- Email: [email protected]
Media coverage of chronic diseases in the Netherlands
Media Coverage of Chronic Diseases in the Netherlands Elly M. van der Wardt, Erik Taal, Johannes J. Rasker, and Oene Wiegman
Objective: Little is known about the quantity or quality of information on rheumatic diseases provided by the mass media, The aim of this study was to gain insight into the media coverage of rheumatic diseases compared w i t h other chronic diseases in the Netherlands. Materials and Methods: Newspaper articles, magazine articles, and medical television programs that appeared or were broadcast during a 1-year period, and contained information on rheumatic diseases, heart disease, cancer, chronic lung disease, or diabetes mellitus, were selected for content analysis. For each article and program, it was determined whether coverage concentrated on treatment, influence of lifestyle, scientific progress, or disease consequences. It was also determined whether professional experts and patients were featured. Results: Nine hundred seventeen newspaper articles, 304 magazine articles, and 163 medical programs were found. Most dealt w i t h cancer (43%) and heart disease (37%). The amount of media attention given to each of the five disease categories was found to correspond w i t h mortality but not w i t h prevalence. The contents of the articles and programs differed significantly according to disease topic. The main focus in rheumatic diseases was on patients" experiences, as well as regular and alternative medications. In heart disease and cancer, the main focus was on professional medical viewpoints, operations, and mortality, whereas in chronic lung disease and diabetes it was on treatments in the context of regular medications, scientific progress, and incurability. The influence of lifestyle on the disease process was mentioned most often in connection w i t h diabetes, rheumatic disease, and chronic lung disease. Conclusions: The amount of attention a disease category received from the media depended on its fatality rates and not on its prevalence. Heart disease and cancer were portrayed as being more serious than the more lingering diseases. Surprisingly, the proportion of articles and programs that included the influence of lifestyle in their coverage was lowest for cancer and heart disease. More frequent and more accurate coverage of chronic diseases, especially rheumatic diseases, is needed if their image is to be brought into line w i t h their importance for and impact on the community. Semin Arthritis Rheum 28:333-341, Copyright © 1999 by W.B, Saunders Company INDEX WORDS: Media; chronic diseases; Netherlands.
From the Department of Communication Studies, University of Twente, and the Department of Rheumatology, Hospital of Twente [Medisch Spectrum Twente], Enschede, the Netherlands. Elly M. van der Wardt, PhD: Communication scholar and Linguist, Department of Communication Studies, University of Twente; Erik Taal, PhD: Social psyehologist, Department of Communication Studies, University of Twente; Johannes J. Rasker, MD, PhD: RheumatoIogist, Department of Rheumatology, Hospital of Twente [Medisch Spectrum Twente], and Professor, Department of Communication Studies, University of
Twente; Oene Wiegman, PhD: Professor in Psychology, Department of Communication Studies, University of Twente. Supported by grants from the Dutch League against Rheumatism [National Reumafonds] and the Min&try of Health, Welfare, and Sport [Ministerie van Volksgezondheid, Welzijn en Sport] of the Netherlands. Address reprint requests to Professor Johannes J. Raske~ MD, PhD, Department of Communication Studies, University of Twente, PO Box 217, 7500 AE Enschede, The Netherlands. Copyright © 1999 by W.B. Saunders Company 0049-0172/99/2805-00055!0.00/0
Seminars in Arthritis and Rheumatism, Voi 28, No 5 (April), 1999: pp 333-341
333
334
TTLE IS KNOWN about the quantity or quality of information on rheumatic diseases provided by the mass media. The aim of this study was to gain insight into the media coverage of rheumatic diseases compared with other chronic diseases in the Netherlands. Public interest in health issues has increased enormously over the last three decades; health columns in newspapers and magazines are flourishing, health claims are prominent in advertising, and more and more health-related programs are broadcast by an increasing number of television channels (1,2). About 30 years ago, a study showed that the printed media were the American public's main source of health information (3). A later study found that newspapers and magazines had been replaced by television as the most important source of health information (4). It is evident, then, that both the printed and broadcast media should be considered when investigating media coverage of chronic diseases. Media coverage of chronic diseases may have a great impact on the general public's perception of the causes, treatments, and consequences of these conditions. An important function of the mass media is setting the public agenda and stimulating public discussion (5,6). During the course of their lives, people have direct contact with only a small sector of the environment, so their impressions of many social realities are greatly influenced by the information they receive from the mass media (7). The images presented by the media, particularly television, may influence beliefs concerning health and health behavior. Television contributes to viewers' conceptions of social roles and human relations (8). Frequent portrayal of violence in connection with mental illness, for example, persuades audiences that the two are associated (9). Frequent television viewing can lead to a high degree of confidence in the medical profession's ability to cure disease, which may justify unhealthy lifestyles and lack of interest in disease prevention (4). The mass media have been held responsible both for the public's misconceptions about arthritis (10) and for its increased awareness of asthma (11). Newspaper coverage of homicides, accidents, and disasters influences the public's ideas about the main causes of death (12). The mass media can influence the opinions of the public, but public opinion also shapes the style and presentation of media news. Audience members have well-formed preferences and opinions that
E
VAN DER WARDT ET AL
allow them to selectively choose what to read, watch, or hear in the media. To attract and maintain audiences, journalists and program-makers must bear in mind the public's preferences (13). There have been few studies on the coverage of chronic diseases and health by the mass media (12,14-22). These studies were subject to methodological restrictions on data collection. Most studies concentrated on one medium only or on one category of presentation within one medium: television news reports (14,15), medical programs on television (16,17), newspaper articles (12,18-20), or magazine articles (21). In some, the period studied was 4 months or less (16-19). Nearly all of the studies relied on the indexes of news archives for their data collection (14,15,19,21,22); therefore, results were dependent on the key words provided by the indexers. Media coverage of medical topics often is biased. Several American studies have focused on news coverage of cancer. Newspapers provided scarcely any information on incidence rates, distinct types of cancer, prevention, detection, or treatment (19), and few focused on scientific background information (20). The media overstated the risks from pollution, food additives, and nuclear power, and understated those inherent in the use of tobacco and exposure to sunlight (22). The popular press paid little or no attention to skin cancer morbidity or mortality (21). Two studies on medical programming by the British Broadcasting Corporation showed that most television programs were hospital-based, highlighted aspects of medical technology, and that hospital specialists were the most prominent communicators. Few programs dealt with health maintenance, preventive medicine, risk factor avoidance, or common diseases (16,17). It seems important to know how and when a particular disease becomes newsworthy. Does media coverage of chronic disorders, such as rheumatic disease, chronic lung disease, or diabetes, differ in meaningful ways from coverage of conditions that are potentially fatal, such as heart disease and cancer? Although heart disease and cancer can be fatal, they are increasingly becoming chronic, because treatments for these diseases have become more effective. To answer these questions, we compared the coverage of rheumatic illness with the coverage of heart disease, cancer, chronic lung disease, and diabetes meUitus by the Dutch mass
CHRONIC DISEASES IN THE NETHERLANDS
media. The amount of coverage and the nature of the information presented were analyzed. The units analyzed were newspaper and magazine articles, and medical programs on television, which represented 12 months of publishing and broadcasting. Although radio continues to play an important role in Europe and North America, for practical reasons this medium was not analyzed. METHODS
