Content-Type: text/html Accent on the POZitive? A Content Analysis of Biomedical AIDS Stories in POZ Magazine by William P. Cassidy Department of Communication University of Wisconsin-Whitewater 600 W. Blackhawk Dr. #1 Fort Atkinson, WI 53538 (262) 472-5022 [log in to unmask] Submitted to the Science Communication Interest Group of the Association for Education in Journalism and Mass Communication Accent on the POZitive? A Content Analysis of Biomedical AIDS Stories in POZ Magazine (1996-2000) Abstract This study examines biomedical AIDS stories published in POZ, a magazine targeted at HIV-positive men and women in the U.S., during 1996-2000. A monthly stratified sample of 12 issues were content analyzed. Results show that the overall tone of coverage was .94 on a scale of 0 to 2 (0=negative, 2=positive) and that medical or scientific (non-governmental) sources, unaffiliated HIV-positive sources and AIDS activist sources figured most prominently in coverage. A short report written by Dr. Michael Gottleib and colleagues in June 1981 appearing in the federal Centers for Disease Control and Prevention's (CDC) weekly newsletter noted that a rare parasitic lung infection, pneumocystis carinii pneumonia (PCP), had struck "five gay men, all active homosexuals" in Los Angeles (Begley, 2001). Almost immediately after the report's publication, Gottleib began receiving reports of similar cases in San Francisco and New York (Brown, 2001). The report signaled the beginning of the AIDS epidemic. Since then more than 25 million people worldwide have died from AIDS or AIDS-related complications (Eberstadt, 2002). As of December 2002, 42 million people were estimated to have AIDS and 5 million people were newly infected with HIV, the virus that causes AIDS during 2002 (UNAIDS, 2002). The CDC reports 816,149 AIDS cases in the U. S. since the beginning of the epidemic, and 467,910 deaths (CDC, 2002). Despite the above and the fact that over 1 million Americans are HIV-positive, there is a growing perception in the U. S. that AIDS is no longer a cause for serious concern and has become a manageable chronic illness (Ackerman, 2001a). A survey by the Henry J. Kaiser Family Foundation found that the proportion of people who view AIDS as the leading health problem in the nation has declined from 44% in 1995 to 26% in 2000 (Staples, 2001). Futhermore, the number of AIDS deaths in the U. S. has declined each year since 1996 (CDC, 2002). According to Laurie Garrett, Pulitzer Prize-winning author of The Coming Plague and a Newsday reporter, the media play a role in this complacency. "The slowing down of AIDS reporting in the United States and abroad," she says, "has eliminated all sense of urgency in the public, in politicians, and in communities" (Ng, 1999). Another major complaint regarding the media is that since the 1996 introduction of new drug therapies, which have played a significant part in lowering death rates in the United States, coverage has been overly optimistic (Cohen, 1997). "Stories focus on the drop in deaths and ignore the steady increase in infections," says Mario Cooper, founder of Leading for Live, a national campaign launched by the Harvard AIDS Institute to raise HIV awareness among African-American and Latin leaders. "The media treat AIDS like they would a sports game, proclaiming victory and making people think there's been a cure when nothing could be further from the truth" (Ng, 1999). That complaint is the focus of this paper, which analyzes AIDS coverage in POZ, a magazine targeted at the HIV-positive men and women in the U.S., as well as their friends and families. This paper also seeks to assess the tone of coverage in the years following the introduction of new drug therapies. Experts have said that because, in part, of the tone of media coverage, increases in AIDS-related deaths are likely to occur in the U.S. (Ng, 1999). There is already some evidence this is happening. The CDC reported that infection rates among gay men in San Francisco, a bellwether city for AIDS trends, have significantly risen ("AIDS at 20," 2001). Literature Review Media Coverage of AIDS The mainstream media were slow to publish reports about AIDS, even as the number of cases grew. The first newspaper article, which appeared in July 1981 in the New York Times, focused on eight deaths from PCP that had occurred in a two-year period, Initially, the syndrome was termed GRID for "gay-related immunodeficiency." At that time the syndrome was thought to be confined to gay men (Ng, 1999). Many have attributed the early lag in coverage to reluctance by the media to publish stories about homosexuals (Altman, 1986; Baker, 1986; Nelkin, 1991; Shilts, 1987). As evidenced by the initial name coined for AIDS, it was perceived by many in the media and among the public as