WK 11 DIS. DATA

The Tuskegee Study of Untreated Syphilis and Public Perceptions of Biomedical Research: A Focus Group Study Benjamin R. Bates, PhD and Tina M. Harris, PhD Athens, Ohio and Athens, Georgia

African Americans are less likely than European Americans to participate in biomedical research. Researchers often attribute nonparticipation to the "Tuskegee effect." Using critical qualitative analysis of focus group data, we exam- ined the public's use of the Tuskegee Study of Untreated Syphilis (TSUS) to discuss biomedical research. Our partici- pants articulated three prmary themes in relation to TSUS: 1) that TSUS made them suspicious about biomedical research; 2) that other values had to weigh against con- cerns about TSUS; and 3) that Afncan Americans could take steps to resolve their concerns about TSUS. African Ameri- cans were more likely to discuss TSUS than were European Americans. African Americans did not use TSUS to express simple fear. African Americans suggested issues other than TSUS that influence the decision to participate in research. African Americans indicated specific reforms that would increase participation in research. We discuss how a better understanding of African Americans' use of TSUS can enhance research participation and allay concerns about "another Tuskegee."

Key words: genetic research U Afrcan Amencans-l Tuskegee * Public Health Service Study of Untreated Syphilis in the Negro Male * focus group methodology U crtical qualitative analysis

© 2004. From School of Communication Studies, Lasher Hall, Ohio University, Athens, OH (Bates) and the Department of Speech Communication, Terrell Hall, University of Georgia, Athens, GA (Harnis). Send correspondence and reprnt requests for J Natl Med Assoc. 2004;96:1051-1064 to: School of Com- munication Studies, Lasher Hall, Ohio University, Athens, OH 45710; phone: (740) 593-9163; fax: (740) 593-4810; e-mail: [email protected]

Thejudgments made in 1973 about the conduct ofthe Tuskegee Study in 1932 are made with the advantage ofhindsight, acutely sharpened over some 40 years concerning an activity in a different age with differ- ent social standards. Nevertheless, onefundamental ethical rule is that a person should not be subjected to avoidable risk ofdeath orphysical harm unless he freely and intelligently consents.I p. 12

It is important to be clear about what this study meant for the participants. It meant thatfor up to 30 months, several hundred people with HIV infection were observed but not treated... Such a study could not have been performed in the United States, where it would be expected that patients with HIV and other sexually transmitted diseases would be treated.2 p. 967

When discussions of informed consent and med- ical ethics take place, a common reference point is the Public Health Service Study of Untreated Syphilis in the Negro Male, more commonly called "The Tuskegee experiment" or the Tuskegee Study of Untreated Syphilis (TSUS). Reactions to TSUS caused a reevaluation of how medical research should be conducted.

The reforms enacted after TSUS did not necessari- ly change the practice of all biomedical research. The first epigraph comes from the Final Report of the advisory panel that suggested these reforms. Although their recommendations have been adopted in the Unit- ed States, research infected with problems similar to TSUS is still conducted. The second epigraph comes from Angell's New England Journal ofMedicine edi- torial2 discussing a study performed by Quinn and col- leagues.3 In Quinn's study, HIV-positive persons in rural Uganda were observed but not offered treatment. In addition, it is unclear whether these person's sexual partners were informed about their risk. Although institutional review boards approved Quinn's study, Angell indicates that some bioethicists disapproved of the study because it failed to learn the lessons of TSUS. Like Quinn's study, TSUS was not conducted

JOURNAL OF THE NATIONAL MEDICAL ASSOCIATION VOL. 96, NO. 8, AUGUST 2004 1051

PUBLIC PERCEPTIONS OF BIOMEDICAL RESEARCH

in secret. Like Quinn's study, TSUS studied persons of African descent. Like Quinn's study, TSUS did not make treatment available to research "subjects." Like Quinn's study, TSUS denied informed consent by not fully explaining risks and treatment options to these "subjects." And, like Quinn's study, larger institutions approved and funded the study.

TSUS was publicly challenged in 1972. White reports that articles about TSUS were published throughout the life of the experiment, including prominent outlets, such as the Journal ofthe Ameri- can Medical Association, Archives ofInternal Medi- cine, and the Journal of Chronic Disease.4-'0 These publications indicated that the biomedical community accepted TSUS. Yet, after Heller's report on the study in The New York Times, public reaction was swift and condemnatory.'1 A national review board found that TSUS was unjustified as science.'2 A special commis- sion was appointed, reforms were suggested, and new regulations were enacted."3-5Despite these changes, the legacy ofTSUS lives on. Tuskegee has become a metaphor for problematic research and serves as a historical marker for discontent with the biomedical community by racial groups chosen as "guinea pigs.''l6 l7 Moreover, research conducted in the pres- ent-research like Quinn's-seems to confirm fears that there could be "another Tuskegee," research that knowingly deprives people ofmedical care to observe the progression of a disease without the informed consent ofthe people being studied.

The purpose of this paper is to explore whether members of the public recall TSUS and whether they use the study to interpret current biomedical research. In other words, does this historical marker influence public discussion ofbiomedicine? The public has dis-

Table 1. Summary of Demographic Characteristics of Focus Group Participants

African-American European-American Hispanic Multiracial Total* N 118 71 15 11 216

Fall 2001 Moderator Guide 60 46 8 0 115 Spring 2002 Moderator Guide 58 25 7 11 101 Urban area 61 19 0 1 61 University town 42 19 1 10 82 Regional hub 82 33 14 0 72 Male** 75 38 8 6 127 Female 41 33 7 5 86 Mean age 34.36 30.93 34.60 25.45 32.7 Median Income $50,000 $55,000 $39,500 $50,000 $50,000 Median educational level Some College College Some Some

college degree degree college college * One participant declined to self-identify racially on the demographic forms. ** Two participants declined to self-identify a sex on the demographic forms.

played the ability to recognize TSUS.'8-20 There has not been research into the public's spontaneous use of TSUS in conversations about research. In addition to recognition studies, the public's recall ofTSUS, with- out prompting, should be investigated. Recall memo- ry requires a higher level of cognitive processing than does recognition. Recognition requires the identifica- tion of previously learned material only; recall requires a person to integrate contextual cues with old information and to articulate a new statement.21'22 As such, this study will examine unprompted, sponta- neous discussion ofTSUS as it emerges in the context of a larger discussion ofbiomedical research.

The discussions ofTSUS examined in this paper come from our research group's larger investigation of the public's understanding of genetics. Although we have reported on specific questions asked to our focus groups elsewhere,2325 in this paper, we investi- gate the use ofTSUS as an emergent theme in public discussion. We begin by reviewing previous discus- sions ofTSUS. After noting limitations of this work, we present a focus group study of how members of the public use TSUS as a reference point. We outline differences between African Americans and other ethnic groups in their use of the Tuskegee Study nar- rative and indicate how TSUS influences public understanding of biomedical research. Finally, we discuss how understanding the ways that TSUS is discussed in public can be used to encourage African Americans' participation in biomedical research.