All editions of nine Dutch newspapers and magazines and the programs on four Dutch language television channels that have appeared, or were transmitted, between April 1, 1992 and March 31, 1993, were reviewed for their coverage of rheumatic disease, heart disease, cancer, chronic lung disease, and diabetes mellitus. The newspapers and magazines belonged t o four widely read categories. They comprised three national daily newspapers with circulations of approximately 726,000; 414,000, and 342,000 at the time of the study, two regional daily newspapers with circulations of 141,000 and 136,000, two gossip maga, zines with circulations of 492,000 and 417,000, and two women's weekly magazines with circulations of 778;000 and 535,0001 Each was among the most wide!y circulated within its category (population of the Netherlands = 15 238i000). The television channels included all of those that, during the review, directed their p r o g r a ~ g a t a Dutch-speaking audience and were widely available throughout the Netherlands, Three were: public channels, and one was an independent commercial channel. The first author and aresearch assistant independently identified ~ o s e art~c!es and programs dealing with the diseases under consideration. The articles were found by reading an entire copy of every edition of each newspaper and magazine. To assess the exhaustiveness o f this identification procedure, the intercoder reliability (et = 1 - (observed disagreement/expected disagreement) [23]) was calculated for 10% of the editions under review, selected at random. This intercoder reliability was 0.90. The p r o g r m s were found by watching videotapes of all of the programs described by a major television guide as dealing with health or medical topics Each reviewer independently analyzed the contents of all articles and programs that either had identified. Programs repeated during the study period were analyzed oniy once. A data sheet was
335
completed for each disease category covered by an article or program, except when coverage consisted of no more than causal mention. The title of the article or program, and the date when published or broadcast, were recorded. The area of the page or pages covered by an article was recorded in square centimeters, and the duration of a program was recorded in minutes. It also was noted whether articles were illustrated, whether they were within health columns, and at what time of the day television programs were broadcast. Statistics for prevalences, numbers of deaths. and numbers of work disabilities in the Netherlands were provided for each of the five disease categories by previously published sources (24,25). Pearson correlation coefficients were calculated among the numbers of articles and programs dealing with the five disease categories, and the associated prevalences, numbers of deaths, and numbers of work disabilities. Values of P less than .05 (twosided) were taken to indicate statistical significance. For each article and program, the information presented was classified as concerning treatment, disease consequences, influence of lifestyle, or scientific progress. Treatments were subclassified as regular medications, operations, or alternative therapy, and whether they were described as positive. negative, or neutral. If disease consequences were included in the coverage, it was noted whether they included incurability or fatality. The influence of lifestyle was defined as a way of living by which patients may themselves avoid, relieve, or cure their illness, for example, by not smoking or by exercising. It was also noted whether the opinions of patients or professionals (ie, medical doctors and scientists) were provided. These classifications provided information about the intentions and opinions of the journalists or programmers. The intercoder reliability for all items was assessed with Cohen's Kappa (26). For all items. Cohen's Kappa levels were sufficiently high (0.64 < Kappa < 0.88) to justify statistical analysis of the results. The associations among the five disease categories with respect to the items traced in the content analyses were tested by chi-square analysis. Values of P less than .01 were taken to indicate statistical significance.
336
VAN DER WARDT ET AL
RESULTS Eight hundred fifty-one newspaper articles and 285 magazine articles were found that covered one or more of the five disease categories. Some articles covered more than one disease category. In the 851 newspaper articles, one of the five disease categories was covered 917 times, and in the 285 magazine articles, a single disease category was covered 304 times. Thirty-two percent of the newspaper articles and 47% of the magazine articles were illustrated with pictures. The average areas covered by text and illustration in newspaper and magazine articles were 328 cm 2 and 639 cm 2, respectively; 9% of these newspaper articles and 36% of the magazine articles were within health columns. One hundred twenty-five television programs, with a mean duration of 27 minutes, gave more than casual mention to 163 individual disease categories. The mean duration of the segments dealing with an individual category was 7 minutes. Forty-nine percent of the broadcasts took place during weekends and 47% during prime time (8:00 to 11:00 PM). The most frequently featured disease categories, regardless of medium, were heart disease and cancer (Table 1). Newspapers, magazines, and television differed significantly in the relative amounts of attention they gave to cancer, diabetes, and rheumatic diseases (P < .001). The proportions of newspaper and magazine articles that gave attention to cancer was greater than the proportion of television programs covering cancer. The proportion of newspaper articles about rheumatic diseases was smaller than the proportions of magazine articles and television programs about rheumatic diseases. The proportions of newspaper and maga-
zine articles covering diabetes was smaller than the proportion of television programs covering the same topic. Composite prevalences, numbers of deaths, and numbers of work disabilities are given in Figure 1. The amount of media attention received was generally related to the numbers of deaths, but not to prevalences or numbers of work disabilities. The correlation with numbers of deaths was significant for both newspapers and television programs (newspapers, r = .91, P < .05; magazines, r -- .84, NS; television, r = .97, P < .01). There were no statistically significant correlations between the amount of media attention and prevalences (newspapers, r = - . 4 3 , NS; magazines, r = - . 2 9 , NS; television, r = - . 3 6 , NS) or numbers of work disabilities (newspapers, r = - . 2 1 , NS; magazines, r = - . 0 6 , NS; television, r = - . 0 3 , NS). The proportion of television programs featuring individuals was larger than that of articles in either newspapers or magazines, regardless of disease category (Table 2). Patients, but not professionals, were portrayed in a large proportion of magazine articles. Heart disease included the largest proportion of articles and programs featuring professional experts; the coverage of rheumatic diseases showed the largest proportion of articles and programs featuring patients. However, the highest proportion of doctors or scientists was for programs on cancer. The lowest proportions of articles or programs featuring patients was found for newspaper coverage of chronic lung diseases and diabetes. The proportion of newspaper articles dealing with treatment was generally smaller than that of magazine articles or television programs (Table 3). Treatment with regular medications received most attention in diabetes, less attention in rheumatic
Table 1: Chronic Diseases in Newspaper and Magazine Articles, and on Medical Television Programs
Rheumatic disease Heart disease Cancer Chronic lung disease Diabetes
Newspaper Articles
Magazine Articles
46 349 419 67 36 917
41 (13%) 99 (33%) 132 (43%) 15 (5%) 17 (6%) 304 (100%)
(5%) (38%) (46%) (7%) (4%) (100%)
Medical TV Programs 20 64 40 15 24 163
(12%) (39%) (25%) (9%) (15%) (100%)
Total
X 2 (dr = 2)*
107 (8%) 512 (37%) 591 (43%) 97 (7%) 77 (6%) 1,384 (100%)
28.3t 3.4 25.4t 3.3 30.7t
*X2With respect to differences between media coverage of each disease and the media coverage of the other four diseases. t P < .001.
CHRONIC DISEASES IN THE NETHERLANDS
337
219.000
52.300
458.000
34.200
Pravalences
Numbers of deaths 7.200
251
Numbers of work disabilities [ ] Heart d i m
[ ] Cancer [] Rheumatism [ ] Lung disease [ ] Diabetes
Fig 1. Prevalence, numbers of deaths, and numbers of work disabilities due to rheumatic diseases, heart disease, cancer, chronic lung disease and diabetes mellitus in the population of the Netherlands. Prevalences in 1991 (24); numbers of deaths and numbers of work disabilities in 1988 {25), Table 2: Professionals and Patients in Newspaper and Magazine Articles, and on Medical Television Programs per Five Chronic Diseases Rheumatic Disease
Heart Disease
Cancer
Chronic Lung Disease
Diabetes
x2(df=4)t
Professional expert* Newspapers Magazines TV programs Total Patients*
19 (41%) 7 (17%) 9 (45%) 35 (33%)
174 22 47 243
(50%) (22%) (73%) (47%)
165 30 36 231
(39%) (23%) (90%) (39%)
22 4 8 34
(33%) (27%) (53%) (35%)
13 (36%) 3 (18%) 12 (50%) 28 (36%)
12.5 1.0 19.95 14.8§
Newspapers Magazines TV programs Total
17 (37%) 30 (73%) 16 (80%) 63 (59%)
103 71 41 215
(30%) (72%) (64%) (42%)
118 99 22 239
(28%) (75%) (55%) (40%)
3 6 11 20
(4%) (40%) (73%) (21%)
6 (17%) 10 (59%) 16 (67%) 32 (42%)
22.95 9.4 4.2 31.15
*Numbers and percentages of articles and programs on each disease in which professional experts and patients are portrayed. t× 2 With respect to differences between diseases, considering the portrayal of professional experts and patients in articles and programs. SP < .001. §P< .01.