a "gay story." Coverage expanded in 1983 following publication of a news release based on an editorial in the Journal of the American Medical Association (JAMA). Dr. Anthony Fauci, director of the National Institute of Allergy and Infectious Diseases, and author of the editorial, suggested that routine household contact might spread AIDS. Thus, the editorial implied that AIDS could infect the general population. With society at large now deemed at potential risk, the news value of AIDS increased (Rogers, Dearing & Chang, 1991). A dramatic increase in coverage occurred in 1985 after actor Rock Hudson was diagnosed with AIDS. Hudson's plight and that of Ryan White, a 13-year-old hemophiliac with AIDS who was barred from attending public school the same year, personalized and humanized the epidemic and was critical in changing journalists' perceptions of the epidemic (Rogers, Dearing & Chang, 1991). Although AIDS was now considered important news, Shilts (1987) wrote that one aspect of the epidemic was ignored by the media: the federal government's role in combating the virus. At this time, for example, the presidential administration of Ronald Reagan had not launched a coordinated AIDS prevention program. Reagan also did not discuss AIDS during the first six years of his presidency (Perez & Dionisopoulos, 1995). But, according to Shilts (1987) and Perez and Dionisopoulos (1995) media coverage changed substantially with the October 1986 publication of the Surgeon General's Report on Acquired Immune Deficiency Syndrome. Written personally by Surgeon General Dr. C. Everett Koop, the 36-page report called for AIDS education for children, advocated widespread condom use and stated that any form of quarantine, or compulsory identification of those infected with HIV would be useless in fighting the epidemic (Shilts, 1987). The report, according to Shilts (1987), "galvanized the media and allowed AIDS to achieve the critical mass to make it a pivotal social issue of 1987" (p. 587). Perez and Dionisopoulos (1995) assert the report greatly influenced the "socio-political environment in which the AIDS crisis was discussed" (p. 30). In a content analysis of newspaper coverage before and after the Surgeon General's report, Cassidy (2000) found substantial increases in the number of stories, as well as the number of government, gay rights and AIDS activists and HIV-positive sources cited by journalists. Medical sources also made significantly more socio-political comments about AIDS following the report's publication. Other notable surges in coverage came when three-time National Basketball Association most valuable player Earvin "Magic" Johnson announced he was HIV-positive in November 1991 and after the death of tennis great Arthur Ashe from AIDS-related complications in Februrary 1993 (Kaiser Family Foundation, 1996). While most early studies of AIDS in the media were critical of the coverage (Altman, 1986; Baker, 1986; Kinsella, 1989; Nelkin, 1991; Shilts, 1987), more recent studies have documented a decrease in the number of stories reported. For example, Agoston (1996) in her examination of AIDS coverage in the New York Times, Washington Post and Los Angeles Times, found that the number of AIDS stories published decreased almost every year between 1987 (when the number of stories hit their peak) and 1994. The exception came in 1991, when there was an increase from 726 to 879. However, by 1994 the number of stories had dropped to 299. Kim's (1998) content analysis of AIDS coverage in Time and Newsweek reported similar results. Coverage peaked in 1987, dropped significantly the following year, then remained steady until 1992 when the number or stories increased to 253. The numbers declined substantially in 1993 and continued to do so through 1995. However, the number of stories increased again in 1996. While there is a fairly rich stream of literature addressing mainstream media coverage, little attention has been given to how the gay and lesbian press reports on AIDS. Streitmatter (1995) gave mixed reviews to the gay and lesbian press regarding coverage during the early stages of the epidemic. He said the New York Native, a biweekly with minimal resources, provided the most complete and thorough coverage and on several occasions "scooped" the New York Times. Other publications receiving praise from Streitmatter (1995) were the Washington Blade and Gay Community News. On the other hand, Streitmatter (1995) said that two San Francisco publications, the Bay Area Reporter and the San Francisco Sentinel, failed to adequately cover AIDS because they feared losing advertising revenues from bathhouses, even though these establishments were described by the medical community as "death camps" (p. 254). The Advocate, often referred to as the "Time or Newsweek