Background The Tuskegee Study is a landmark in the history of

medical ethics. The study is more than a testament to the ability of some people to dehumanize other people

1052 JOURNAL OF THE NATIONAL MEDICAL ASSOCIATION VOL. 96, NO. 8, AUGUST 2004

PUBLIC PERCEPTIONS OF BIOMEDICAL RESEARCH

or of the fascination with science over other human values. TSUS has become a metaphor for concerns that members of the public have about biomedical research. The medical literature emphasizes TSUS's exploitation of vulnerable populations.'7'26-3' Although there has been much attention paid to the fact that the subjects in TSUS were all African-American, this pop- ulation was also exploited because of its low socioeco- nomic status and because of the lack of formal educa- tion received by these subjects. TSUS also symbolizes the arrogance of the government and of physicians in deciding which patients "deserve" proper medical treatment and which patients do not.17'28'30

The combination of exploitation and arrogance may have led the Public Health Service to feel that they owned not only the research findings but the very bodies of the men participating in the study. Roy32 has concluded that the use of these 600 men's bodies for profit at the expense of the men's humanity amounted to a new form of slavery. In addition to exploitation of "subjects," TSUS has also become a metaphor for medical malpractice in research.'7 0'33-34 The decision to withhold initial treatments of heavy metals (the best available therapy for syphilis in the 1930s) and of penicillin (following its discovery in the 1940s) so that untreated syphilis could be observed across the lifes- pan has come under much scrutiny.20'26-28'35-36 Although TSUS was defended in 197237 and has defenders in the present,4 the withholding of therapy has been com- pared to genocide-the systematic killing of a racial or cultural group-because race was used to select those who would go untreated.30'33'38

Given the associations made by medical ethicists and historians among TSUS, genocide, and slavery, it is unsurprising that African Americans are more aware ofTSUS than are European Americans.30'33'39 Additionally, African Americans are more likely to distrust the biomedical system than are European Americans.40 This distrust, it is often asserted, is because of TSUS. For example, Cox,4' p. 2 claims that TSUS is fundamental to the "lore of prejudice and persecution children learn in many black com- munities." TSUS is assigned blame for low African- American participation in clinical trials, blood dona- tion, vaccination, and other medical research and intervention. 1920'26'3842-45 TSUS has also been identi- fied as the root for African-American cultural myths. For instance, beliefs that HIV/AIDS was designed by the government to exterminate African Americans have been laid at TSUS's door.16"17'29-30,33,34 Likewise, TSUS is faulted for myths that crack cocaine and Agent Orange are products of the Cen- tral Intelligence Agency aimed at African Ameri- cans.2045Because ofthe prevalence of these beliefs, there have been calls for greater cultural sensitivity in biomedical research and for the redress of larger

discrimination in society to rebuild trust.'6"17 These calls, some believe, have gone unanswered

by biomedical researchers and by the U.S. federal government. Because of the complexity of the "Tuskegee effect" and the difficulty of addressing public concerns about biomedical research, public health authorities may tend to ignore the "Tuskegee effect" when attempting to recruit participants of col- or for research studies.'6'20 Moreover, the failure to address concerns linked to TSUS may lessen the cred- ibility of public health authorities for persons of color because rumors about biomedical research are unad- dressed. Other scholars indicate that there may not be a "Tuskegee effect," as mistrust of biomedical research predates TSUS.'7 The "Tuskegee effect" may be minimal, as people are more likely to cite time constraints, familial obligations, or a fear of having blood drawn as reasons for not participating than they are to cite concerns about TSUS.4'

The ability to address these concerns is also con- strained by false understandings of TSUS. That is, because people do not know what "really" happened in TSUS, the public may need to be educated about historical facts and the protections for human sub- jects that have since been enacted. According to a 1999 poll by the Kaiser Family Foundation, only 38% of the public claims to have heard ofTSUS.46 A greater percentage of African Americans (59%) claims to have heard ofTSUS. Among all groups, the foundation found that a substantial portion of those who have heard of the study reported that TSUS was the training of the Tuskegee Airmen, a heart disease study, or did not know. Public understanding of TSUS is not necessarily bound by historical accura- cy, as some of those who know that TSUS was about syphilis believe that the government deliberately infected the participants.2728 Although many people claim to be familiar with TSUS, they often misstate the nature of the study, its extent, and its duration.20

As such, some researchers believe that concerns linked to TSUS are a problem of education. These researchers maintain that once the public receives accurate information, their concerns will evaporate. Indeed, White4 goes so far as to defend TSUS. He claims that TSUS was nonracist, that it followed the best-known norms of treatment regarding penicillin, and that nontreatment of syphilis was a standard prac- tice. Few researchers go as far as White. They do indi- cate, however, that describing substantive changes between TSUS and current research will reduce poten- tial participants' concerns about enrolling in biomed- ical research.47A8

Research Question Despite debate over the "meaning" of TSUS,

explorations of the public's use of TSUS in reacting

JOURNAL OF THE NATIONAL MEDICAL ASSOCIATION VOL. 96, NO. 8, AUGUST 2004 1053

PUBLIC PERCEPTIONS OF BIOMEDICAL RESEARCH

to biomedical research have been rare. Most studies that outline public concerns linked to TSUS are his- torical reviews or ethics commentaries designed to promote better research norms. Freimuth and col- leagues20 showed HBO's Miss Evers'Boys and asked focus groups for reactions. Although references to fiction are not unproblematic because of their free use of "artistic license," films like Miss Evers'Boys can provide a dramatic presentation of ethical issues and serve as a starting point for discussion.49-50 A few studies have used survey techniques to evaluate the public's self-reported knowledge of TSUS and its ethical implications.'8-'946

Studies showing films or utilizing survey tech- niques rely on the principle of recognition to explore public understandings of TSUS. Recognition-an awareness that something perceived has been per- ceived before-requires that the participant have associations when a person mentions TSUS.21'22 In recognition, a stimulus is presented to which a person reacts. The awareness that emerges from recognition may be incomplete or incorrect. These studies do not address recall. Recall-the remembrance of what has been previously learned or experienced-requires response that draws on more complete understand- ings of the term or concept.21'22 Recall goes beyond simple awareness in that it assigns both denotative and connotative values to the person, place, or thing recalled. Recall is also more spontaneous. Recall also allows free association, the expression of new ideas, and the integration ofnew knowledge with old knowl- edge. Even when a person is not specifically asked about a topic, they may think of that topic in a larger discussion. The questions in survey research are less able to take advantage of recall because surveys must be consistent and are highly structured. As such, they generally rely on recognition.5" Because recognition and recall operate through different cognitive path- ways, different kinds of research may be needed to access recall than are used to access recognition.52-53 Qualitative methodologies tend to have a more free- flowing nature than do quantitative methodologies. In addition, qualitative methods open space for partici- pants to build on the ideas presented by other partici- pants in addition to questions posed by the researcher. Qualitative methods allow the research participant to indicate the topics and concepts that he or she believes are important, rather than those that the researcher has predetermined are important. As such, qualitative methods are more useful for study of how concepts are recalled than are quantitative methods.54

The central research question in the public under- standing ofTSUS and its implications for biomed- ical research should go beyond recognition. The cen- tral research question, when one is interested in how public decision-making is influenced by TSUS,

should be centered on recall. As such, our central research question is: In discussions of biomedical research, do members of the public use Tuskegee as a touchstone for articulating concerns about such research? That is, do members of the public sponta- neously recall TSUS in these discussions? In addi- tion, we should explore whether the public draws on TSUS to formulate arguments. If there is a substan- tial "Tuskegee effect," members of the public should articulate TSUS as a concern without the urging of the investigators.

Participants and Procedures To explore if members of the public draw on

TSUS to express their understanding of biomedical research, a focus group methodology was employed. As part of a larger project that is investigating the public's understanding of medical genetics, our study group conducted 24 focus groups with 215 total participants from August 2001 through March 2002 in the southeastern United States. Twelve groups had persons self-identifying as "black" or "African-American," seven groups had persons self- identifying as "white" or "European-American," and two groups had approximately equal numbers of African and European Americans. One group had persons self-identifying as multiracial, and two groups had persons self-identifying as Hispanic or Latino. Moderators were matched by self-identified race, with one integrated focus group being moder- ated by an African-American female and the other moderated by a European-American female. Moder- ators were of the same race as focus group partici- pants in an effort to elicit honest and open dialogue about genetics, race, and human genetics research. Sessions lasted about two hours.

