contact with the press should seek
science education in college. A Freedom Forum survey training in order to better understand reporters' needs and
showed that most medical reporters majored in journal- constraints.
ism and liberal arts; only 17% had advanced degrees in
science.5
EMBARGOES: WHEN TO REMAIN SILENT
Decide on 1±3 messages that you believe are most
important to explain your research. If you're discussing a An embargo is an agreement that you make with a
multicenter study or there will be multiple spokespersons, publisher when your manuscript is accepted for publica-
agree in advance on the messages and make sure you are tion in a peer-reviewed journal. It states that you agree
all reading from the same script. `Òff message'' or conflict- not to disclose data in your article until its publication
ing messages may end up being ``the story'' rather than date. Embargoes don't apply to researchers presenting
your health care finding. Ensure that your key messages papers at scientific meetings, Congressional hearings, or
make it into the news story by putting them in quotable other government proceedings. If reporters cover these
(i.e., short, everyday) language. Write them down so you presentations and approach you afterward with questions,
can see them in black and white. Make sure you're giving it's okay to discuss your findings, but take care to limit
the message you intend. comments to the information addressed in your presenta-
Answer the audience's unspoken question, ``What's in it tion. Refrain from making illustrations available to the
for me?'' Specify the group(s) to whom your message interviewer that you may wish to publish in a journal at a
applies and provide concrete suggestions for the indi- later time.


Table 4. The Interview: Rules of the Road Table 4. (continued)

General tips Television cameramen will shoot additional footage following
an interview. Be sure your smiles and gestures are compati-
Never say ``no comment.'' It's like waving a red flag in front of ble with your posture during the interview; you don't know
a bull. Reporters will look elsewhere for information you can't how the story will be edited.
or don't provide. Even a general statement is better than none
at all. If you can't give an answer, explain why. Plan the background carefully. Remember, a picture is worth
a thousand words.
Don't answer hypothetical questions. Questions that begin
with ``what if'' or ``let's suppose'' might land you in hot water.
Instead, respond positively about the way you've handled
actual situations, or use a phrase such as, ``I can't speculate
about that, but we do know for a fact that . . . .'' An embargo also refers to an agreement that journals
have with journalists. Journals release advance copy about
The interview isn't over until it's over. Even after the reporter
has snapped shut the notebook, he's still listening to a manuscript to the news media with the understanding
everything you're saying. Anything you say might be used in the media will defer coverage until final publication. In
the story. those cases, the authors are encouraged to talk with
Telephone interviews reporters, although it's advisable to first secure a reporter's
promise to honor the embargo.
Your voice transmits how you're feeling. Demonstrate the
same animation and excitement in a telephone interview that Similar to the embargo is an agreement between
you would in person. Take the call standing up to avoid authors and journals known as the Ingelfinger Rule.21
relaxing too much and letting your guard down. Under this rule, the journal considers a paper for publica-
Ask if the interview is being recorded. Print media reporters tion only on the condition that its substance is not
will tape interviews to get the facts straight and avoid submitted or reported elsewhere. The Ingelfinger Rule
misquoting you. If the interview is for radio, it means that any (named for the New England Journal of Medicine editor
answer or partial answer can be edited for use on the air.
who founded it) prohibits publicity about an unpublished
Incorporate the question into your answer if the interview is report from the time of submission, while the embargo
being taped for later editing and broadcast. The interviewer's
applies only to the time between the printing of the journal
questions that put your answers in context may not be aired
or printed. and its date of publication.
In tandem, the 2 policies are designed to ensure that
Make your point first when responding to a question. Then
back it up with facts. Most importantly, don't start answering journal readers have the full article when the media report
the question before you think through what you want to say. on it. It's also believed that organized dissemination of
Ask before the interview if it will air in its entirety or will be important, technical information through the public press
edited. If the interview won't be edited, proceed as if it were leads to better quality news coverage. The embargo period
``live.'' Everything you say will be on tape. allows reporters time to analyze and report on the complex
Television interviews stories behind the data. Reporters who violate a journal's
embargo (proved by conclusive evidence) can be excluded
Look open, relaxed and responsive. Communications studies
show that in television, the impression you create outweighs from future embargoed information.
the message you state. These policies have been criticized by some who argue
Look at the reporter, not the camera. Maintain good eye that the public has a right to know before scientific
contact. Looking away can make you look uncertain and publication about trial data that appear to have significant
nervous. It also distracts the viewer from what you're saying. implications for patient care or safety, especially when
Speak at your normal voice level in a conversational tone. funded by the government.21±23 Under such circumstances,
Modern recording equipment will get your message across. the prepublication embargo has been waived in a few
Avoid sitting in a chair that swivels. If you can't, plant your instances. A case study of 1 such instance suggests that
feet on the floor in a way that will prevent you from inadver- the premature release of research data may have unintend-
tently turning the chair. ed and adverse effects on clinical practice.24 Results of 2
Sitting fairly erect, but leaning slightly forward will help you clinical trials on the use of carotid endarterectomy (CEA) to
look attentive and eager to talk. Pressing the small of your prevent stroke that were released prepublication were
back firmly against the chair centers your energy. Sitting on
associated with rapid increases in the use of CEA. Unfortu-
your suit coat or lab coat will prevent wrinkling at the
shoulders. Men should make sure socks cover their calves nately, the increased use of CEA was not always applied
when legs are crossed. Take a minute beforehand to appropriately, resulting in more harm than benefit to some
straighten your hair and clothing. patients.24 It is the responsibility of authors, sponsoring
Use audience language. Reporters probably will not air any institutions, and reporters to correct misinformation.4
response that contains jargon or technical words that need
further explanation.
After the interview, feel free to speak out if you don't feel you
CONCLUSION
did your best. The reporter may be willing to let you try again If successful, this article demonstrates that physicians
on the important questions.
can help journalists write accurate medical stories with an