of the gay community," with a circulation triple that of any other gay publication, was also criticized by Streitmatter (1995). He said the magazine initially ignored the epidemic even though its associate editor was one of the first AIDS fatalities. Instead, The Advocate, Streitmatter (1995) said, focused on gay culture and politics. There have been no empirically-based studies addressing how the gay and lesbian press covers AIDS. This paper seeks to remedy this oversight by examining how POZ reports on the epidemic. History of POZ From March 1994 when the first issue appeared on newsstands, (then) publisher Sean Strub had definite goals for the magazine. "We want to be entertaining, useful and challenging to my parents in Iowa or to my siblings who want to know how to handle issues," he said. "We want POZ to have the readability of People, the literacy of The New Yorker and the cutting-edge information of the New England Journal of Medicine" (Hochwald, 1994, p. 35). The enormous impact of AIDS on the gay community vaulted gay and lesbian issues into the public eye. But, this was a steep price to pay of visibility. According to Streitmatter (1995) "For many gay men…far too much of life had become devoted to visiting sick friends, attending memorial services and reading about the latest research dead end" (p. 319). POZ was an effort to bring a more positive tone to coverage of the epidemic. "If we believe ourselves to be terminally ill, we void everything about our lives," said Strub, who has been HIV-positive for over 20 years and sold his own life insurance to cover start-up costs for the magazine. "Our love, passion, vision and vitality. Our hopes and dreams. We might as well just plan the funeral and wait to die. But, if we are survivors we have a future. Something to strive for. Places to go. People to meet. Things to do. Love to share" (Strub, 1994, p. 60). Despite the altruistic sentiment of such statements, Hicks (1997) said other factors, such as a stable advertiser base contributed to the start of the magazine. Tragically, he noted that the number of those affected by AIDS also had to grow large enough to sustain a national-circulation magazine. Streitmatter (1995) said profiles of HIV-positive men and women who remained active and productive quickly became the magazine's trademark. Other sections of POZ include new treatment options and research developments, and columns on living with HIV and AIDS. During the early years the magazine also regularly reported on Strub's health, complete with the results of his latest blood work and analysis by two doctors, who sometimes disagreed on treatment. However, POZ is not without its critics. Burkett (1995) said that while the articles in the magazine offer hope to those infected, to the uninfected the message is that an HIV-positive diagnosis isn't a huge problem. But, the most vehement attacks on the magazine have come because of its advertisers. Although many "mainstream" companies advertise in POZ, some of its biggest advertisers are companies known as viaticals, which offer AIDS sufferers cash for their life insurance policies (Streitmatter, 1995). Critics said it was unnerving that while the editorial content of the magazine offered optimistic messages, some of the major advertisers "trafficked in death futures" (Burkett, 1995, p. 139). POZ began as a bimonthly and first turned a profit in 1997 (Kerr, 1998). The magazine, with a circulation of 100,000, has undergone some tough times in recent years. Advertising revenue has dropped 50% since 1999. The magazine let go of one-third of its staff in 2001 and in that year also began reducing the number of pages in each edition (Smith, 2003). However, POZ, which is currently published 10 times a year, remains free for anyone who is HIV-positive and can't pay the subscription fee ("About POZ", 2003). News Sources One way of assessing AIDS coverage in POZ is by studying the sources used in the magazine's stories about AIDS. The sources journalists use are important to study because they determine not only what information is given to the public, but what image of society is presented (Soloski, 1989). Gans (1979) wrote that the sources used by journalists reflect society's desire for moral order. Reese, Grant and Danielson (1994) believe reliance on these elite sources restricts the views expressed and implicitly defines which opinions and individuals are important. It has been well established that government officials are the sources used most frequently in news reports. For example, Sigal (1973) in his content analysis of front page stories in the New York Times and Washington Post found that 81.3% of the sources were government officials. A similar study of regional and local newspapers reported that 55.4% of sources were affiliated