Participants were recruited by nomination from a community advisory board in three areas: a large urban area, a regional hub that was primarily agri- cultural but is now in transition, and a university town. Community advisory boards were first asked to discuss what constituted their community. Then they were asked to nominate individuals not on the board who would represent the perspectives in their community. Those individuals were then telephoned by research team members and invited to participate if they did not self-identify as "experts" in genetics. These individuals were considered to have "lay knowledge" of genetics because they did not have in-depth academic or practical experience with genetic science that would qualify them as "experts." Participants were told by whom they were nominated and offered $50 in compensation.

As indicated in Table 1, participants were 118 African Americans, 71 European Americans, 15 His- panics, and 11 self-identified multiracial persons, and

1054 JOURNAL OF THE NATIONAL MEDICAL ASSOCIATION VOL. 96, NO. 8, AUGUST 2004

PUBLIC PERCEPTIONS OF BIOMEDICAL RESEARCH

one person declined to answer; 86 were female, 127 were male, and two declined to answer. All participants provided written informed consent, and appropriate institutional review boards approved the study. People with a lay understanding of genetics were recruited from urban (61), suburban (82), and rural (72) commu- nities in the southeastern United States. Participants ranged from 18-64 years of age (M=32.7). Family income ranged from $7,500 to $700,000 with a median of $50,000 (median income for the state is $36,372). Educational levels ranged from less than high school (3%) to a terminal degree (1%). Most participants had either a high school diploma (20%), some college edu- cation (31%), or had completed a bachelor's degree (31%). The remainder had some graduate (9.5%) or doctoral (5.2%) education.

Standard qualitative focus group procedures were followed. Moderators employed a series of questions, which had been developed by the research team and then reviewed and revised by the three community advisory boards. Groups that met from August to December, 2001 were asked common questions from a 21-question moderator guide, while groups that met from January to March 2002 were asked com- mon questions from a 29-question moderator guide. Both moderator guides are available upon request to the authors. Moderators included follow-up probes and revised wordings in questions to seek full explo- ration of the issues by as many participants as possi- ble. Audiotapes of the sessions were transcribed. These transcripts were then compared to the video- tape and corrected as necessary. Transcripts were not corrected for grammar or for formality in order to capture the oral nature of the discussion.

Specific questions asked by the moderators have been reported on elsewhere. These questions includ- ed, among others, what participants thought the phrase "a gene for heart disease" meant (Fall 2001 groups),23 what concerns participants had about con- suming race-based pharmacogenomics medications (Spring 2002 groups),24 and the roles that inherited genes, social factors, personal behaviors, and ecologi- cal environments played in the development ofhuman attributes, such as height, weight, mental abilities, and risk of genetic disease (all groups).25 In our research group's previous studies, we performed our analyses primarily to answer questions that our research team posed. In the current study, we drew on the standards of critical qualitative research to investigate one ofthe significant topics articulated by our participants that were not posed by our moderator guides.

In this study, a critical qualitative analysis of talk turns related to Tuskegee was performed. Critical qualitative focus group research differs substantially from market-based methods offocus group research.55- 59 Rather than beginning with traditional norms of

logico-deductive research wherein a question is asked and the responses are coded for trends in response to that question and reduced to a numerical categoriza- tion, critical qualitative research begins with anom- alous data that does not fit preconceived questions and expected types of answers. After conducting previous studies, members of our research team noted several references to TSUS that cut across questions and did not fit into the expected answers to our questions. Having noticed this anomalous data, two members of our research team (the authors) collected TSUS-relat- ed terms such as "Tuskegee," "syphilis," and "Macon County, AL." These terms were then used to identify all talk turns where TSUS was discussed. Because the- matic analysis is contextual, we (the authors) also examined the discourse surrounding these mentions. Sections where participants used TSUS to discuss their understanding of biomedical research were iso- lated for closer analysis. The purpose of this analysis was not to reduce the comment made to fit into previ- ous questions but to allow the data to generate ideas that could be further investigated in later studies. Criti- cal qualitative investigations are designed to, a) allow participants a voice in shaping the research that emerges from focus group sessions, b) make sense of anomalous human behavior that does not fit precon- ceived questions, and c) provide data points that can be investigated in later research for further generalizabili- ty. As such, we (the authors) do not claim that all TSUS-related phenomena can be explained through a single theory, such as the fear-based paradigm articu- lated in the literature review, but instead seek to open space for a greater diversity ofviews about TSUS that are articulated in reference to biomedical research. We report examples from this use ofTSUS below.

Results There were apparent ethnic differences in the use

of Tuskegee. Out of the 24 focus groups, only those groups with African-American participants men- tioned TSUS. No group composed solely of Euro- pean Americans, Hispanics, or multiracial persons referenced TSUS. The only mention ofTSUS by any European-American participants in our study was after an African-American male mentioned TSUS as a reason for being suspicious of genetic engineering. In this integrated group, two European-American female participants agreed with him.

2i-JJJ: As an African-American, I'd be suspicious. I want to say that, basing it on history. I think about things like the Tuskegee experiment and things like that. So as an African-American, I'd be aware of what s going on. You know? Yeah, because ofhistory, what happened.

JOURNAL OF THE NATIONAL MEDICAL ASSOCIATION VOL. 96, NO. 8, AUGUST 2004 1055

PUBLIC PERCEPTIONS OF BIOMEDICAL RESEARCH

2i-202: Right. I would be suspicious too. People use things improperly all the time. People do manipulate society.

2i-206: Yeah, manipulate society with the data.

2i-202: Manipulate things for their own needs all the time. So ifthere was any kind ofreference like that, I would be suspicious.

This conversation was the only discussion of TSUS by non-African-American participants. After TSUS was mentioned by an African-American par- ticipant, some European-American participants vali- dated his concern. Their concerns, however, were not as potent or specific as those ofAfrican Ameri- cans. The European-American participants appeared to be more concerned with the propensity for manip- ulating data than they were with the methodological and racial implications ofTSUS. Although partici- pant 2i- 11 1 did not clarify his concerns about TSUS, the statements made by other African Americans (as described below) indicated that TSUS is a potent marker for African Americans.

The relative absence ofmentions of discussion of TSUS by European Americans, Hispanics, and mul- tiracial persons, as compared with African Ameri- cans, may indicate that TSUS has greater salience for the African-American community than it has for members of other ethnic communities. This finding, in itself, is not surprising. Because African Ameri- cans were the "subjects" in TSUS, there is more likelihood of identification between contemporary African Americans and the TSUS subjects than there is for other ethnic groups. When contemporary experiences with discrimination in medicine are tak- en into consideration, the indication that TSUS is a touchstone for African Americans and not for others should be expected.

Suspicions Based in TSUS Indeed, TSUS was a marker for distrust ofbiomed-

ical research for African Americans. African-Ameri- can participants in our focus groups indicated that the experience of TSUS made them less likely to trust medical research. TSUS was not raised as an issue by the moderators. Rather, the participants in our African- American groups used the discussion of biomedical research to raise the issue ofTSUS themselves. This conversation between an African-American female and an African-American male is indicative:

4-108: Um, this isjust a general question. Do people still have concerns about being tested? I mean, I know that a lot ofpeople ofour race don 't believe in being part ofresearch studies or being tested, and

when you 're talking about genetics, I mean, they may, you know, conjure up some negative connotations...

4-101: They have reason to.

4-108: They may still have problems with the Tuskegee experiment.

4-101: Exactly.