338
VAN DER WARDT ET AL
Table 3: Conventional Medications, Surgery, and Alternative Treatments in Newspaper and Magazine Articles, and on Medical Television Programs for Five Chronic Diseases Rheumatic Disease
Regular medications* Newspapers Magazines TV programs Total Surgery* Newspapers Magazines TV programs Total Alternative treatments* Newspapers Magazines TV programs Total
Heart Disease
Cancer
Chronic Lung Disease
Diabetes
23 (34%) 7 (47%) 10 (67%) 40 (41%)
17 (47%) 10 (59%) 19 (79%) 46 (60%)
x2(df = 4)t
20 (43%) 19 (46%) 7 (35%) 46 (43%)
78 31 17 126
(22%) (31%) (27%) (25%)
80 (19%) 43 (33%) 13 (33%) 136 (23%)
4 8 9 21
119 43 29 191
(34%) (43%) (45%) (37%)
94 63 20 177
(22%) (48%) (50%) (30%)
0 (0%) 0 (0%) 1 (7%) 1 (1%)
0 2 2 4
(0%) (12%) (8%) (5%)
59.0* 26.6* 19.6, 80.75
14 (4%) 7 (7%) 11 (17%) 32 (6%)
27 8 2 37
(6%) (6%) (5%) (6%)
2 (3%) 2 (13%) 1 (7%) 5 (5%)
6 2 3 11
(17%) (12%) (13%) (14%)
65.65 45.75 10.7 125.5,
(9%) (20%) (45%) (20%)
16 (35%) 18 (44%) 7 (35%) 41 (38%)
30.4* 8.1 25.6* 68.2*
*Numbers and percentages of articles and programs on each disease in which conventional medications, surgery, and alternativ~ treatments are mentioned. 1-×2with respect to differences between diseases,considering the mentioning of conventional medications, surgery, and alternative treatments in articles and programs. ~P<.OOl.
and chronic lung diseases, and least attention in heart disease and cancer. Surgery was reported the most in heart disease and cancer, regardless of medium, and in television programs about rheumatic disease. Alternative treatments were reported most often, in all media, in relation to rheumatic diseases. Standard medications and surgery mostly were described in a neutral way, whereas mainly positive aspects of alternative treatments were highlighted (Table 4). Conventional medications for heart disease were described more often in a positive way than for the other diseases, and standard medications for rheumatic diseases were described more often in a negative way compared with other diseases. For diabetes, conventional medications frequently were described in a neutral way. Cancer surgery was often described negatively compared with other diseases. No significant differences were found in the way alternative treatments for the five disease categories were described in articles and programs. The proportion of articles mentioning influence of lifestyle was highest for newspaper articles on rheumatic diseases and diabetes (Table 5). Influ-
ence of lifestyle was covered least often in articles and programs about cancer and heart disease. There were no significant differences in the coverage of scientific progress among the five disease categories (Table 6). Incurability coverage was highest for diabetes and chronic lung disease (Table 7). The proportion of magazine articles covering incurability was greatest for cancer and lowest for heart disease. Much media attention was given to fatality in cancer and heart disease (Table 7). A high proportion of magazine articles reported fatality in cancer patients. Diabetes occasionally was mentioned as a cause of death in newspapers, but not in magazines or television programs. Rheumatic disease was reported as a cause of death only once. DISCUSSION The amount of attention given to a disease or disease category by the media depends on its fatality rate, not its prevalence. Both in our study and in an American study of chronic diseases presented on national television news (14,15), the diseases receiving the most attention were those with the highest death rates, heart disease and
CHRONIC DISEASES iN THE NETHERLANDS
339
Table 4: Positive, Neutral or Negative Judgments of Conventional Medications, Surgery, and Alternative Treatments in Newspaper and Magazine Articles, and on Medical Television Programs for Five Chronic Diseases Rheumatic Disease
Heart Disease
Cancer
Chronic Lung Disease
Diabetes
Regular medications* Positive Neutral Negative Total x2(df = 2)t
8 (19%) 22 (47%) 16 (34%) 46 (100%) 11.5~
56 (44%) 54 (43%) 16 (13%) 126 (100%) 17.4§
30 (22%) 75 (55%) 31 (23%) 136 (100%) 9.1
13 (33%) 23 (57%) 4 (10%) 40 (100%) 1.1
12 (26%) 34 (74%) 0 (0%) 46 (100%) 13.6$
Surgery* Positive Neutral Negative Total x2(df = 2)t
8 (38%) 12 (57%) 1 (5%) 21 (100%) 3,4
54 (28%) 111 (58%) 26 (14%) 191 (100%) 9.0
48 (27%) 79 (45%) 50 (28%) 177 (100%) 16,2~
0 (0%) 1 (100%) 0 (0%) 1 (100%) 1.0
1 (25%) 3 (75%) 0 (0%) 4 (100%) 1.2
Alternative treatments* Positive Neutral Negative Total ×2(dr = 2)1"
25 (62%) 9 (23%) 6 (15%) 40 (100%) 1.6
25 (78%) 5 (16%) 2 (6%) 32 (100%) 1,8
27 (73%) 8 (22%) 2 (5%) 37 (100%) 1.3
5 (100%) 0 (0%) 0 (0%) 5 (100%) 2.3
4 (36%) 4 (36%) 3 (28%) 11 (100%) 6.8
*Numbers and percentages of articles and programs on each disease in which conventional medications, surgery, and alternative treatments are judged positive, neutral, or negative. I-X2with respect to differences in judgment of treatments for each disease and judgment of treatments for the other four diseases. tP < .01. §P< .001.
cancer. Neither rheumatic diseases nor chronic lung disease were featured nearly as often, yet both are more prevalent and responsible for more work disability. No correlation between the frequency of coverage and mortality statistics was found in a study of newspaper coverage of 27 disorders (18). However, the latter included many rare diseases
that seldom cause death, such as rabies, Down syndrome, and food poisoning. Cancer and heart disease are serious conditions because they are the biggest killers, are frequently disabling, and are costly to the health care system. Our study showed that these diseases were covered more frequently than and differently from the more
Table 5: Influence of Lifestyle in Newspaper and Magazine Articles, and on Medical Television Programs for Five Chronic Diseases Rheumatic Disease Influence of lifestyle* Newspapers Magazines TV programs Total
23 (50%) 15 (37%) 7 (35%) 45 (42%)
Heart Disease 111 23 31 165
(32%) (23%) (48%) (32%)
Cancer 115 (27%) 17 (!3%) 14 (35%) 146 (25%)
Chronic Lung Disease 22 4 11 37
(33%) (27%) (73%) (38%)
Diabetes
xZ(df= 4)t
17 (47%) 5 {29%) 12 (50%) 34 (44%)
14.75 12,5 7.7 26,0§
*Numbers and percentages of articles and programs on each disease in which influence of lifestyle is mentioned, t× 2With respect to differences between diseases, considering the influence of lifestyle in articles and programs. SP< .01. §P< .001.