with some governmental body (Brown, Bybee, Wearden & Straughan, 1987). A study of the sourcing patterns of national security reporters found that U. S. government sources represented 73.3% of the citations in these types of stories (Hallin, Manoff & Webble, 1993). Bennett (1990) said that journalists base the range of views expressed in their stories on the range of views expressed in mainstream government debate about the topic of the story. Non-officials are included in stories and editorials only when they express opinions already emerging in public circles. Alexseev and Bennett (1995) reported that sustained governmental conflict opened the "news gates" to broader political statements from non-governmental social groups and organized interests. Regarding sources in AIDS coverage, Colby and Cook (1991) said that as the news value of AIDS increased, the epidemic was covered by multiple reporters from multiple beats which allowed for a greater variety of sources to be heard. However, Hallett and Canella (1994) found that the Arizona Republic newspaper cited HIV-positive sources in only 20% of their AIDS reports from 1986 through 1990. Cassidy (2000), as mentioned earlier, found increases in the number of attributed comments by HIV-positive sources in his study, but the percentage of comments attributed to these sources rose to only 2.6%. A later study by Cassidy (2002) found that HIV-positive public figures were the second most cited source type in stories about the epidemic in 1995 and 1996 editions of Newsweek, Time, and U. S. News & World Report. Research Questions This study seeks to analyze POZ's AIDS coverage by studying the sources used in biomedical stories about the epidemic. Based on the above, the following research question is posed: RQ1: Which types of sources are most prominent in biomedical stories about the AIDS epidemic in POZ? This study is also interested in the tone of biomedical stories about AIDS in POZ, especially since the magazine tries to bring a more positive tone to its AIDS coverage (Burkett, 1995; Streitmatter, 1995; Strub, 1994). The tone of coverage following the 1996 introduction of promising new drug therapies known as protease inhibitors, is of particular interest. These therapies have dramatically lowered AIDS death rates and given many AIDS patients a higher quality of life (CDC, 2000). The mainstream media have come under criticism for carelessly hyping these advances. For example a 1997 Science article by Jon Cohen said: A casual browser of recent newsstands might conclude that AIDS is now all but cured. Last fall, the New York Times Magazine featured a cover story headlined "When AIDS Ends." Newsweek's cover wondered about "The End of AIDS?" and a Time cover toasted AIDS researcher David Ho as its "Man of the Year." As these magazines attest, AIDS research had a banner year in 1996, making dramatic strides with potent new drug combinations…But, as delighted as AIDS researchers are about the progress, a growing number are concerned that many popular media stories cross the line that separates hope from hype (p. 298). In addition, previous research on mainstream media and AIDS (Cassidy, 2000; Kaiser Family Foundation, 1996; Rogers Dearing & Chang, 1991) has shown that specific events can have a strong impact on coverage. However, Cassidy (2002) found in his analysis of the three major newsmagazines that coverage actually became slightly more negative in 1996 following the introduction of protease inhibitors. Therefore, an additional research question is advanced: RQ2: What is the overall tone of biomedical stories about the AIDS epidemic published in POZ after the introduction of protease inhibitors? Methodology Content analysis was used to answer the research questions. The time period analyzed was January 1996 through December 2000. Protease inhibitors were approved by the Food and Drug Administration in December 1995 and became readily available in early 1996. While acknowledging that changes in coverage may not be immediate, the initial time period of this study seems the logical starting point, as stories about trial studies of protease inhibitors began appearing in 1995. Thus, the media was aware of the new drugs before FDA approval and began covering their effectiveness shortly thereafter. The staff and editorial content reductions at POZ are the reasons for not including issues published in 2001 and 2002. Since ostensibly nearly all stories in POZ are, at least indirectly, about AIDS, and given that a major interest of this study is coverage after the introduction of protease inhibitors, only biomedical AIDS stories were studied. The definition of biomedical stories, as suggested by Cassidy (2000) is: stories regarding scientific information and research findings about AIDS, including its