4-108: [They use it] as a toolfor why they may not trust their healthcare provider orfeel like they can open up enough to talk to their healthcare provider related to this, because they don 't want any testing to be done on them. I mean, I still see it as aproblem.

4-101: It is a problem.

In the conversation, the mention of a concern about being tested was linked to TSUS. The implica- tions ofTSUS for the African-American community were clearly explained. Some African Americans may believe that participating in the biomedical sys- tem will expose them to further dangers. TSUS does not force African Americans to distrust medicine. Instead, the way that African Americans use their understanding of TSUS may be what undermines trust in healthcare.

TSUS does not influence all African-American understandings of the biomedical system. Instead, some African Americans may generally be suspi- cious of the system and raise questions about it. When the issue ofTSUS is inserted into the debate, however, mention ofTSUS can alter the direction of the conversation. For example, in the middle of a discussion about possible race-based pharmaco- genomics,24 participant 12-108, an African-Ameri- can female, indicated that she was influenced by ear- lier discussions ofTSUS. She said,

When he mentioned the Tuskegee experiment, I'm like, Oh, gad! This is... That's scary. I would have run the other way. I would. Simplyfor, I mean, our government's history in this country, I would seri- ously have to question, "What are they trying to do to us now? " I really would. 'And what is the drug? " Is it going to kill more ofus before we 're 55? I mean, what's going on and historically-I mean, before he mentioned the Tuskegee issue, I was thinking like, "Well, maybe." [LAUGHS] And then we said that, I said, "Oh yeah, they're lying." So I couldn 't-I don 't know ifI could trust something like that.

This participant indicated that when the issue of TSUS was incorporated into the debate, it enhanced her concerns about race-based pharmacogenomics.

1056 JOURNAL OF THE NATIONAL MEDICAL ASSOCIATION VOL. 96, NO. 8, AUGUST 2004

PUBLIC PERCEPTIONS OF BIOMEDICAL RESEARCH

The acceptability of race-based pharmacogenomics was not fixed when this participant first considered it. The "maybe," though, was converted into a firm "no" when the specter ofTSUS arose. Because the salience ofTSUS was strong for this participant, race-based pharmacogenomics became something that could not be trusted and was, in all probability, based on lies.

TSUS and the Intersection with Other Issues

Previous discussions of how TSUS influences perceptions of biomedical research have largely assumed that most African Americans adopt views similar to those of participant 12-108. Although TSUS is a strong and salient marker for many African Americans, consideration ofTSUS does not exist in a vacuum. Instead, African Americans may also weigh concerns about TSUS with and against other issues relevant to biomedical research.

Some African-American participants indicated that concerns about TSUS needed to be balanced against the need for sufficient African-American rep- resentation in biomedical research. That is, even if TSUS raised concerns about exploitation, for a med- ication to be considered safe and effective for African Americans, enough African Americans had to be included in the medication's evaluation. This need is illustrated in a conversation in one ofour groups:

9-104: You know, there's a great amount ofpeople out there who don 't know to think about beyondjust what they're saying. And think more about, "Well, who did they test and actually who are these peo- ple?" Thinking beyond exactly what they're saying. And saying, "Well, okay. I'm black so this is going to be greatfor me " and then go out and get it and it s not the whole story.

9-101: But you wouldn't think they wouldjust test blacks? You got the Tuskegee incident. That was when theyjust tested the black guysfor Syphilis. I'm sure they wouldjust test blackpeople, I mean.

9-114: But I'm saying not all blackpeople are alike.

9-101: No, heck no.

9-114: But I'm saying, who 's to say that they're test- ing what s extensive enough to cover all the bases?

9-101: Thats true. You never know. Imean, what might workforyou, might not workfor other black women.

As this conversation developed, the need for rep- resentative research was made more important than concerns about TSUS. As such, medications that

work for European Americans could not be assumed to be equally safe or efficacious for African Ameri- cans. The participants in our groups had enough understanding of the scientific method to realize that a small sample does not reflect the whole popu- lation. Moreover, the inclusion ofjust a few African Americans would be insufficient as well. As such, for these participants, the issues ofTSUS and of rep- resentative research needed to be balanced.

The desire to have effective medication was not a theoretical issue for our participants. For example, when an African-American male tried to raise TSUS as a reason for not participating in research, an African-American female responded with a thor- ough explanation for why she set concerns about TSUS aside and chose to participate in an NIH- sponsored research study. In this group, the follow- ing conversation took place:

6-105:A lot ofpeople don 't trust thefederalgovernment.

6-108: They 'rejust testing.

6-105:A lot ofpeople don 't trust things. Like somebody mentioned the Tuskegee thing. Iain 'tgoing to go....

6-108: I didsome ofthose studiesjust to get money off of[them]. AndI didn 't see nothing wrong with some of them. But it was a sleep study. They wanted to monitor me and see how I sleep. You know see how my mind is thinking, you know, howfar it carries on and stufflike that. So I thought that was a good studyfor me to get on and everything and stuff So when this person had, like, a sleeping disorder and everything, and they was likejudging me with them and stuff I thought that was goodfor thefact Iwas helping anotherperson out. But you know, a lot ofpeople don 't be interested in studies like that that help otherpeople out.

Participant 6-108 transformed participant 6-105's TSUS-based concerns into a weak excuse. For 6-108, the need to help sick people was a more important value than concerns about TSUS. Because of her per- sonal experience with this sleep study, 6-108 indicat- ed that the central variables under study compared a person with an illness to a person without that illness. The issue of race-and, thus, the reference to TSUS-was not valid when considering the utility of this study. By changing the ground rules and placing a different system ofjudgment on biomedical studies, concerns about TSUS could be suppressed in favor of an ethic that cares for the needs ofothers.

Reconsidering the weight ofTSUS did not always require that a second participant provide alternative values to place against concerns linked to TSUS. In fact, our participants carefully weighed TSUS

JOURNAL OF THE NATIONAL MEDICAL ASSOCIATION VOL. 96, NO. 8, AUGUST 2004 1057

PUBLIC PERCEPTIONS OF BIOMEDICAL RESEARCH

against the potential advantages of biomedical research. This complexity is reflected well in partici- pant 12-103's attempt to decide whether his con- cerns about TSUS should outweigh his desire to take advantage of genetic engineering. He said:

After the Tuskegee experiment, Ijust about run another way. ... Ijust don 't trust people that have that kind of genetic knowledge and all that. Ijust, you know-I mean-I'm conflicted because ... itsjust one ofthose things thatyou want to giveyour child every advantage in this society, getting more and more competitive. I don 't think I'd do anything like looks-wise, you know, but I would do something, you know, to insure that she was disease-free, or, you know, something, you know, to just try to give her an edge over ... disease or things that might hurt her in life or something like that. Kind of you know, vaccinate her, ifyou will, against all that kind ofstuff But this right here, you know, Idon 't know. I'd stay awayfrom medicines because I'm just con- vinced that, you know, when it comes down to AIDS and stufflike that, you know, I'm convinced that it was meant to, you know, for some reason. Its an old rumor and it never really substantiated, you know, conspiracy theory. But Ijustfeel like, you know, this government and this world don 't really mean blackpeople no good.

This statement points to many of the issues identi- fied in previous research. 12-103 indicated thatTSUS has made him suspicious ofthe biomedical communi- ty. He also related these concerns to what he explicitly identifies as conspiracy theories about AIDS. These concerns, however, are placed into balance with the desire to "vaccinate" his daughter through genetic engineering against disease. Thus, rather than con- cluding that TSUS is a sufficient reason to not partici- pate in the biomedical system, 12-103 indicated that it is an important issue that must, nonetheless, be con- sidered among others. Although 12-103 ultimately concluded that he did not want to participate in genet- ic engineering, his complex decision-making indi- cates that TSUS does not cause African Americans to act irrationally because TSUS took place.