340
VAN DER WARDT ET AL
Table 6: Scientific Progress in Newspaper and Magazine Articles, and on Medical Television Programs for Five Chronic Diseases Chronic Lung
Rheumatic Disease
Scientific Progress* Newspapers Magazines TV programs Total
15 5 3 23
(33%) (12%) (15%) (21%)
Heart Disease
138 15 15 168
(40%) (15%) (23%) (33%)
Cancer
183 14 8 205
(44%) (11%) (20%) (35%)
Disease
Diabetes
37 2 3 42
18 5 8 31
(55%) (13%) (20%) (43%)
x2(df =
(50%) (29%) (33%) (40%)
4)t
8.6 4.9 2.5 13.0
*Numbers and percentages of articles and programs on each disease in which scientific progress is mentioned. tx 2With respect to differences between diseases, considering scientific progress in articles and programs.
ences in media attention they received may suggest that heart disease and cancer are more common than rheumatic or chronic lung diseases. Experts were more often featured and fatality more often mentioned in the coverage of heart disease and cancer, which may add to the impression of their severity. Reports on the influence of lifestyle and alternative treatments in rheumatic diseases may lead to the misleading impression that they can be easily avoided, relieved, or even cured by patients themselves. Many important questions remain: What can professionals in medical science and journalism do to ensure responsible media coverage of medical issues? Who is to blame for public confusion about medical science: the media, medical scientists, or an uncritical public? To what extent does press coverage of medicine influence politicians? How important is the role of the lay press in the
lingering, chronic disorders. The proportion of articles and programs on heart disease and cancer featuring professionals was striking. The specific subjects dealt with often included surgery as well as fatality. Rheumatic and chronic lung disease and diabetes coverage were more likely to feature standard medications and the influence of lifestyle. The large amount of attention on lifestyle in relation to rheumatic diseases is particularly surprising. One might expect to see more attention given to the influence of lifestyle on cancer (eg, smoking habits, sunlight exposure) or heart disease (smoking, diet, exercise), which are widely accepted associations. Many articles and programs on rheumatic diseases featured patients and treatments involving alternative medicine. If the public's notions of the five diseases are similarly influenced by media coverage, several misconceptions are likely to prevail. The differ-
Table 7: Incurability and Fatality in Newspaper and Magazine Articles, and on Medical Television Programs for Five Chronic Diseases Rheumatic Disease
Incurability* Newspapers Magazines TV-programs Total Fatality* Newspapers Magazines mV-programs Total
9 3 3 15
(20%) (7%) (15%) (14%)
0 (0%) 1 (2%) 0 (0%) 1 (1%)
Heart Disease
36 (10%) 2 (2%) 1 (2%) 39 (8%) 147 33 20 200
(42%) (33%) (31%) (39%)
Cancer
Chronic Lung Disease
59 22 5 86
(14%) (17%) (13%) (15%)
15 2 3 20
(22%) (13%) (20%) (21%)
159 79 11 249
(38%) (60%) (28%) (42%)
11 2 3 16
(16%) (13%) (20%) (16%)
Diabetes
x2(df
8 (22%) 1 (6%) 5 (21%) 14 (18%) 6 0 0 6
(17%) (0%) (0%) (8%)
*Numbers and percentages of articles and programs on each disease in which incurability and fatality are mentioned. tx 2With respect to differences between diseases, considering the incurability and fatality in articles and programs. ~:P< .01. §P< .OOl.
=
4)t
11.1 14.55 10,3 21.7§ 49.5§ 64.1§ 16.95 112.0§
CHRONIC DISEASES IN THE NETHERLANDS
341
transmission of medical knowledge not only to the public but also to the scientific community (27)? Our findings indicate the need for more frequent and more accurate media coverage of chronic lung disease, diabetes mellitus, and especially rheumatic disorders. Foundations, doctors, and scientists concerned with rheumatic diseases should take more advantage of the opportunities for public education
provided by the mass media. More attention from the media could lead to greater public awareness of rheumatic diseases and, in turn, contribute to a better understanding of people suffering from them. ACKNOWLEDGMENT
The authors thank W. Burke and J:M. Cornwell for their help with English.
REFERENCES 1. Atkin C, Arkin EB. Issues and initiatives in communicating health information to the public. In: Atkin C, Wallack L, editors. Mass communication and public health. Newbury Park, CA: Sage, 1990:13-40. 2. Signorielli N. Mass media images and impact on health. Westport, CT: Greenwood, 1993. 3. Wade S, Schramm W. The mass media as sources ofpublic affairs, science, and health knowledge. Publ Opin Q 1969;33:197209. 4. Gerbner G, Gross L, Morgan M, Signorielli N. Special report: health and medicine on television. N Engl J Med 1981;305:901-4. 5. McCombs ME, Shaw DL. The agenda-setting function of mass media. Publ Opin Q 1972;36:176-87. 6. Wallack L. Improving health promotion: Media advocacy and social marketing approaches. In: Atkin C, Wallack L, editors. Mass communication and public health. Newbury Park, CA: Sage, 1990:147-63. 7. Bandura A: Social foundations of thought and action: A social cognitive theory. Englewood Cliffs, NJ: Prentice-Hall, 1986. 8. Gerbner G, Gross L. Living with television: The violence profile. J Commun 1976;26:t73-99. 9. Philo G, Secker J, Platt: S, Henderson L, McLaughfin G, Burnside J. The impact of the mass media on public images of mental illness: Media:content and audience belief. Health Educ J 1994;53:271-81. 10. Price JH, Hillman KS, Toral ME, Newell S. The public's perceptions and misperceptions of arthritis. Arthritis Rheum 1983;26:1023-& 11. Banman A, Antic R, Rubinfeld A, Zimmerman R Gutch D, Bacon B, et al. "Could it be astma?": The impact of a mass media campaign aimed at raising awareness about astma in Australia. Health Educ Res 1993;8:581-7. 12. Combs B, Slovic R Newspaper coverage of causes of death. Journalism Q 1979;56:837-43. 13. Ettema JJ, Whitney DC: Audiencemaking: how the
media create the audience. Thousand Oaks. CA: SAGE Publications. 1994. 14. Pichert JW. Diabetes in the national TV-news: 19711981. Diabetes Care 1983:6:92-4. 15. Pichert JW, Hanson SL. Arthritis in the national TVnews: 1971-1981. J Rheum 1983:10:323-5. 16. Garland R. Images of health and medical science conveyed by television. J R Coll Gen Pract 1984:34:316-9. 17. Johnson CA, Johnson BE. Medicine on British television: A content analysis. J Commun Health 1993:18:25-35. 18. Kristiansen CM. Newspaper coverage of diseases and actual mortality statistics. Eur J Soc Psychnl 1983:13:193-4. 19. Freimuth VS. Greenberg RH, DeWitt J, Romano ILM. Coveting cancer: Newspapers and the public interest. J Commun 1984;34:62-73. 20. Greenberg M, Wartenberg D. Newspaper coverage of cancer clusters. Health Educ Q 1991; 18:363-74. 21. Miner KJ, Baker JA. Media coverage of suntanning and skin cancer: Mixed messages of health and bean~y. J Health Educ 1994:25:234-8. 22. ReynoldsT. Media coverage of cancer risks may mislead. J Nad Cancer Institute 1993;85:1366-8. 23. Krippendorff K. Content analysis: An introduction to its methodology. Newbury Park. CA: Sage, 1980. 24. Commissie voor Keuzen in de Zorg. Kiezen en delen: Achtergrondstndies. [Committee for Choices in Care. Choose and divide: Background studies.] Den Hang: Albani, 1991. 25. Raad voor Gezondheidsonderzoek. Advies chronische aandoeningen: Prioriteiten voor onderzoek. [Board of Health Research: Advice on chronic diseases: Priorities for research.] Den Hang: Stantsuitgeverij, 1991. 26. Reynolds HT. The analysis of cross-classifications. New York: Free Press. 1977. 27. Phillips DP, Kanter EJ. Bednarczyk B. Tastad PL. Importance of the lay press in the transmission of medical knowledge to the scientific community. N Engl J Med 1991:325: 1180-3.