symptoms and the opportunistic infections associated with AIDS. Also in this category are stories regarding the medical treatments physicians give to their patients who are HIV-positive or have AIDS. A monthly stratified sample of 12 issues of POZ were analyzed. Lacy, Riffe and Randle (1998) in a study of consumer magazines, found that when sampling from a five-year period (like this study), a constructed year of 12 issues, with one issue selected randomly from among the issues in each month, was a more efficient and accurate sample than 20 randomly selected issues. While noting that articles about specific topics tended to vary more than the other variables in their study, they reported that the minimal efficiency standard (68% of sampled means within one standard error of population mean and 95% within two standard errors of population mean) was still reached on this variable with the 12 issue monthly stratified sample. The unit of analysis was the attributed comment (direct and indirect quotes). The types of sources, taken from Cassidy (2000) were coded as: (1) government (non-medical/appointed officials, (2) government (elected) officials, (3) government (medical or scientific/appointed), (4) government (unspecified), (5) medical and/or scientific (non-governmental), (6) medical and/or scientific (unspecified), (7) public figures (HIV-positive), (8) public figures (HIV-negative), (9) religious, (10) gay rights and AIDS activists, (11) legal, (12) unaffiliated HIV-positive, (13) unaffiliated HIV-negative, (14) corporate and (15) other. Definitions of the categories are available from the author by request. For RQ1, prominence was operationalized as how often a source type was cited making attributed comments and where in a story a source was first cited. Placing a source higher in a story gives it greater importance, while the frequency of a source being cited is an indication of the sources overall newsworthiness (Hallin, Manoff and Weddle (1993); Massey, 1998). Sources were coded once per paragraph and the stories were divided into blocks of three paragraphs, with the tenth graph through the conclusion of a story serving as the fourth block. This is a slight variation of methods used by Hallin, Manoff and Weddle (1993) and Massey (1998), who divided stories into blocks of five paragraphs. The reason for the change in this study is because Cassidy's (2002) analysis of articles about AIDS in newsmagazines found that the mean story length was less than eight paragraphs. Therefore dividing the stories into "shorter" blocks seems more likely to yield richer information on prominence here. To determine the tone of coverage, methodologies employed by Cassidy (2002), guided by the earlier studies of the Kaiser Family Foundation (1996) and Kim (1998) were utilized. The Kaiser Family Foundation (1996) surveyed AIDS print and broadcast coverage from 1985 to 1996 and found that during a typical week 48% of stories were optimistic and 23% pessimistic, with the remaining 29% being neutral/ambiguous. Kim (1998) said coverage became more negative was time went on. During the first time period of his study (1982-1984) 31% of stories were deemed positive, with 15% negative. However, in the final time period (1994-1996) only 11% of the stories were positive, while 29% were negative. For this study, attributed comments (direct and indirect quotes) were also the unit of analysis in determining the tone of a story. Each attributed comment made in a story was evaluated as to whether its tone was positive, ambiguous or negative. The definitions of these categories are: Positive: Comments regarding successful aspects of AIDS treatment and research, including comments noting improvements in the quality of life for those people with AIDS. Also included in this category are comments regarding any declines in AIDS death rates. Ambiguous: Comments that are neither positive or negative, or are mixed (a combination of positive and negative). Negative: Comments regarding problems and setbacks in AIDS treatment and research, as well as those noting debilitating effects of the disease. Also included in this category are comments regarding any increases in AIDS death rates. Each attributed comment in each AIDS story was coded into one of the above three categories and given a score of 0 (negative), 1 (ambiguous) or 2 (positive). The number of attributed comments and scores in each category were then tallied for each story and each story was given an attributed comment score. This part of the coding procedure is similar to the one used by Waldman and Devitt (1998) in their evaluation of newspaper photographs of presidential candidates in the 1996 election. Each story was also assessed for overall tone and given a score of 0 (negative), 1 (ambiguous) or 2 (positive). The