The question ofrational objection versus irrational objection to biomedical research because of TSUS was raised by our participants. Our participants indi- cated that TSUS was a good reason to object to bio- medical research but also held that education about how medical studies are performed and the availabili- ty of opportunities for participation would allow these concerns to be addressed and overcome. The need for rational, informed decision-making about participa- tion was outlined by our participants in conversations similar to the following. Participant 10-110, an African-American female, said:

I agree with P112 on the education aspect ofwhat she is saying, but I also-I don 't just think it came from the Tuskegee catastrophe. We have been, as a race, have been shoved aside-it's not like we don 't- we 're not capable ofgetting the education, it is that they make it that much harderfor us to get the educa- tion, and you know, they got all the resources. So it's like we, you know, like we don't want to learn, or want to a, know about these studies, but every time, it seems like when you try to get involved in something they either want to talk over your head, and see I'm one of them people, I don't mind, if I don't know something, I will ask. I don 't care ifI'm hanging with the President. Ifhe say a word that I don't under- stand, I'll stand up in front of anybody and say, "excuse me, that word you just used, you need to define thatfor me, cause I want to know what you 're talking about." There's nothing wrong with not know- ing-you know, and I will ask you, so you know Ijust think that we have a lack ofdistrust, but it's not mis- placed. We kinda get the short end ofthe stick all the time, and it's timefor that to turn around some.

Distrust because ofTSUS was only one reason that African Americans do not participate in bio- medical research studies. Other reasons create a more complex picture. According to participants like 10-1 10, a lack of education on the part ofmany African Americans was a leading cause of nonpar- ticipation. Blame was assigned primarily to a racist educational system that does not provide equal opportunities for African Americans. To remedy this problem, 10-110 suggested that African Americans need to insist that things that they do not understand be clearly explained. Thus, 10-110 claimed that when African Americans are recruited for participa- tion in biomedical research studies, they need to ask questions about the study to see what is involved and to make an informed decision about participation.

Solutions to the "Tuskegee Effect" Requesting additional information was only one

of the solutions to the "Tuskegee effect" proposed by our participants. Because they recognized advan- tages to African Americans from participation in biomedical research, our participants indicated that African Americans should participate more often. Encouraging participation, however, required more than additional education. Two additional compo- nents were suggested. Both suggestions require the active involvement ofAfrican Americans in the bio- medical system.

The first addendum provided by our participants was the need for greater involvement by African- American physicians in conducting biomedical research. Participant 6-1 10, an African-American

1058 JOURNAL OF THE NATIONAL MEDICAL ASSOCIATION VOL. 96, NO. 8, AUGUST 2004

PUBLIC PERCEPTIONS OF BIOMEDICAL RESEARCH

female, suggested that African Americans would be more likely to trust African-American physicians than they would be to trust European-American physicians. She said,

I thinkpeople seem to remember like the Tuskegee incident, a lot ofpeople that are really afraid to be involved in that kind ofstuffanyway. I mean, that's why I said, dealing with a black doctor person. Because I know that his motives might be a little bet- ter when I talk to an African-American person than a white doctor

This higher level oftrust accorded to a physician of the same ethnic background than to one of a different ethnic background is not unique. African Americans have repeatedly expressed greater trust in, and higher preference for, African-American healthcare providers than in European-American healthcare providers.40 This preference does not correspond to objective measures of patient care quality. Indeed, there is no significant difference in the objective quality of care provided to African-American patients by African- American physicians than that provided by European- American physicians.6' Instead, this preference may be because African-American patients may believe that African-American providers are more likely to share similar cultural values, beliefs, and experiences that affect healthcare provision than are European-Ameri- can providers.62 As such, if African-American providers become more involved in the recruitment of African-American participants in biomedical research and in the conduct of studies, it is possible that African Americans will be more likely to participate.

In addition to African-American healthcare providers becoming more involved, our participants suggested that African Americans also need to demand opportunities for participation in biomed- ical research while protecting their rights as "human subjects." Responsibility could not be devolved to healthcare providers. Even if healthcare providers intended to include more African Americans in research, our participants indicated that actual change would occur if, and only if, change was demanded by African Americans. This conversation is indicative:

10-118: It's going to comefrom within us to turn any- thing around, so we need to look within ourselves to make the change, because they're not going to do it- it's going to be like that. Iknow we're going.

10-110: But they have the control, we can change, but.

10-101: We can change but we don 't have the control.

10-110: Right, we can change, and we can progress, but we can 'tforce them to progress, and, and accept us as equals, do you know what I'm saying-we got- ta step up to the plate.

10-101: They already think we are equal already, and we're not.

10-110: And that is what they are basically so threatened by.

The "they" to whom they are referring in this conversation is unclear. "They" might be the bio- medical research community. Alternatively, "they" might be a larger system of white privilege that per- vades the United States. Whoever "they" are, they are outside of the African-American community. "They" are not all-powerful, though. These partici- pants indicated that, to have systemic change in the conduct of biomedical research, African Americans need to demand such change. Changes could include treating African-American participants in research studies as active and involved participants, rather than as objectified "subjects," or putting greater emphasis on conditions that affect African Ameri- cans disproportionately. These changes could threat- en established biomedical interests, a concern of which our participants were aware. Nevertheless, our participants indicated that, for African Ameri- cans to trust biomedical research, they needed to voice demands for change.

DISCUSSION Our results indicate that TSUS is a touchstone for

some people's understandings ofbiomedical research. TSUS is used differently by different communities. The way that TSUS is used by the public-particular- ly the African-American public-is also more com- plex than suggested by previous research. Rather than serving as a base of fear, our findings suggest a multi- faceted situation. Yes, TSUS makes some people decline participation in biomedical research. Yet, TSUS is placed into a decision-making matrix in which multiple values are weighed against one anoth- er to decide whether TSUS is a sufficient reason to refuse to participate. Finally, the response to TSUS is not the simple rejection of participation but is, instead, the articulation of demands for change in the way that biomedical research is conducted.

TSUS is a touchstone for some communities in coming to understand biomedical research. 16"7202830'33'34'39 African Americans are the ethnic group most likely to use TSUS as a way of interpreting biomedical research. European Americans, Hispanics, and multiracial per- sons did not raise the question ofTSUS in their groups. As indicated by the single European-American discus-

JOURNAL OF THE NATIONAL MEDICAL ASSOCIATION VOL. 96, NO. 8, AUGUST 2004 1059

PUBLIC PERCEPTIONS OF BIOMEDICAL RESEARCH

sion ofTSUS, a discussion in response to an African- American statement, these ethnic groups may recognize TSUS but are not likely to recall it on their own. Because recall requires that the concept under discus- sion be used to interpret new data or ideas on a sponta- neous basis, it is possible that European Americans, Hispanics, and multiracial persons do not place TSUS at a high level in their cultural reservoirs for immediate interpretation. As indicated by previous research, African Americans are more likely to have thoughts of TSUS "at their fingertips" in public discussions. This cultural touchstone may be less relevant to non-African Americans. As outlined above, African Americans are more likely to perceive consubstantiality with the par- ticipants in TSUS than are members of other ethnic groups. As such, African Americans should be more likely to recall TSUS spontaneously than members of other ethnic groups. In this area, our findings agree with those ofprevious researchers.

