Reporting Epidemiologic Research by the News Media

Quit Smoking Magic

Quit Smoking Magic

Get Instant Access

Most of the information that Americans receive about science and technology comes from television news programs and newspapers (Nelkin 1987). The news media has become increasingly interested in reporting the results of research as they are published in journals or presented at scientific conferences. Science and health writers scour the hundreds of studies published each week. Their selection of which published research will be played in a lead story shapes the public image of the critical health issues of the day. Many health professionals first learn about the results of an important study by reading it in the morning paper or hearing it on the six o'clock news.

The role of the media in reporting on scientific studies was recently assessed (Wilkes and Kravitz 1992). Two journals in the United States—the New England Journal of Medicine and the Journal of the American Medical Association—account for many of the national and local stories that present health information to the public. Wilkes and Kravitz (1992) reviewed reports published in these two journals during a 10-month period in 1989, and found 414 original scientific articles. The most common type of study published during this time dealt with epidemiology or health services research (42%), compared with basic science (21%), clinical trials (26%), and other topics (11%).

The first authors of these reports stated that their research was often discussed in the lay press; coverage included print and electronic media at the local and national level (Table 11-1). In this study, most first authors were contacted by the media and most authors made special efforts to make themselves available. Many were aided by press releases issued by their institutions or the journal, or by press conferences. Thus, the dissemination of this information is a result of efforts by both media and researchers.

The media's interest in reporting health information is a result of the public interest and demand for such information. Reporters are looking for new information, startling results, or studies that contradict the status quo. Preliminary findings or seemingly trivial results are often reported boldly as "important new information" or "a study with startling findings." At times it may seem more important to sell newspapers and advertising space than to present a concise and effective health message.

Authors may conclude their work with a cautionary statement and caveat such as, "These data are preliminary and should be interpreted with caution." However, once in the press, these caveats are useless (Taubes 1995). In response to this, professional journals have published guidelines on how to deal with reporters. These guidelines serve as defensive tactics, warning epidemiologists to be aware of the reporter's motives. The New England Journal of Medicine warns, "Never even whisper to a reporter anything you would not care to see in screaming headlines." Potential problems can be identified if authors do a dry run with public relations staff in their agency.

At the other end of the spectrum from overly aggressive reporters is the promotion of research findings by scientists themselves. A presentation of "major research findings" at a press conference may promote the researcher's—or institution's—stature in the community. Many institutions

Table 11-1. Types of Media Coverage of Reports Published in the New England Journal of Medicine and the Journal of the American Medical Association, 1989



Report received media attention



Newspaper coverage







Electronic media coverage







Trade journals



All reports



Source: Wilkes and Kravitz (1992).

Source: Wilkes and Kravitz (1992).

employ a media or public relations professional to increase coverage. These professionals serve as an important link between the media and scientists, helping to package complex material in a manageable form (Nelkin 1987).

Epidemiologic research takes an incremental approach in determining the causes of disease and effectiveness of preventive interventions. Many of the major risk factors for diseases were identified decades ago. Wynder recalls that in his early study of smoking and lung cancer, "the relative risks were so large that, in fact, our paper published in 1950 included no statistical testing. For similar reasons, other studies such as those on alcohol and cancer of the upper alimentary tract, sexual habits and cervical cancer, radiation and leukemia were also easy targets to explore by even the inexperienced epidemiologist" (Wynder 1996).

Occasionally, causal relationships have been established following a few studies. For example, aspirin was determined to cause Reye syndrome, an often fatal disease of young children, as a result of a few studies published in the early 1980s (Starko et al. 1980; Waldman et al. 1982; Halpin et al. 1982). The media attention given to these studies caused a dramatic decline in aspirin use and an associated decline in the number of Reye syndrome cases (Remington et al. 1986).

Despite these occasional important health discoveries, most of the understanding about disease causation accumulates over years, and even decades, often requiring dozens of different studies using a variety of epidemiologic methods. Much of today's epidemiologic research focuses on risk factors that

Table 11-2. Media Highlights of Controversies in Epidemiology

Headline or Implication

Causes pancreatic cancer (McMahon et al. 1981)

Risk Factor Coffee

Type A personality

Margarine Pesticides

Estrogen replacement therapy

Beta carotene

Oral contraceptives

Causes heart disease (Barefoot et al. 1983)

Cause breast cancer (Falck et al. 1992)

Does not cause breast cancer (Kaufman et al. 1984)

Prevents cancer

Do not cause breast cancer (Cancer and Steroid Hormone Study Group 1986)


Does not cause pancreatic cancer (Feinstein et al. 1981)

Does not cause heart disease (Shekelle et al. 1987)

Is bad for the heart (Willett and Asherio 1994)

Do not cause breast cancer (Krieger et al. 1994)

Causes breast cancer (Steinberg et al. 1991)

Causes cancer (Omenn et al. 1996)

Cause breast cancer (Miller DR et al. 1989).

Is good for the heart


Figure 11-1. Conflicting health messages received by the public (Reprinted with special permission of King Features Syndicate)


Figure 11-1. Conflicting health messages received by the public (Reprinted with special permission of King Features Syndicate)

are prevalent in society, such as alcohol consumption or exposure to environmental tobacco smoke or pesticides. Even with a small relative risk of disease among those exposed (e.g., a relative risk of 1.3), the widespread exposure to these factors creates a potentially large public health problem. Despite the potential importance of the issue, however, many of the studies that examine these risks are plagued by biases, errors in measurement, and méthodologie weaknesses. Thus, some have stated that we now approach the "limits of epidemiology" (Taubes 1995). When results from studies appear to conflict, the media highlights the controversy (Taubes 1995; Angell and Kassirer 1994). Examples of this abound in the literature (Table 11-2 and Figure 11-1).

In conclusion, communicating epidemiologic information through media coverage of published research often leaves the public confused, bewildered, and in a state of "information overload." As a result, the public may begin to doubt all the information, or select only the information that conforms with their desires and existing beliefs. More organized strategies are needed to improve the public's understanding about these issues.

Was this article helpful?

0 0
5 Easy Ways To Stop Smoking

5 Easy Ways To Stop Smoking

Your first day without cigarettes can be difficult, but having a plan will make it easier! Learn what steps to take on the day you quit smoking.

Get My Free Ebook

Post a comment