investment of time, creative energy, and skill. To decide also value medical research librarian Molly Harris, AHIP,
whether the payoff is worth the investment, consider the MLS, MA, for her expertise in searching and reference
following. News reports are the first way that most people, management.
including physicians and scientists, learn about new This material is the result of work supported in part by a
grant from the John A. Hartford Foundation of New York and
developments in medicine.25 A Washington Post survey
with resources from the South Texas Veterans Health Care
found that television is the primary source of health
System and the Department of Veterans Affairs Medical Center
information for 39% of those surveyed; newspapers, 28%; in Durham, NC. The views expressed in this article are those of
magazines, 23%; and radio, 10%.22 News reports not only the authors, and they do not necessarily represent the views of
inform a physician's patients and colleagues, they also the Department of Veterans Affairs.
confer credibility. Furthermore, researchers are more likely
to cite papers that have been publicized in the popular
press. One study found that articles that appeared in the REFERENCES
New England Journal of Medicine and that were covered by
1. Moynihan R, Bero L, Ross-Degnan D, et al. Coverage by the news
the New York Times received 73% more scientific citations
media of the benefits and risks of medications. N Engl J Med. 2000;
in the Science Citation Index than control articles.25 342:1645±50.
Although it's not clear that newspaper coverage of a 2. Goldberg JP, Hellwig JP. Nutrition research in the media: the
single research study can influence health policy, news challenge facing scientists. J Am Coll Nutr. 1997;16:544±50.
reporting influences public opinion, which in turn, helps set 3. Molitor F. Accuracy in science news reporting by newspapers: the
case of aspirin for the prevention of heart attacks. Health
public policy agendas.26±28 Consider media coverage of the
Commun. 1993;5:209±24.
health hazards of smoking. Studies suggest that the mass 4. Shuchman M, Wilkes MS. Medical scientists and health news
media may have slowed public acceptance of smoking's reporting: a case of miscommunication. Ann Intern Med. 1997;
health risks and subsequently delayed regulatory action, in 126:976±82.
spite of accumulated evidence indicating that cigarette 5. Rubin R, Rogers HL Jr. Under the microscope: the relationship
between physicians and the news media Nashville, Tenn: The
smoking and passive smoking are associated with increased
Freedom Forum First Amendment Center at Vanderbilt University;
mortality.29,30 Media influences that may have served to 1993.
obscure smoking's heath risks included tobacco industry± 6. Woloshin S, Schwartz LM. Press releases: translating research into
funded public relations campaigns, tobacco advertising news. JAMA. 2002;287:2856±8.
dollars, and journalists' tradition of balanced reporting.29,30 7. de Semir V, Ribas C, Revuelta G. Press releases of science journal
articles and subsequent newspaper stories on the same topic.
Providing balance meant that reporters routinely quoted
JAMA. 1998;280:294±5.
tobacco industry spokespersons, suggesting that these 8. Brody JE. The challenges of accurate health reporting [speech].
opposing voices carried equal scientific weight with pub- University of Minnesota School of Public Health. Dec. 1, 2000. Avail-
lished peer reviewed research.29 It is the role of news able from: URL: http://www.ahcj.umn.edu/files/brodyspeech.
organizations to provide information to the public, give voice pdf.
9. Schwartz LM, Woloshin S, Baczek L. Media coverage of scientific
to different viewpoints, and facilitate discussion, but to
meetings: too much, too soon? JAMA. 2002;287:2859±63.
avoid advocating solutions to maintain journalistic objec- 10. Schwartz LM, Woloshin S, Black WC, Welch HG. The role of
tivity.31 When appropriate, medical researchers should numeracy in understanding the benefit of screening mammogra-
explicitly clarify areas where significant scientific agree- phy. Ann Intern Med. 1997;127:966±72.
ment exists and comment on implications their findings 11. Sheridan SL, Pignone M. Numeracy and the medical student's
ability to interpret data. Eff Clin Pract. 2002;5:35±40.
may have for individual behavior or health policy.32
12. Kirsch IS, Jungeblut A, Jenkins L, Kolstad A. Adult Literacy in
Doctors' patients ultimately suffer the most when news America. Washington, D.C.: Office of Education Research and
reports sensationalize information. A case in point: news Improvement, U.S. Department of Education; 1993.
wire services reported in 1995 that 6 million people in the 13. Budescu DF, Wallsten TS. Processing linguistic probabilities:
United States who were taking calcium channel blockers general principles and empirical evidence. In: Busemeyer JR,
Hastie R, Medlin DL, eds. Decision Making from a Cognitive
for hypertension might be increasing their risk for a heart
Perspective. New York: Academic Press; 1995:275±318.
attack by 60%.4 Some people stopped taking their medica- 14. Merz JF, Druzdzel MJ, Mazur DJ. Verbal expressions of prob-
tion and doctors' offices were flooded with calls. The report ability in informed consent litigation. Med Decis Making. 1991;11:
stemmed from a presentation at the American Heart 273±81.
Association that was covered by journalists who had been 15. Schwartz LM, Woloshin S, Welch HG. Misunderstandings about the
effects of race and sex on physicians' referrals for cardiac
given a news release but who had little access to scientists
catheterization. N Engl J Med. 1999;341:279±83.
other than the presenter. Situations such as this are 16. Schwartz LM, Woloshin S, Welch HG. Risk communication in
perhaps the most compelling reason for internists to help clinical practice: putting cancer in context. J Natl Cancer Inst
journalists get it right. Monogr. 1999;25:124±33.
17. Lang TA, Secic M. How to report statistics in medicine: annotated
guidelines for authors, editors, and reviewers. Philadelphia:
American College of Physicians;1997.
18. Zhang J, Yu KF. What's the relative risk? A method of correcting the
We appreciate the thoughtful critique given this manuscript odds ratio in cohort studies of common outcomes. JAMA. 1998;
by our internal reviewer, Eric M. Mortensen, MD, MSc. We 280:1690±1.