Objective: Little is known about the quantity or quality of information on rheumatic diseases provided by the mass media, The aim of this study was to gain insight into the media coverage of rheumatic diseases compared w i t h other chronic diseases in the Netherlands. Materials and Methods: Newspaper articles, magazine articles, and medical television programs that appeared or were broadcast during a 1-year period, and contained information on rheumatic diseases, heart disease, cancer, chronic lung disease, or diabetes mellitus, were selected for content analysis. For each article and program, it was determined whether coverage concentrated on treatment, influence of lifestyle, scientific progress, or disease consequences. It was also determined whether professional experts and patients were featured. Results: Nine hundred seventeen newspaper articles, 304 magazine articles, and 163 medical programs were found. Most dealt w i t h cancer (43%) and heart disease (37%). The amount of media attention given to each of the five disease categories was found to correspond w i t h mortality but not w i t h prevalence. The contents of the articles and programs differed significantly according to disease topic. The main focus in rheumatic diseases was on patients" experiences, as well as regular and alternative medications. In heart disease and cancer, the main focus was on professional medical viewpoints, operations, and mortality, whereas in chronic lung disease and diabetes it was on treatments in the context of regular medications, scientific progress, and incurability. The influence of lifestyle on the disease process was mentioned most often in connection w i t h diabetes, rheumatic disease, and chronic lung disease. Conclusions: The amount of attention a disease category received from the media depended on its fatality rates and not on its prevalence. Heart disease and cancer were portrayed as being more serious than the more lingering diseases. Surprisingly, the proportion of articles and programs that included the influence of lifestyle in their coverage was lowest for cancer and heart disease. More frequent and more accurate coverage of chronic diseases, especially rheumatic diseases, is needed if their image is to be brought into line w i t h their importance for and impact on the community. Semin Arthritis Rheum 28:333-341, Copyright © 1999 by W.B, Saunders Company INDEX WORDS: Media; chronic diseases; Netherlands.
From the Department of Communication Studies, University of Twente, and the Department of Rheumatology, Hospital of Twente [Medisch Spectrum Twente], Enschede, the Netherlands. Elly M. van der Wardt, PhD: Communication scholar and Linguist, Department of Communication Studies, University of Twente; Erik Taal, PhD: Social psyehologist, Department of Communication Studies, University of Twente; Johannes J. Rasker, MD, PhD: RheumatoIogist, Department of Rheumatology, Hospital of Twente [Medisch Spectrum Twente], and Professor, Department of Communication Studies, University of
Twente; Oene Wiegman, PhD: Professor in Psychology, Department of Communication Studies, University of Twente. Supported by grants from the Dutch League against Rheumatism [National Reumafonds] and the Min&try of Health, Welfare, and Sport [Ministerie van Volksgezondheid, Welzijn en Sport] of the Netherlands. Address reprint requests to Professor Johannes J. Raske~ MD, PhD, Department of Communication Studies, University of Twente, PO Box 217, 7500 AE Enschede, The Netherlands. Copyright © 1999 by W.B. Saunders Company 0049-0172/99/2805-00055!0.00/0
Seminars in Arthritis and Rheumatism, Voi 28, No 5 (April), 1999: pp 333-341
333
334
TTLE IS KNOWN about the quantity or quality of information on rheumatic diseases provided by the mass media. The aim of this study was to gain insight into the media coverage of rheumatic diseases compared with other chronic diseases in the Netherlands. Public interest in health issues has increased enormously over the last three decades; health columns in newspapers and magazines are flourishing, health claims are prominent in advertising, and more and more health-related programs are broadcast by an increasing number of television channels (1,2). About 30 years ago, a study showed that the printed media were the American public's main source of health information (3). A later study found that newspapers and magazines had been replaced by television as the most important source of health information (4). It is evident, then, that both the printed and broadcast media should be considered when investigating media coverage of chronic diseases. Media coverage of chronic diseases may have a great impact on the general public's perception of the causes, treatments, and consequences of these conditions. An important function of the mass media is setting the public agenda and stimulating public discussion (5,6). During the course of their lives, people have direct contact with only a small sector of the environment, so their impressions of many social realities are greatly influenced by the information they receive from the mass media (7). The images presented by the media, particularly television, may influence beliefs concerning health and health behavior. Television contributes to viewers' conceptions of social roles and human relations (8). Frequent portrayal of violence in connection with mental illness, for example, persuades audiences that the two are associated (9). Frequent television viewing can lead to a high degree of confidence in the medical profession's ability to cure disease, which may justify unhealthy lifestyles and lack of interest in disease prevention (4). The mass media have been held responsible both for the public's misconceptions about arthritis (10) and for its increased awareness of asthma (11). Newspaper coverage of homicides, accidents, and disasters influences the public's ideas about the main causes of death (12). The mass media can influence the opinions of the public, but public opinion also shapes the style and presentation of media news. Audience members have well-formed preferences and opinions that
E
VAN DER WARDT ET AL
allow them to selectively choose what to read, watch, or hear in the media. To attract and maintain audiences, journalists and program-makers must bear in mind the public's preferences (13). There have been few studies on the coverage of chronic diseases and health by the mass media (12,14-22). These studies were subject to methodological restrictions on data collection. Most studies concentrated on one medium only or on one category of presentation within one medium: television news reports (14,15), medical programs on television (16,17), newspaper articles (12,18-20), or magazine articles (21). In some, the period studied was 4 months or less (16-19). Nearly all of the studies relied on the indexes of news archives for their data collection (14,15,19,21,22); therefore, results were dependent on the key words provided by the indexers. Media coverage of medical topics often is biased. Several American studies have focused on news coverage of cancer. Newspapers provided scarcely any information on incidence rates, distinct types of cancer, prevention, detection, or treatment (19), and few focused on scientific background information (20). The media overstated the risks from pollution, food additives, and nuclear power, and understated those inherent in the use of tobacco and exposure to sunlight (22). The popular press paid little or no attention to skin cancer morbidity or mortality (21). Two studies on medical programming by the British Broadcasting Corporation showed that most television programs were hospital-based, highlighted aspects of medical technology, and that hospital specialists were the most prominent communicators. Few programs dealt with health maintenance, preventive medicine, risk factor avoidance, or common diseases (16,17). It seems important to know how and when a particular disease becomes newsworthy. Does media coverage of chronic disorders, such as rheumatic disease, chronic lung disease, or diabetes, differ in meaningful ways from coverage of conditions that are potentially fatal, such as heart disease and cancer? Although heart disease and cancer can be fatal, they are increasingly becoming chronic, because treatments for these diseases have become more effective. To answer these questions, we compared the coverage of rheumatic illness with the coverage of heart disease, cancer, chronic lung disease, and diabetes meUitus by the Dutch mass
CHRONIC DISEASES IN THE NETHERLANDS
media. The amount of coverage and the nature of the information presented were analyzed. The units analyzed were newspaper and magazine articles, and medical programs on television, which represented 12 months of publishing and broadcasting. Although radio continues to play an important role in Europe and North America, for practical reasons this medium was not analyzed. METHODS