Our research also suggests that the African-Ameri- can public's use ofTSUS to understand biomedical research may be much more complex than previous researchers have suggested.l8-204748 This simple model ofAfrican Americans' decision-making in contexts of biomedical research is illustrated in Figure 1. Previ- ous researchers have implied that African Americans recall TSUS and then refuse to participate in biomed- ical research. These assertions take two facts-that TSUS took place and that African Americans have lower participation in biomedical research than do European Americans-and assert a causal link

Figure 1. Presumed Decision Flow for African Americans in Deciding Whether to Participate or

Not Participate in Biomedical Research

Do Recall the Tuskegee Study?

If Yes If No

Distrust Trust Biomedical System Biomedical System

Do Not Participate Participate

between these facts. Although fewer African Ameri- cans participate in biomedical research than is desir- able for representative and reliable research,631-1 recall ofTSUS is not an absolute barrier to participation. Nevertheless, many researchers assert that, because TSUS happened, it caused African Americans to not participate. This sort of reasoning represents the post hoc ergo propter hoc fallacy in that it assumes a causal relationship between a chronologically prior fact and a chronologically subsequent fact. Of course, there sometimes is a relationship, but chronological subsequence does not necessitate logical conse- quence. In addition, African Americans have never withdrawn entirely from biomedical research. Although African Americans may be critical of or cynical about participation, a substantial number of African Americans participate nonetheless. IfTSUS led directly to nonparticipation, then no African Americans would participate in biomedical research.

Rather than assigning a simple cause-effect rela- tion between TSUS and low rates ofAfrican-Ameri- can participation, our participants indicated that their understandings ofTSUS did not lead to a sim- ple refusal to participate. Our participants articulat- ed three primary themes in relation to TSUS: 1) that TSUS made them suspicious about biomedical research; 2) that other values had to weigh against concerns about TSUS; and 3) that African Ameri- cans could take steps to resolve their concerns about TSUS. Previous research accounts well for the first only. 19,20,26,38,40-45 Our findings agree with previous researchers who claim that TSUS has made African Americans suspicious of participating in biomedical research. Our findings differ from previous research in that we do not claim that this suspicion leads to nonparticipation directly.

Suspicion may not stop participation in and of itself. The possibility of participation despite suspi- cion may be because humans are likely to include multiple factors in their decision-making. Previous research suggests that TSUS is a frightening concept that overwhelms all other components of decision- making.6"1720,29'30,33,34'38'45 Our findings suggest that other values may be weighed against TSUS. As illustrated in Figure 2, we suggest that African Americans may engage in a complex process of decision-making when deciding whether or not to participate in new biomedical research. African Americans are not likely to recall TSUS and withdraw from participation. Instead, our participants suggested that the desire for safe and effective medications, the imperative to help themselves and others, and the need for informed decision-making must be considered alongside con- cerns linked to TSUS. The desire for safe medication may require African Americans to participate in bio- medical research despite concerns about TSUS, as the

1060 JOURNAL OF THE NATIONAL MEDICAL ASSOCIATION VOL. 96, NO. 8, AUGUST 2004

PUBLIC PERCEPTIONS OF BIOMEDICAL RESEARCH

cost-benefit ratio may balance best in favor of partici- pation. The imperative to help others-a Samaritan ethic-may require that service to others be placed above one's personal concerns or desires. The need for education may indicate that making decisions out of fear, rather than informed choice, is a blameworthy, not praiseworthy, process. Regardless of the ethic chosen, our participants indicated that fear-based decision-making was less desirable than a more com- plex way ofthinking through decision-making.

When memories ofTSUS are faulted for the lower rates of African-American participation in the bio- medical system, the current biomedical system is absolved of responsibility because the indirect vic- tims of TSUS-African Americans today-are accused of transferring fears improperly to a new medical environment. The Institute of Medicine (IOM)'s report, Unequal Treatment, however, indi- cates that patient beliefs are only one variable in a larger set of causal factors for healthcare and research disparities.40 African-American interpretations of dis- parity in medical research and treatment, and their likelihood of requesting and receiving treatment, are influenced by conscious and unconscious stereotyp- ing and prejudice by both patients and providers, by disparities in the social, legal, and economic environ- ments, and by actual differences in physical examina- tions performed and diagnostic test results reported. When these factors are combined, the IOM indicates that patient input is only one part, and a small part, of differences between African Americans and others in choosing to take part in medical research and treat- ment. TSUS may be a "representative anecdote" about biomedical research, not an absolute barrier to participation. TSUS chose its subjects not just because they were African-American, but also because they were poorer, less educated, of lower sta- tus, and seen as less-deserving than European Ameri- cans by physician and governmental agents.'7'26-3' As such, TSUS may reflect more than the problems of a single study. TSUS may also reflect social, economic, cultural, legal, and ethical forces that have traditional- ly disadvantaged and may continue to disadvantage African Americans and serve as a representative anec- dote that articulates these themes. This anecdote, however, is countered by other stories and discussions that tell the advantages of research participation. As such, we question the simple link between concerns about TSUS and nonparticipation. Instead, it is proba- ble that the decision not to participate in biomedical research is a multifaceted decision, not a simplistic fear-based reaction.

Finally, our participants suggested possible solu- tions to the "Tuskegee effect." Although the "Tuskegee effect" is probably much weaker and much less causal than commonly assumed in previous research,26-28 32 35 36

the "Tuskegee effect" should be addressed for that small part ofthe population that reacts out of fear. Our participants provided valuable ways to address con- cerns about TSUS. Each of these ways of addressing TSUS may also influence choices in the decision flow, illustrated in Figure 2, that enhance the likelihood that African Americans will participate. Our African- American participants called for more community education so that African Americans would better understand what biomedical research was, why it was being conducted, and why they should participate. This community education might help to allay con- cerns linked to TSUS and might allow the considera- tions of other values that weigh against TSUS. Our participants also called for more African-American physicians to be involved in biomedical research. Just as the recognition of ethnic consubstantiality with the participants in TSUS may raise concerns about research, our participants suggested that the recogni- tion of ethnic consubstantiality with researchers could allay these concerns. Lastly, our participants suggested that African Americans become more involved in framing research concerns. As part of a larger plat- form advocating ethnic equality, our participants indi- cated that biomedical research could be one plank in that advocacy. Altogether, these suggestions were not fear-based reactions. Instead, they were responsive discussions of how the "Tuskegee effect" could be countered. Ifthese suggestions were adopted, it is pos- sible that African Americans would be more likely to participate in biomedical research.

CONCLUSION When our findings are taken together, they indi-

cate that TSUS is an issue that emerges in the African-American public's understanding of bio- medical research. Our findings expand the discus- sion ofTSUS beyond a paradigm of fear and suspi- cion. The African-American public may engage TSUS more critically than previous researchers have suggested. Although TSUS is articulated as a reason to be suspicious of the biomedical community, dis- cussion does not stop at that point. Instead, the African-American public may weigh the issues encapsulated in the Tuskegee metaphor against other needs for the African-American community.

The African-American public does not want "another Tuskegee." Reforms in biomedical research may not have been enough to fully allay suspicions of "another Tuskegee." Nevertheless, the African-Amer- ican public may recognize the benefits of African- American participation in biomedical research. African Americans may recognize that their participa- tion may be necessary for the development of safe and effective medications. African-American partici- pation in biomedical research studies, however, is

JOURNAL OF THE NATIONAL MEDICAL ASSOCIATION VOL. 96, NO. 8, AUGUST 2004 1061

PUBLIC PERCEPTIONS OF BIOMEDICAL RESEARCH

lower than European-American participation. If these benefits were better emphasized when recruiting African-American participants, African-American participation might rise. Additionally, if further reforms suggested by our participants were enacted, African-American participation might rise.