two scores were then combined to give each story a final score. Examining both the "nuts and bolts" (attributed comments by sources) as well as the overall tone of stories seems a more rigorous procedure than that employed by other studies. The author served as the primary coder of this study. To determine intercoder reliability, approximately 20% of the articles were coded by a second trained coder. Using Scott's pi, the level of agreement was .92 for source affiliation, .90 for attributed comment tone, .93 for story type (whether or not a story was a biomedical AIDS story) and 1.00 for overall story tone. Results A total of 100 stories were analyzed (N=100) in the 12 editions of POZ sampled (M=8.33 biomedical AIDS stories per issue). The stories yielded 467 attributed comments (M=4.67) (see Table 1) made by 258 different sources (M=2.58). RQ1 asks which types of sources are most prominent in biomedical stories about the AIDS epidemic in POZ? Table 1 shows that three source types were most prominent in coverage: medical and/or scientific (non-governmental) sources; unaffiliated HIV-positive sources, and; gay rights and AIDS activists. Together these three source types accounted for nearly 82% of the attributed comments made in the stories. Medical and/or scientific (non-governmental) sources were the leading source type cited with 199 attributed comments (M=1.99). Unaffiliated HIV-positive sources were next, with 118 attributed comments (M=1.18), followed by gay rights and AIDS activists with 64 attributed comments (M=.64). Table 2 shows that of the three leading source types, attributed comments by medical and/or scientific (non-governmental sources) tended to be introduced earlier in a story than were unaffiliated HIV-positive and gay rights and AIDS activist sources. Forty-six percent of the initial attributed comments by medical and/or scientific (non-governmental) sources came in the first three paragraphs of the story. Unaffiliated HIV-positive sources were introduced in the first three paragraphs 39% of the time, followed by gay rights and AIDS activists at 35%. Unaffiliated HIV-positive sources and gay rights and AIDS activists were more likely to be introduced in the second block of paragraphs (25% of both types) than were medical and/or scientific (non-governmental) sources (21%). Gay rights and AIDS activists were more likely to be introduced in the fourth block of paragraphs (35%) than were unaffiliated HIV-positive sources (27%) and medical and/or scientific (non-governmental) sources (23%). The stories analyzed in the sample tended to be either quite short or very lengthy. This, in part, accounts for the relatively high percentages of initial attributed comments by these three source types coming in the fourth block of paragraphs. RQ2 asks what is the overall tone of biomedical stories about AIDS in POZ? Table 3 shows that the mean tone score of the 100 stories examined was .94 on a scale of 0 (negative) to 2 (positive) Discussion This study examined biomedical stories about the AIDS epidemic in POZ published after the introduction of successful new drug therapies, known as protease inhibitors. Therefore, it is not surprising that medical and scientific sources were most prominent in coverage. These results are similar to Cassidy's (2002) analysis of newsmagazine AIDS coverage during 1995-1996. However, another study looking at newspaper coverage of AIDS during an earlier critical time period of the epidemic (1986-1987) found that government (medical or scientific/appointed) sources were cited more than any other source type (Cassidy, 2000). It seems that by the mid-1990s journalists covering AIDS had a more thorough understanding of the epidemic and realized which non-official sources were legitimate and did not feel compelled to rely on government (medical or scientific/appointed) sources. Unaffiliated HIV-positive sources also figured very prominently in coverage in this study. This could be for two reasons. The first being that POZ is targeted at HIV-positive individuals in the U.S. Therefore, it seems logical that the voices of HIV-positive individuals would have a prominent role. The second is that by the mid-90s the stigma associated with being HIV-positive had decreased, making it easier for HIV-positive sources to speak publicly. Interestingly, mainstream news magazines essentially ignored unaffiliated HIV-positive sources until after the 1996 introduction of protease inhibitor drugs (Cassidy, 2002), but these sources did not have nearly as prominent a role in coverage as they do in POZ Another notable difference in coverage here is the prominent voice given to activists in POZ. According to Epstein (1996) AIDS activists played a major role in developing the treatments for AIDS and also pressured the medical