All editions of nine Dutch newspapers and magazines and the programs on four Dutch language television channels that have appeared, or were transmitted, between April 1, 1992 and March 31, 1993, were reviewed for their coverage of rheumatic disease, heart disease, cancer, chronic lung disease, and diabetes mellitus. The newspapers and magazines belonged t o four widely read categories. They comprised three national daily newspapers with circulations of approximately 726,000; 414,000, and 342,000 at the time of the study, two regional daily newspapers with circulations of 141,000 and 136,000, two gossip maga, zines with circulations of 492,000 and 417,000, and two women's weekly magazines with circulations of 778;000 and 535,0001 Each was among the most wide!y circulated within its category (population of the Netherlands = 15 238i000). The television channels included all of those that, during the review, directed their p r o g r a ~ g a t a Dutch-speaking audience and were widely available throughout the Netherlands, Three were: public channels, and one was an independent commercial channel. The first author and aresearch assistant independently identified ~ o s e art~c!es and programs dealing with the diseases under consideration. The articles were found by reading an entire copy of every edition of each newspaper and magazine. To assess the exhaustiveness o f this identification procedure, the intercoder reliability (et = 1 - (observed disagreement/expected disagreement) [23]) was calculated for 10% of the editions under review, selected at random. This intercoder reliability was 0.90. The p r o g r m s were found by watching videotapes of all of the programs described by a major television guide as dealing with health or medical topics Each reviewer independently analyzed the contents of all articles and programs that either had identified. Programs repeated during the study period were analyzed oniy once. A data sheet was
335
completed for each disease category covered by an article or program, except when coverage consisted of no more than causal mention. The title of the article or program, and the date when published or broadcast, were recorded. The area of the page or pages covered by an article was recorded in square centimeters, and the duration of a program was recorded in minutes. It also was noted whether articles were illustrated, whether they were within health columns, and at what time of the day television programs were broadcast. Statistics for prevalences, numbers of deaths. and numbers of work disabilities in the Netherlands were provided for each of the five disease categories by previously published sources (24,25). Pearson correlation coefficients were calculated among the numbers of articles and programs dealing with the five disease categories, and the associated prevalences, numbers of deaths, and numbers of work disabilities. Values of P less than .05 (twosided) were taken to indicate statistical significance. For each article and program, the information presented was classified as concerning treatment, disease consequences, influence of lifestyle, or scientific progress. Treatments were subclassified as regular medications, operations, or alternative therapy, and whether they were described as positive. negative, or neutral. If disease consequences were included in the coverage, it was noted whether they included incurability or fatality. The influence of lifestyle was defined as a way of living by which patients may themselves avoid, relieve, or cure their illness, for example, by not smoking or by exercising. It was also noted whether the opinions of patients or professionals (ie, medical doctors and scientists) were provided. These classifications provided information about the intentions and opinions of the journalists or programmers. The intercoder reliability for all items was assessed with Cohen's Kappa (26). For all items. Cohen's Kappa levels were sufficiently high (0.64 < Kappa < 0.88) to justify statistical analysis of the results. The associations among the five disease categories with respect to the items traced in the content analyses were tested by chi-square analysis. Values of P less than .01 were taken to indicate statistical significance.
336
VAN DER WARDT ET AL
RESULTS Eight hundred fifty-one newspaper articles and 285 magazine articles were found that covered one or more of the five disease categories. Some articles covered more than one disease category. In the 851 newspaper articles, one of the five disease categories was covered 917 times, and in the 285 magazine articles, a single disease category was covered 304 times. Thirty-two percent of the newspaper articles and 47% of the magazine articles were illustrated with pictures. The average areas covered by text and illustration in newspaper and magazine articles were 328 cm 2 and 639 cm 2, respectively; 9% of these newspaper articles and 36% of the magazine articles were within health columns. One hundred twenty-five television programs, with a mean duration of 27 minutes, gave more than casual mention to 163 individual disease categories. The mean duration of the segments dealing with an individual category was 7 minutes. Forty-nine percent of the broadcasts took place during weekends and 47% during prime time (8:00 to 11:00 PM). The most frequently featured disease categories, regardless of medium, were heart disease and cancer (Table 1). Newspapers, magazines, and television differed significantly in the relative amounts of attention they gave to cancer, diabetes, and rheumatic diseases (P < .001). The proportions of newspaper and magazine articles that gave attention to cancer was greater than the proportion of television programs covering cancer. The proportion of newspaper articles about rheumatic diseases was smaller than the proportions of magazine articles and television programs about rheumatic diseases. The proportions of newspaper and maga-
zine articles covering diabetes was smaller than the proportion of television programs covering the same topic. Composite prevalences, numbers of deaths, and numbers of work disabilities are given in Figure 1. The amount of media attention received was generally related to the numbers of deaths, but not to prevalences or numbers of work disabilities. The correlation with numbers of deaths was significant for both newspapers and television programs (newspapers, r = .91, P < .05; magazines, r -- .84, NS; television, r = .97, P < .01). There were no statistically significant correlations between the amount of media attention and prevalences (newspapers, r = - . 4 3 , NS; magazines, r = - . 2 9 , NS; television, r = - . 3 6 , NS) or numbers of work disabilities (newspapers, r = - . 2 1 , NS; magazines, r = - . 0 6 , NS; television, r = - . 0 3 , NS). The proportion of television programs featuring individuals was larger than that of articles in either newspapers or magazines, regardless of disease category (Table 2). Patients, but not professionals, were portrayed in a large proportion of magazine articles. Heart disease included the largest proportion of articles and programs featuring professional experts; the coverage of rheumatic diseases showed the largest proportion of articles and programs featuring patients. However, the highest proportion of doctors or scientists was for programs on cancer. The lowest proportions of articles or programs featuring patients was found for newspaper coverage of chronic lung diseases and diabetes. The proportion of newspaper articles dealing with treatment was generally smaller than that of magazine articles or television programs (Table 3). Treatment with regular medications received most attention in diabetes, less attention in rheumatic
Table 1: Chronic Diseases in Newspaper and Magazine Articles, and on Medical Television Programs
Rheumatic disease Heart disease Cancer Chronic lung disease Diabetes
Newspaper Articles
Magazine Articles
46 349 419 67 36 917
41 (13%) 99 (33%) 132 (43%) 15 (5%) 17 (6%) 304 (100%)
(5%) (38%) (46%) (7%) (4%) (100%)
Medical TV Programs 20 64 40 15 24 163
(12%) (39%) (25%) (9%) (15%) (100%)
Total
X 2 (dr = 2)*
107 (8%) 512 (37%) 591 (43%) 97 (7%) 77 (6%) 1,384 (100%)
28.3t 3.4 25.4t 3.3 30.7t
*X2With respect to differences between media coverage of each disease and the media coverage of the other four diseases. t P < .001.
CHRONIC DISEASES IN THE NETHERLANDS
337
219.000
52.300
458.000
34.200
Pravalences
Numbers of deaths 7.200
251
Numbers of work disabilities [ ] Heart d i m
[ ] Cancer [] Rheumatism [ ] Lung disease [ ] Diabetes
Fig 1. Prevalence, numbers of deaths, and numbers of work disabilities due to rheumatic diseases, heart disease, cancer, chronic lung disease and diabetes mellitus in the population of the Netherlands. Prevalences in 1991 (24); numbers of deaths and numbers of work disabilities in 1988 {25), Table 2: Professionals and Patients in Newspaper and Magazine Articles, and on Medical Television Programs per Five Chronic Diseases Rheumatic Disease
Heart Disease
Cancer
Chronic Lung Disease
Diabetes
x2(df=4)t
Professional expert* Newspapers Magazines TV programs Total Patients*
19 (41%) 7 (17%) 9 (45%) 35 (33%)
174 22 47 243
(50%) (22%) (73%) (47%)
165 30 36 231
(39%) (23%) (90%) (39%)
22 4 8 34
(33%) (27%) (53%) (35%)
13 (36%) 3 (18%) 12 (50%) 28 (36%)
12.5 1.0 19.95 14.8§
Newspapers Magazines TV programs Total
17 (37%) 30 (73%) 16 (80%) 63 (59%)
103 71 41 215
(30%) (72%) (64%) (42%)
118 99 22 239
(28%) (75%) (55%) (40%)
3 6 11 20
(4%) (40%) (73%) (21%)
6 (17%) 10 (59%) 16 (67%) 32 (42%)
22.95 9.4 4.2 31.15
*Numbers and percentages of articles and programs on each disease in which professional experts and patients are portrayed. t× 2 With respect to differences between diseases, considering the portrayal of professional experts and patients in articles and programs. SP < .001. §P< .01.