The reforms suggested by our participants are not impossible. In fact, one of the reforms suggested by our participants is easily adopted. Biomedical research teams can and should include more African-American researchers and physicians. Greater participation by African-American investigators would be beneficial not just from an "affirmative action" standpoint. Greater participation by African-American investiga- tors may make African Americans more likely to par- ticipate in studies. A second reform- broader com- munity education-is somewhat more difficult. Members of the public are unlikely to read scientific journals and may not always recognize the links between biomedical research and medical practice. Biomedical researchers may want to make this con- nection clearer in their reporting and may wish to pres-

Figure 2. Decision Flow for African Americans in Deciding Whether to Participate or Not Participate in Biomedical Research

Do Recall the Tuskegee Study?

Is TSUS a Salient Referent? O If No

Is Racial Representation Necessary O If Yes to Good Science?

Is there a Personal Advantage o If Yes from My Participation?

Do Others' Needs Outweigh My Own Needs? o If Yes

Have Had a Positive Personal Experience If Yes with Biomedical Research?

Do Trust the Biomedical System in General? Yes

Do Not Participate Participate in New Study

ent their findings in more public forums, such as the newspaper and television news. Physicians may want to explain how their practice is informed by research and indicate how the public's participation in research has made these practices possible. By taking these steps, researchers and physicians may display a better face of biomedical research and raise the general awareness ofthe need for participation. Given the larg- er disconnect between research and medicine for African Americans than for European Americans, these explanations may need to be directed towards African-American patients in the medical office and the so-called "black media." The final reform-and the most difficult-is the claim that African Ameri- cans need to be more proactive in representing their concerns to the biomedical community. If such con- cerns are expressed to the biomedical community, it is the role of researchers to take these concerns seriously and adapt research programs.

When evaluating these concerns, it is important to consider the communities that articulate these con- cerns. The concerns articulated by our participants

1062 JOURNAL OF THE NATIONAL MEDICAL ASSOCIATION VOL. 96, NO. 8, AUGUST 2004

PUBLIC PERCEPTIONS OF BIOMEDICAL RESEARCH

about TSUS are representative of the participants in these groups. Although the majority of research indi- cates that African Americans are likely to share views of the biomedical system regardless of socioe- conomic status, educational level, geographical loca- tion, age, sex, or other factors,40'46 alternative con- cerns linked to TSUS may be named by participants from different social locations than were named by the participants in our focus groups. When attempt- ing to involve more African Americans in biomedical research, specific communities should be asked what their particular concerns are and what recommenda- tions these communities offer for making participa- tion more amenable. The African Americans who made recommendations in our study were from a slightly higher socioeconomic group than are many African Americans. African Americans from a lower socioeconomic group may offer different concerns linked to TSUS and recommendations for overcom- ing them. In addition, many of our participants knew community leaders, making it possible that our par- ticipants were more likely to be politically and cul- turally active. Although community leader contact is one of the most effective strategies for recruiting par- ticipants of color and participants who are economi- cally disadvantaged,65 participants who are more dis- connected from community leaders may view biomedical research beyond their ability to change and may be less likely to offer recommendations. Finally, our participants were all residents of the southeastern United States. It is possible that African Americans from Macon County, AL, the site of TSUS, would express more reluctance to participate in biomedical research, whereas African Americans from Chicago would express less reluctance because of geographic removal from the site ofTSUS. Geog- raphy may also matter to the kind of recommenda- tions that African Americans make for having more racially inclusive biomedical research, if such recom- mendations are made at all. Additional study that will offer additional avenues for representative and reli- able research is needed.

In investigating reasons for the relative lack ofpar- ticipation by African Americans in biomedical research, we also need to be careful about the assumptions that we make. These assumptions are not just about African Americans as a population, but our assumptions ofwhat African Americans believe. Ifwe assume that TSUS is a prohibitive barrier to African- American participation, then the ability to recruit greater numbers ofAfrican-American participants is already lost. If we take a deeper look at how the African-American public uses TSUS to understand biomedical research, however, opportunities for adapting to concerns exemplified by TSUS become available. The African-American public is not just a

population infected with fear of participation. The African-American public is a well of ideas for making research more representative and for enhancing African-American participation in research studies. By investigating their concerns as well as considering their suggestions, research can become more reflec- tive ofthe needs ofthe whole population.

ACKNOWLEDGEMENT This paper was supported by a grant from the

National Institutes of Health (HGO2191-02). The authors would like to thank Celeste Condit and the Health and Heritage Project research team at the University of Georgia and the Pennsylvania State University.

REFERENCES 1. Tuskegee Syphilis Study Ad Hoc Advisory Panel. Final report. Washington: U.S Department of Health, Education, and Welfare/Public Health Service; 1973. 2. Angell M. Investigators' responsibilities for human subjects in developing countries. N EngI J Med. 2000;342:967-969. 3. Quinn TC, Wawer MJ, Sewankambo N, et al. Viral load and heterosexual transmission of human immunodeficiency virus type 1. N Engl J Med. 2000; 342:921-929. 4. White RM. Unraveling the Tuskegee study of untreated syphilis. Arch Intem Med. 2000;1 60:585-598. 5. Vonderehr RA, Clark T, Wegner OC, et al. Untreated syphilis in the male Negro: a comparative study of treated and untreated cases. JAMA. 1936;107:856-860. 6. Schuman SH, Olansky S, Rivers E, et al. Untreated syphilis in the male Negro: background and current status of patients in the Tuskegee Study. J Chronic Dis. 1955;2:543-558. 7. Peters JJ, Peers JH, Olansky S, et al. Untreated syphilis in the male Negro: pathologic findings in syphilitic and nonsyphilitic patients. J Chronic Dis. 1955:1:127-148. 8. Olanksy S, Schuman SH, Peters JJ, et al. Untreated syphilis in the male Negro: 20 years of clinical observations of untreated syphilitic and presum- ably nonsyphilitic groups. J Chronic Dis. 1956;4:177-185. 9. Rockwell DH, Yobs AR, Moore MB. The Tuskegee Study of Untreated Syphilis: the 30th year of observation. Arch Intern Med. 1964;1 14:792-798. 10. Caldwell JG, Price EV, Schroeter AL. Aortic regurgitation in the Tuskegee Study of Untreated Syphilis. J Chronic Dis. 1973;26:187-194. 11. Heller J. Syphilis victims in U.S. study went untreated for 40 years. NY Times. 1972;Jul 26;A1,A8. 12. Syphilis experiment termed unjustified. NY Times. 1 973;Mar 29;A54. 13. H.E.W. will study syphilis project. NY Times. 1 972;Aug 25;A40. 14. Kennedy says 45 babies died in a test. NY Times. 1 972;Oct 1 2;A22. 15. Hicks N. Regulation urged in human testing. NY Times. 1 973;Mar 21 ;A30. 16. Thomas SB, Quinn SC. The Tuskegee syphilis study, 1932-1972: implica- tions for HIV education and AIDS risk education programs in the black community. Am J Public Health. 1991;81:1498-1505. 17. Gamble VN. Under the shadow of Tuskegee: African Americans and healthcare. Am J Public Health. 1997;87:1773-1778. 18. Coughlin SS, Etheredge GD, Metayer C, et al. Remember Tuskegee: public health student knowledge of the ethical significance of the Tuskegee Syphilis Study. Am J Prev Med. 1996;12:242-246. 19. Shavers VL, Lynch CF, Burmeister LF. Knowledge of the Tuskegee study and its impact on the willingness to participate in medical research studies. J NatI Med Assoc. 2000;92:563-572. 20. Freimuth VS, Quinn SC, Thomas SB, et al. African Americans' views on research and the Tuskegee syphilis study. Soc Sci Med. 2001;52:797-808. 21. Reisberg D. Cognition: exploring the science of the mind, 2nd ed. New