establishment to use more pragmatic criteria in their drug trials, as well as more diverse subject populations. This study suggests, that as a result of these efforts, these sources gained "a place at the table" in biomedical AIDS coverage, especially considering the minor role such sources played in coverage during the early stages of the epidemic (Cassidy, 2001). Perhaps the most interesting results of this study center on the tone of coverage in POZ. Many have complained that coverage became overly optimistic or "too positive" following the introduction of protease inhibitors (e.g. Cohen, 1997; Ng, 1999). However, the results here show that coverage during this time in POZ could hardly be deemed "too positive." This is even more surprising since it's inception, a major goal of the magazine has been to bring a more positive tone to coverage (Streimatter, 1995; Strub, 1994). In fact, the tone of coverage here (.94 on a scale of 0 to 2), was more negative than that in the mainstream newsmagazines (1.17 on the same scale) examined by Cassidy (2002). One of the reasons that coverage was found to be far from overly optimistic could be due to the coding scheme of this study. Cohen (1997) in his criticism of AIDS coverage after the introduction of protease inhibitors notes that "casual browser[s]…might think that AIDS is now all but cured." However, this study called for coders to closely analyze the attributed comments of sources, as well as the overall tone of the text. The coverage was not "casually browsed" over. Cohen (1997) also highlights cover stories and headlines signaling the end of the AIDS epidemic and uses them as evidence that coverage may have descended into hype. Variables such as headlines and where in the magazine stories appeared were not used in this analysis. But, perhaps they should have been. It is possible that only after thorough, rigorous examination of the text of AIDS stories does one realize that coverage is far from overly optimistic or positive. However, many readers examine the text more causally, and instead, pay more attention to things such as headlines, covers and photographs. So, perhaps future studies should include these variables in analyzing coverage of the AIDS epidemic. It would also be interesting to conduct a study comparing coverage during the same time frame between mainstream media outlets and publications such as POZ, and/or publications aimed at the gay community, such as The Advocate. References About POZ (2003). POZ. Available: http://www.poz.com/about/index.html Ackerman, T. (2001a, May 31). Blacks lead HIV infection rate among young gay men. Houston Chronicle, p. A1. Ackerman, T. (2001b, June 2). AIDS in America turns 20. Houston Chronicle, p. A1. AIDS at 20. (2001, June 5). USA Today, p. 14A. Alexseev, M. & Bennett, W.L. (1995). For whom the gates open: News reporting and government source patters in the United States, Great Britain and Russia. Political Communication, 12, 395-412. Altman, D. (1986). AIDS in the mind of America. Garden City, NY: Anchor Press/Doubleday. Agoston, J. (1996, August). Coverage of AIDS between 1985 and 1994 in the New York Times, Washington Post and Los Angeles Times. Paper presented at the annual conference of the Association for Education in Journalism and Mass Communication, Anaheim, CA. Baker, A. (1986). The portrayal of AIDS in the media. In D. Feldman and T. Johnson (Eds.), The social dimensions of AIDS. (pp. 179-196). New York: Praeger. Begley, S. (2001, June 11). AIDS at 20. Newsweek, p. 34-37. Bennett, W. L. (1990). Toward a theory of press-state relations in the United States. Journal of Communication, 40(2), 103-125. Brown, D. (2001, June 5). Making medical history. Washington Post, p. CO1. Brown, J., Bybee, C., Wearden, S., & Straughan, D. (1987). Invisible power: Newspaper sources and the limits of diversity. Journalism Quarterly, 64(1), 45-54. Burkett, E. (1995). The gravest show on earth.Boston: Houghton Mifflin. Cassidy, W. P. (2000) The politicizing of AIDS. Newspaper Research Journal, 21(1), 26-40. Cassidy, W.P. (2001). The Surgeon General's Report on AIDS: A political opportunity for AIDS activists. In J. Biberman and A. Alkhafaji (Eds.), Business research yearbook: Global business perspectives. (pp. 708-712). Saline, MI: McNaughton & Gunn. Cassidy, W. P. (2002, August) Newsmagazine coverage of AIDS: Progress and hope or hype? Paper presented at the annual conference of the Association for Education in Journalism and Mass Communication, Miami Beach, FL. Centers for Disease Control and Prevention. (2000) HIV/AIDS surveillance report, 11(1). Atlanta: Author. Centers for Disease Control and Prevention. (2002). HIV/AIDS surveillance report, 13(2). Atlanta: Author. Cohen, J. (1997, January 17). The media's love affair with AIDS research: Hope