19. Davies HT, Crombie IK, Tavakoli M. When can odds ratios mislead? 26. Broder DS. Behind the Front Page: A Candid Look at How News Is
BMJ. 1998;316:989±91. Made. 1st Touchstone ed. New York: Simon and Schuster; 1988.
20. Farberman RK. What the media need from news sources. In: Schwartz 27. Jordan DL. Newspaper effects on policy preferences. Public Opin Q.
LL, ed. Psychology and the Media: A Second Look, 1st ed. Washing- 1993;57:191±204.
ton, DC: American Psychological Association; 1999: 9±23. 28. Otten AL. The influence of the mass media on health policy. Health
21. Altman LK. The Ingelfinger rule, embargoes, and journal peer Aff (Millwood). 1992;11:111±8.
review±Part 1. Lancet. 1996;347:1382±6. 29. Kennedy GE, Bero LA. Print media coverage of research on passive
22. Altman LK. The Ingelfinger rule, embargoes, and journal peer smoking. Tob Control. 1999;8:254±60.
review±Part 2. Lancet. 1996;347:1459±63. 30. Warner KE. Cigarette advertising and media coverage of smoking
23. Kassirer JP, Angell M. Prepublication release of Journal articles. and health. N Engl J Med. 1985;312:384±8.
N Engl J Med. 1997;337:1762±3. 31. Logan RA. Three opportunities to work with the news media.
24. Gross CP, Steiner CA, Bass EB, Powe NR. Relation between Arthritis Care Res. 1998;11:53±9.
prepublication release of clinical trial results and the practice of 32. Wellman NS, Scarbrough FE, Ziegler RG, Lyle B. Do we facilitate
carotid endarterectomy. JAMA. 2000;284:2886±93. the scientific process and the development of dietary guidance
25. Phillips DP, Kanter EJ, Bednarczyk B, Tastad PL. Importance of the when findings from single studies are publicized? An American
lay press in the transmission of medical knowledge to the scientific Society for Nutritional Sciences controversy session report. Am J
community. N Engl J Med. 1991;325:1180±3. Clin Nutr. 1999;70:802±5.

Helping Journalists Get It Right

Helping Journalists Get It Right

More Related Content

Viewers also liked (19)

Similar to Helping Journalists Get It Right (20)

Recently uploaded (20)

Helping Journalists Get It Right