JOURNAL OF THE NATIONAL MEDICAL ASSOCIATION VOL. 96, NO. 8, AUGUST 2004 1063

PUBLIC PERCEPTIONS OF BIOMEDICAL RESEARCH

York: W.W. Norton; 2001. 22. Sternberg RJ. Cognitive psychology. Florence, KY: International Thom- son Publishing; 1997. 23. Bates BR, Templeton A, Achter PJ, et al. What does "a gene for heart disease" mean?: a focus group study of public understandings of genetic risk factors. Am J Med Genet. 2003;1 1 9A:1 56-16 1. 24. Condit CM, Templeton A, Bates BR, et al. Attitudinal baniers to delivery of race-targeted pharmacogenomics among informed lay persons. Genet Med. 2003;5:385-392. 25. Parroff RL, Silk K, Condit CM. Diversity in lay perceptions of the sources of human traits: genes, environments, and personal behaviors. Soc Sci Med. 2003;1099-1 109. 26. Corbie-Smith G. The continuing legacy of the Tuskegee syphilis study: considerations for clinical investigation. Am J Med Sci. 1999;317:5-8. 27. Corbie-Smith G. Tuskegee as metaphor. Science. 1999;285:48. 28. Francis CK. Medical ethics and social responsibility in clinical medicine. J Urban Health. 2001;78:29-45. 29. Kirp DL. Blood, sweat, and tears: the Tuskegee experment and the era of AIDS. Tikkun. 1995;10:50-52. 30. Fairchild AL, Bayer R. Uses and abuses of Tuskegee. Science. 1999;284: 919-921. 31. Gamble VN. The Tuskegee syphilis study and women's health. J Am Women's Med Assoc. 1997;52:195-196. 32. Roy B. The Tuskegee syphilis experiment: biotechnology and the admin- istrative state. J Natl Med Assoc. 1995;87:56-67. 33. Jones JH. The Tuskegee legacy: AIDS and the black community. Hast- ings Cent Rep. 1992;22:38-40. 34. Schmidt TA. The legacy of the Tuskegee syphilis experiments for emer- gency exception from informed consent. Ann Emerg Med. 2003;41:79-81. 35. Reverby SM. Tuskegee: could it happen again? Postgraduate Med J. 2001 ;77:553-554. 36. Talone P. Establishing trust after Tuskegee. Int J Radiat Onc Biol Phys. 1 998;40:3-4. 37. Ex-chief defends syphilis project. NY Times. 1 972;Jul 28;A29. 38. Quinn S. Belief in AIDS as a form of genocide: implications for HIV pre- vention programs for Afrcan Amercans. J Health Educ. 1997;28:S6-S1 1. 39. Green B, Maisiak R, Wang M, et al. Participation in health education, health promotion, and health research by African Americans: effects of the Tuskegee syphilis experment. J Health Educ. 1997;28:196-201. 40. Smedley BD, Stith AY, Nelson AR. Unequal treatment: confronting racial and ethnic disparities in healthcare. Washington: National Academies Press; 2003. 41. Fouad MN, Partridge E, Wynn T, et al. Statewide Tuskegee alliance for clinical trials: a community coalition to enhance minority participation in medical research. Cancer. 2001;91 :S237-S241. 42. Cox D. Paternalism, informed consent, and Tuskegee. Int J Radiat Onc Biol Phys. 1998;40:1-2. 43. El-Sadr W, Capps L. The challenge of minority recruitment in clinical tials for AIDS. JAMA. 1992;267:954-957. 44. Swanson G, Ward A. Recruiting minorities into clinical trials: toward a participant-friendly system. J Natl Cancer Inst. 1995;87:1747-1759. 45. Turner P. heard it through the grapevine: rumor in African-American culture. Berkeley: University of California Press, 1994. 46. Race, ethnicity, and medical care: a survey of public perceptions and experiences. Kaiser Family Foundation: Menlo Park, CA, October 1999. 47. Grunfeld GB. Dissimilarties between Tuskegee study and HIV/AIDS pro- grams emphasized. Am J Public Health. 1992;82:1176. 48. White RM. Sociocultural issues in clinical research: unraveling the Tuskegee syphilis studies. Arthritis Rheum. 2002;47:457-458.

49. Farre M, Bosch F, Roset PN, et al. Putting clinical pharmacology in con- text: the use of popular movies. J Clin Pharmacol. 2004;44:30-36. 50. Banks A, Banks SP. Fiction and social research: by ice or fire. Walnut Creek: AltaMira; 1998. 51. Dominick J, Wimmer, R. Training the next generation of media researchers. Mass Communication and Society. 2003;6:3-9. 52. Dobbins IG, Kroll NEA, Yonelinas AP, et al. Distinctiveness in recognition and free recall: the role of recollection in the rejection of the familiar. Jour- nal of Memory and Language. 1998;38:381-400. 53. Lum PH, Boone KB, Cozolino L, et al. Effectiveness of the Rey-Osterreth complex figure test and the Meyers and Meyers recognition trial in the detection of suspect effort. Clin Neuropsychol. 2003;17:426-440. 54. Andsager JL, Austin EW, Pinkleton BE. Questioning the value of realism: young adults' process of messages in alcohol-related public service announcements and advertising. Joumal of Communication. 2001;51:121 - 142. 55. Carspecken P, Apple M. Crtical qualitative research: theory, methodol- ogy, and practice. In: LeCompte MD, Millroy WL, Preissle J, eds. The hand- book of qualitative research in education. San Diego: Academic Press, 1992:507-554. 56. Cunningham-Burley S, Kerr A, Pavis S. Theorizing subjects and subject matter in focus group research. In: Barbour RS, Kitzinger J, eds. Developing focus group research: politics, theory and practice. Thousand Oaks: Sage, 1999:186-199. 57. Myers G, Macnaughten P. Can focus groups be analyzed as talk? In: Barbour RS, Kitzinger J, eds. Developing focus group research: politics, the- ory and practice. Thousand Oaks: Sage, 1999:173-185. 58. Glaser BG, Strauss AL. The discovery of grounded theory and applying grounded theory. In: Denzin NK, Lincoln YS, eds. The American tradition in qualitative research. V.2. London: Sage, 2001:229-243. 59. Charmaz K. Grounded theory. In: Denzin NK, Lincoln YS, eds. The Amer- can tradition in qualitative research. V.2. London: Sage, 2001:244-270. 60. Bevan JL, Lynch JA, Dubriwny TN, et al. Informed lay preferences for delivery of racially varied pharmacogenomics. Genet Med. 2003;5:393-399. 61. Cooper-Patrick L, Gallo JJ, Gonzalez HE, et al. Race, gender, and part- nership in the patient-physician relationship. JAMA. 1999;282:583-589. 62. Chen J, Rathore SS, Radford MJ, et al. Racial differences in the use of cardiac catheterization after acute myocardial infarction. N EngI J Med. 2001;344:1443-1449. 63. Noah BA. The participation of under-represented minorities in clinical research. Am J Law Med. 2003;29:221-245. 64. King Jr TE. Racial disparities in clinical trials. N Engl J Med. 2002;346:1400- 1402. 65. Madriz E. Focus groups in feminist research. In: Denzin NK, Lincoln YS, eds. The American tradition in qualitative research. V.3. London: Sage, 2001:352-372.0

We Welcome Your Comments The Journal of the National Medical Association welcomes your Letters to the Editor about arti- cles that appear in the JNMA or issues relevant

to minority healthcare. Address correspondence to [email protected].

1064 JOURNAL OF THE NATIONAL MEDICAL ASSOCIATION VOL. 96, NO. 8, AUGUST 2004