vs. hype. Science, 298-299. Colby, D. & Cook, T. (1991). Epidemics and agendas: The politics of nightly news coverage of AIDS. Journal of Health Politics Policy and Law, 16(2), 215-249. Eberstadt, K. (2002, November/December). The future of AIDS. Foreign Affairs, p. 22. Epstein, S. (1996). Impure science: AIDS, activism and the politics of knowledge. Berkeley, CA: University of California Press. Gans, H. (1979). Deciding what's news. New York: Pantheon. Hallett, M., & Cannella, D. (1994). Gatekeeping through media format: Strategies of voice for the HIV-positive via human interest format and organizations. Journal of Homosexuality, 26(4), 111-134. Hallin, D., Manoff, R., & Weddle, J. (1993). Sourcing patterns of national security reporters. Journalism Quarterly, 70(4), 753-766. Hicks, G. (1997, August). The case of POZ magazine: Puttng uses and gratifications to the test. Paper presented at the annual conference of the Association for Education in Journalism and Mass Communication, Chicago. Hochwald, L. (1994, March 15). Accent on the POZitive. Folio, 35. In denial: Not taking AIDS seriously is potentially fatal error. (2001, June 7). Houston Chronicle, p. A36. Kaiser Family Foundation. (1996, July/August). Covering the epidemic: AIDS in the news media, 1985-1996. Columbia Journalism Review (Suppl.), 1-12. Kerr, S. (1998, March 23). Gloss action: POZ, a magazine for HIV-positive people. New York, 26-27. Kim, S. (1998, February). Media's restless search for spotlight: A content analysis of AIDS coverage in Time and Newsweek. Paper presented at the Southeast Regional Colloquium of the Association for Education in Journalism and Mass Communication, New Orleans. Kinsella, J. (1989). Covering the plague: AIDS and the American media. New Brunswick, NJ: Rutgers University Press. Lacy, S., Riffe, D., & Randle, Q. (1998). Sample size in multi-year content analyses of monthly consumer magazines. Journalism and Mass Communication Quarterly, 75(2), 408-417. Massey, B. L. (1998). Civic journalism and nonelite sourcing: Making routine newswork of community connectedness. Journalism and Mass Communication Quarterly, 75(2), 394-407. Nelkin, D. (1991). AIDS and the news media. The Millbank Quarterly, 69(2), 293-307. Ng, H. (1999, Fall). AIDS in the media. Harvard AIDS Review. Available: http://www.aids.harvard.edu/publications/har/fallwin_1999/fallwin99-6html. Perez, T., & Dionisopoulos, G. (1995). Presidential silence, C. Everett Koop and the Surgeon General's Report on AIDS. Communication Studies, 46(1), 18-33. Reese, S., Grant, A., & Danielian, L. (1994). The structure of news sources: A network analysis of CBS News, Nightline, MacNeil/Lehrer and This Week with David Brinkley. Journal of Communication, 44(2), 84-107. Rogers, E., Dearing, J., & Chang, S. (1991). AIDS in the 1980s: The agenda setting process for a public issue. Journalism Monographs, 126, 1-47. Shilts, R. (1987). And the band played on. New York: St. Martin's Press. Sigal, L. (1973). Reporters and officials: The organization and politics of newsmaking. Lexingron, MA: D.C. Heath and Company. Smith, R. (2003, March 14). AIDS drugs aim to be 'real.' Washington Blade, p 5.. Soloski, J. (1989). Sources and channels of local news. Journalism Quarterly, 66(4), 864-870. Staples, G. B. (2001, June 3). The new AIDS crisis. Atlanta Constitution and Journal, p. 1C. Streitmatter, R. (1995). Unspeakable: The rise of the gay and lesbian press in America. Boston: Faber and Faber. Strub, S. (1994, March/April). HIV standard of care, POZ, 60-61. Waldman, P., & Devitt, J. (1998). Newspaper photographs and the 1996 presidential elections: The question of bias. Journalism and Mass Communication Quarterly, 75(2), 302-311. Table 1 Frequency of Attributed Comments by Source Type (N=467) Frequency Percent Government (non-medical) 2 .4 Government (elected) 4 .8 Government (medical and/or scientific) 29 6.2 Medical and/or scientific (non-governmental) 199 42.6 Medical and/or scientific (unspecified) 13 2.7 AIDS activist 64 13.7 Legal 3 .6 Unaffiliated HIV-positive 118 25.3 Unaffiliated HIV-negative 9 1.9 Corporate 11 2.3 Other 15 3.5 Note. There were no attributed comments made by four source types: Government (unspecified), public figures (HIV-positive); public figures (HIV-negative) and religious. Table 2 Paragraphs in Which First Attributed Comment Occurred for Three Prominent Source Types (N=258) Source Type Graphs 1-3 Graphs 4-6 Graphs 7-9 Below Graph 10 Medical and/or Scientific (non- 46% 21 10 23 governmental) Unaffiliated 38 25 10 27 HIV-positive AIDS 35 25 5 35 Activist Note. Percent of first attributed comments in each block of paragraphs. Percentages across total 100%. Table 3 Tone of Coverage in POZ, 1996-2000 Number of Stories Tonescore (Mean) SD 100 .94 .45 Note. Scale of 0=negative to 2=positive.