338
VAN DER WARDT ET AL
Table 3: Conventional Medications, Surgery, and Alternative Treatments in Newspaper and Magazine Articles, and on Medical Television Programs for Five Chronic Diseases Rheumatic Disease
Regular medications* Newspapers Magazines TV programs Total Surgery* Newspapers Magazines TV programs Total Alternative treatments* Newspapers Magazines TV programs Total
Heart Disease
Cancer
Chronic Lung Disease
Diabetes
23 (34%) 7 (47%) 10 (67%) 40 (41%)
17 (47%) 10 (59%) 19 (79%) 46 (60%)
x2(df = 4)t
20 (43%) 19 (46%) 7 (35%) 46 (43%)
78 31 17 126
(22%) (31%) (27%) (25%)
80 (19%) 43 (33%) 13 (33%) 136 (23%)
4 8 9 21
119 43 29 191
(34%) (43%) (45%) (37%)
94 63 20 177
(22%) (48%) (50%) (30%)
0 (0%) 0 (0%) 1 (7%) 1 (1%)
0 2 2 4
(0%) (12%) (8%) (5%)
59.0* 26.6* 19.6, 80.75
14 (4%) 7 (7%) 11 (17%) 32 (6%)
27 8 2 37
(6%) (6%) (5%) (6%)
2 (3%) 2 (13%) 1 (7%) 5 (5%)
6 2 3 11
(17%) (12%) (13%) (14%)
65.65 45.75 10.7 125.5,
(9%) (20%) (45%) (20%)
16 (35%) 18 (44%) 7 (35%) 41 (38%)
30.4* 8.1 25.6* 68.2*
*Numbers and percentages of articles and programs on each disease in which conventional medications, surgery, and alternativ~ treatments are mentioned. 1-×2with respect to differences between diseases,considering the mentioning of conventional medications, surgery, and alternative treatments in articles and programs. ~P<.OOl.
and chronic lung diseases, and least attention in heart disease and cancer. Surgery was reported the most in heart disease and cancer, regardless of medium, and in television programs about rheumatic disease. Alternative treatments were reported most often, in all media, in relation to rheumatic diseases. Standard medications and surgery mostly were described in a neutral way, whereas mainly positive aspects of alternative treatments were highlighted (Table 4). Conventional medications for heart disease were described more often in a positive way than for the other diseases, and standard medications for rheumatic diseases were described more often in a negative way compared with other diseases. For diabetes, conventional medications frequently were described in a neutral way. Cancer surgery was often described negatively compared with other diseases. No significant differences were found in the way alternative treatments for the five disease categories were described in articles and programs. The proportion of articles mentioning influence of lifestyle was highest for newspaper articles on rheumatic diseases and diabetes (Table 5). Influ-
ence of lifestyle was covered least often in articles and programs about cancer and heart disease. There were no significant differences in the coverage of scientific progress among the five disease categories (Table 6). Incurability coverage was highest for diabetes and chronic lung disease (Table 7). The proportion of magazine articles covering incurability was greatest for cancer and lowest for heart disease. Much media attention was given to fatality in cancer and heart disease (Table 7). A high proportion of magazine articles reported fatality in cancer patients. Diabetes occasionally was mentioned as a cause of death in newspapers, but not in magazines or television programs. Rheumatic disease was reported as a cause of death only once. DISCUSSION The amount of attention given to a disease or disease category by the media depends on its fatality rate, not its prevalence. Both in our study and in an American study of chronic diseases presented on national television news (14,15), the diseases receiving the most attention were those with the highest death rates, heart disease and
CHRONIC DISEASES iN THE NETHERLANDS
339
Table 4: Positive, Neutral or Negative Judgments of Conventional Medications, Surgery, and Alternative Treatments in Newspaper and Magazine Articles, and on Medical Television Programs for Five Chronic Diseases Rheumatic Disease
Heart Disease
Cancer
Chronic Lung Disease
Diabetes
Regular medications* Positive Neutral Negative Total x2(df = 2)t
8 (19%) 22 (47%) 16 (34%) 46 (100%) 11.5~
56 (44%) 54 (43%) 16 (13%) 126 (100%) 17.4§
30 (22%) 75 (55%) 31 (23%) 136 (100%) 9.1
13 (33%) 23 (57%) 4 (10%) 40 (100%) 1.1
12 (26%) 34 (74%) 0 (0%) 46 (100%) 13.6$
Surgery* Positive Neutral Negative Total x2(df = 2)t
8 (38%) 12 (57%) 1 (5%) 21 (100%) 3,4
54 (28%) 111 (58%) 26 (14%) 191 (100%) 9.0
48 (27%) 79 (45%) 50 (28%) 177 (100%) 16,2~
0 (0%) 1 (100%) 0 (0%) 1 (100%) 1.0
1 (25%) 3 (75%) 0 (0%) 4 (100%) 1.2
Alternative treatments* Positive Neutral Negative Total ×2(dr = 2)1"
25 (62%) 9 (23%) 6 (15%) 40 (100%) 1.6
25 (78%) 5 (16%) 2 (6%) 32 (100%) 1,8
27 (73%) 8 (22%) 2 (5%) 37 (100%) 1.3
5 (100%) 0 (0%) 0 (0%) 5 (100%) 2.3
4 (36%) 4 (36%) 3 (28%) 11 (100%) 6.8
*Numbers and percentages of articles and programs on each disease in which conventional medications, surgery, and alternative treatments are judged positive, neutral, or negative. I-X2with respect to differences in judgment of treatments for each disease and judgment of treatments for the other four diseases. tP < .01. §P< .001.
cancer. Neither rheumatic diseases nor chronic lung disease were featured nearly as often, yet both are more prevalent and responsible for more work disability. No correlation between the frequency of coverage and mortality statistics was found in a study of newspaper coverage of 27 disorders (18). However, the latter included many rare diseases
that seldom cause death, such as rabies, Down syndrome, and food poisoning. Cancer and heart disease are serious conditions because they are the biggest killers, are frequently disabling, and are costly to the health care system. Our study showed that these diseases were covered more frequently than and differently from the more
Table 5: Influence of Lifestyle in Newspaper and Magazine Articles, and on Medical Television Programs for Five Chronic Diseases Rheumatic Disease Influence of lifestyle* Newspapers Magazines TV programs Total
23 (50%) 15 (37%) 7 (35%) 45 (42%)
Heart Disease 111 23 31 165
(32%) (23%) (48%) (32%)
Cancer 115 (27%) 17 (!3%) 14 (35%) 146 (25%)
Chronic Lung Disease 22 4 11 37
(33%) (27%) (73%) (38%)
Diabetes
xZ(df= 4)t
17 (47%) 5 {29%) 12 (50%) 34 (44%)
14.75 12,5 7.7 26,0§
*Numbers and percentages of articles and programs on each disease in which influence of lifestyle is mentioned, t× 2With respect to differences between diseases, considering the influence of lifestyle in articles and programs. SP< .01. §P< .001.
340
VAN DER WARDT ET AL
Table 6: Scientific Progress in Newspaper and Magazine Articles, and on Medical Television Programs for Five Chronic Diseases Chronic Lung
Rheumatic Disease
Scientific Progress* Newspapers Magazines TV programs Total
15 5 3 23
(33%) (12%) (15%) (21%)
Heart Disease
138 15 15 168
(40%) (15%) (23%) (33%)
Cancer
183 14 8 205
(44%) (11%) (20%) (35%)
Disease
Diabetes
37 2 3 42
18 5 8 31
(55%) (13%) (20%) (43%)
x2(df =
(50%) (29%) (33%) (40%)
4)t
8.6 4.9 2.5 13.0
*Numbers and percentages of articles and programs on each disease in which scientific progress is mentioned. tx 2With respect to differences between diseases, considering scientific progress in articles and programs.
ences in media attention they received may suggest that heart disease and cancer are more common than rheumatic or chronic lung diseases. Experts were more often featured and fatality more often mentioned in the coverage of heart disease and cancer, which may add to the impression of their severity. Reports on the influence of lifestyle and alternative treatments in rheumatic diseases may lead to the misleading impression that they can be easily avoided, relieved, or even cured by patients themselves. Many important questions remain: What can professionals in medical science and journalism do to ensure responsible media coverage of medical issues? Who is to blame for public confusion about medical science: the media, medical scientists, or an uncritical public? To what extent does press coverage of medicine influence politicians? How important is the role of the lay press in the
lingering, chronic disorders. The proportion of articles and programs on heart disease and cancer featuring professionals was striking. The specific subjects dealt with often included surgery as well as fatality. Rheumatic and chronic lung disease and diabetes coverage were more likely to feature standard medications and the influence of lifestyle. The large amount of attention on lifestyle in relation to rheumatic diseases is particularly surprising. One might expect to see more attention given to the influence of lifestyle on cancer (eg, smoking habits, sunlight exposure) or heart disease (smoking, diet, exercise), which are widely accepted associations. Many articles and programs on rheumatic diseases featured patients and treatments involving alternative medicine. If the public's notions of the five diseases are similarly influenced by media coverage, several misconceptions are likely to prevail. The differ-
Table 7: Incurability and Fatality in Newspaper and Magazine Articles, and on Medical Television Programs for Five Chronic Diseases Rheumatic Disease
Incurability* Newspapers Magazines TV-programs Total Fatality* Newspapers Magazines mV-programs Total
9 3 3 15
(20%) (7%) (15%) (14%)
0 (0%) 1 (2%) 0 (0%) 1 (1%)
Heart Disease
36 (10%) 2 (2%) 1 (2%) 39 (8%) 147 33 20 200
(42%) (33%) (31%) (39%)
Cancer
Chronic Lung Disease
59 22 5 86
(14%) (17%) (13%) (15%)
15 2 3 20
(22%) (13%) (20%) (21%)
159 79 11 249
(38%) (60%) (28%) (42%)
11 2 3 16
(16%) (13%) (20%) (16%)
Diabetes
x2(df
8 (22%) 1 (6%) 5 (21%) 14 (18%) 6 0 0 6
(17%) (0%) (0%) (8%)
*Numbers and percentages of articles and programs on each disease in which incurability and fatality are mentioned. tx 2With respect to differences between diseases, considering the incurability and fatality in articles and programs. ~:P< .01. §P< .OOl.
=
4)t
11.1 14.55 10,3 21.7§ 49.5§ 64.1§ 16.95 112.0§
CHRONIC DISEASES IN THE NETHERLANDS
341
transmission of medical knowledge not only to the public but also to the scientific community (27)? Our findings indicate the need for more frequent and more accurate media coverage of chronic lung disease, diabetes mellitus, and especially rheumatic disorders. Foundations, doctors, and scientists concerned with rheumatic diseases should take more advantage of the opportunities for public education
provided by the mass media. More attention from the media could lead to greater public awareness of rheumatic diseases and, in turn, contribute to a better understanding of people suffering from them. ACKNOWLEDGMENT
The authors thank W. Burke and J:M. Cornwell for their help with English.
REFERENCES 1. Atkin C, Arkin EB. Issues and initiatives in communicating health information to the public. In: Atkin C, Wallack L, editors. Mass communication and public health. Newbury Park, CA: Sage, 1990:13-40. 2. Signorielli N. Mass media images and impact on health. Westport, CT: Greenwood, 1993. 3. Wade S, Schramm W. The mass media as sources ofpublic affairs, science, and health knowledge. Publ Opin Q 1969;33:197209. 4. Gerbner G, Gross L, Morgan M, Signorielli N. Special report: health and medicine on television. N Engl J Med 1981;305:901-4. 5. McCombs ME, Shaw DL. The agenda-setting function of mass media. Publ Opin Q 1972;36:176-87. 6. Wallack L. Improving health promotion: Media advocacy and social marketing approaches. In: Atkin C, Wallack L, editors. Mass communication and public health. Newbury Park, CA: Sage, 1990:147-63. 7. Bandura A: Social foundations of thought and action: A social cognitive theory. Englewood Cliffs, NJ: Prentice-Hall, 1986. 8. Gerbner G, Gross L. Living with television: The violence profile. J Commun 1976;26:t73-99. 9. Philo G, Secker J, Platt: S, Henderson L, McLaughfin G, Burnside J. The impact of the mass media on public images of mental illness: Media:content and audience belief. Health Educ J 1994;53:271-81. 10. Price JH, Hillman KS, Toral ME, Newell S. The public's perceptions and misperceptions of arthritis. Arthritis Rheum 1983;26:1023-& 11. Banman A, Antic R, Rubinfeld A, Zimmerman R Gutch D, Bacon B, et al. "Could it be astma?": The impact of a mass media campaign aimed at raising awareness about astma in Australia. Health Educ Res 1993;8:581-7. 12. Combs B, Slovic R Newspaper coverage of causes of death. Journalism Q 1979;56:837-43. 13. Ettema JJ, Whitney DC: Audiencemaking: how the
media create the audience. Thousand Oaks. CA: SAGE Publications. 1994. 14. Pichert JW. Diabetes in the national TV-news: 19711981. Diabetes Care 1983:6:92-4. 15. Pichert JW, Hanson SL. Arthritis in the national TVnews: 1971-1981. J Rheum 1983:10:323-5. 16. Garland R. Images of health and medical science conveyed by television. J R Coll Gen Pract 1984:34:316-9. 17. Johnson CA, Johnson BE. Medicine on British television: A content analysis. J Commun Health 1993:18:25-35. 18. Kristiansen CM. Newspaper coverage of diseases and actual mortality statistics. Eur J Soc Psychnl 1983:13:193-4. 19. Freimuth VS. Greenberg RH, DeWitt J, Romano ILM. Coveting cancer: Newspapers and the public interest. J Commun 1984;34:62-73. 20. Greenberg M, Wartenberg D. Newspaper coverage of cancer clusters. Health Educ Q 1991; 18:363-74. 21. Miner KJ, Baker JA. Media coverage of suntanning and skin cancer: Mixed messages of health and bean~y. J Health Educ 1994:25:234-8. 22. ReynoldsT. Media coverage of cancer risks may mislead. J Nad Cancer Institute 1993;85:1366-8. 23. Krippendorff K. Content analysis: An introduction to its methodology. Newbury Park. CA: Sage, 1980. 24. Commissie voor Keuzen in de Zorg. Kiezen en delen: Achtergrondstndies. [Committee for Choices in Care. Choose and divide: Background studies.] Den Hang: Albani, 1991. 25. Raad voor Gezondheidsonderzoek. Advies chronische aandoeningen: Prioriteiten voor onderzoek. [Board of Health Research: Advice on chronic diseases: Priorities for research.] Den Hang: Stantsuitgeverij, 1991. 26. Reynolds HT. The analysis of cross-classifications. New York: Free Press. 1977. 27. Phillips DP, Kanter EJ. Bednarczyk B. Tastad PL. Importance of the lay press in the transmission of medical knowledge to the scientific community. N Engl J Med 1991:325: 1180-3.