|
 |
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
© 2006 American Public Health Association
DOI: 10.2105/AJPH.2004.043471
ATTAINING HEALTH IN AN INEQUITABLE WORLD |
Both authors are with Family Health International, Durham, NC.
Correspondence: Requests for reprints should be sent to Elizabeth E. Tolley, Family Health International, 2224 East NC 54, Durham, NC 27713 (email: btolley{at}fhi.org).
 |
ABSTRACT |
|---|
TOP
ABSTRACT
INTRODUCTIONIt has been argued that rigid thinking about the types and progression of research needed to evaluate health promotion interventions has stymied the process by which research is translated to action. This argument is particularly salient in the field of HIV/AIDS prevention.
We examined microbicide research and identified challenges that obstruct the integration of clinical trial and behavioral and social science research, thereby reinforcing linear programs of research. We found that behavioral and social science research can both support microbicide clinical trial performance and anticipate the information most needed for a rapid and successful introduction of future microbicide products.
|
INTRODUCTION |
|---|
IN THE AUGUST 2003 ISSUE of the Journal, Glascow et al. argued for a more integrated approach to the development of behavioral interventions rather than the linear process of efficacy to effectiveness and the postmarketing research favored by the pharmaceutical industry.1 We concur and have extended their argument to the field of microbicide development by considering both the challenges and the opportunities of a more integrated program of research.
As researchers enter their third decade of searching for an effective means of preventing HIV transmission, topical microbicides—gels, films, or other substances that can be self-applied vaginally or rectally—have shown great potential. While no topical microbicide has yet been proven effective, these products are appealing for several reasons. First, they are one of the few technologies that can be initiated by women—the fastest growing population of HIV-infected individuals in many parts of the world2—and many believe microbicide use will require less active involvement from a male partner than male or female condoms do.3 Second, topical microbicides may reach the market more quickly than other HIV prevention methods that are in development.4 Third, microbicide products could potentially have additional properties that are desirable to users, such as being a contraceptive or noncontraceptive and having lubricating characteristics.3 Finally, because microbicides are designed as a temporary rather than a systemic barrier to HIV, some potential users may consider microbicides to be a more flexible prevention option than a vaccine would be. One of the many challenges for microbicide researchers has been determining whether candidate products effectively prevent transmission of HIV. This seemingly straightforward requirement is, in fact, rather complex. Some researchers believe the biological ability of a first-generation microbicide product to prevent transmission when used perfectly (i.e., its efficacy) may be quite low when compared with condoms. However, others argue that even with lower efficacy rates, the product’s effectiveness—its ability to prevent HIV transmission during everyday circumstances—may be higher than condoms. Such a paradox could happen if a less efficacious microbicide product were used more consistently and by more people than condoms.5,6 It is clear that effectiveness will ultimately be influenced by women’s (or men’s) willingness and ability to use microbicides consistently and correctly with high-risk partners. In turn, the choice to initiate and to sustain use (i.e., its acceptability) will undoubtedly be influenced by a range of other factors, including how risky a particular sexual partner is perceived to be or how sexually disruptive or enhancing the product is.7 Behavioral and social science (BSS) research that addresses these issues could therefore play a crucial role in determining overall effectiveness.
Some BSS research has been undertaken since the earliest Phase I microbicide clinical trials began. Data collection has included structured questions for assessing microbicide and/or condom use, vaginal cleaning practices, and overall appreciation of the gel or other specific product attributes. Some Phase I trials have incorporated focus group discussions with women who have completed the trial to better understand their experiences with both gel use and participation in a research study.8–10 However, plans for using BSS methods during the Phase II/III trials to examine longer-term acceptability and use patterns have been limited. Whereas some have called for a better incorporation of social science into biomedical research processes,8,11,12 there have been numerous barriers to integrating BSS research and clinical trials research. We examined various perspectives and challenges that have obstructed this integration.
|
CHALLENGE 1: AN EFFICACIOUS PRODUCT WILL BE USED IF IT IS AVAILABLE |
|---|
This perspective stems in part from a belief that "risk factors associated with HIV transmission are well defined, and education and behavioral modification programs proven effective."13(p459) However, we can look at condom use to see that this is a myth. Male condoms, when used consistently and correctly, offer upward of 95% protection from both pregnancy and HIV.14 However, despite high efficacy, worldwide condom use is low, even within populations most at risk for HIV.15,16
Social science literature on the determinants of condom use provides important clues into the kinds of mediators that will influence the uptake and the effectiveness of an eventual microbicide product. For example, cultural context shapes an individual’s or a couple’s perceptions of HIV risk, which can lead married women or those in stable relationships to ignore their partner’s behavior as an important source of HIV risk.15–21 Service delivery policies and provider attitudes may determine whether an unmarried sexually active adolescent will have access to HIV-related information and services, such as condom provision. Additional mediators mitigate the effect of HIV risk perception on condom uptake. For example, as condom use has become associated with illicit sexual behavior, even those with knowledge about and a desire to use condoms may be unable to do so. Thus, even sex workers with a known risk for HIV have been unwilling or unable to use condoms with their husbands or romantic partners.22,23 Also, the immediate circumstances of a sexual encounter will often determine whether or not an available condom is used.22,24
|
CHALLENGE 2: CLINICAL TRIALS DIFFER FROM SERVICE DELIVERY INTERVENTIONS |
|---|
Stein et al. have shown that trial participants will receive the intervention generally under more rigorous and demanding circumstances than they would in a normal service delivery setting.25 For example, trial participants will likely receive more information about the study product and how to use it than would family planning clients or pharmacy customers if a microbicide became widely available. They also will have greater access to tests for HIV, sexually transmitted diseases (STD), and other conditions than would those outside a trial. Sexual partners will be more involved in discussions about trial participation than they would be about everyday use of a microbicide product. Because of this, some might argue that little can be learned about use behavior within a randomized controlled trial.
However, not all clinical trials diverge significantly from the context of individuals’ everyday health services. Trial managers can design the trial to reflect standard care in terms of the types and the amount of information provided to participants (Robinson J, FIBS, oral communication, April 2004). In such situations, lessons drawn from BSS research may be directly transferable to posttrial settings. Even within the most carefully planned and highly controlled trial settings, a BSS research component that focuses on the clinical trial context could provide valuable information for trial implementation and future product introduction. BSS questions can examine how well participants understand written or oral instructions on product use and how the delivery of such instructions and ongoing access to providers can contribute to correct use. For example, misunderstandings associated with the timing of gel application or the quantity of gel necessary could then be rectified to improve adherence. Insight into the relative importance of counseling versus experience via trial-and-error will provide important clues about the kinds of information or counseling that can support more consistent use outside of a trial setting.
One clear difference between a clinical trial and the everyday provision of microbicides is how providers discuss, and users comprehend, messages about product efficacy. Early microbicide acceptability research suggests that use is positively influenced by beliefs about product effectiveness. For example, our unpublished BSS research suggests that some women who have multiple partners may selectively use the study gel, i.e., they will use condoms only with clients, and they will use gel only with steady partners. Strategically targeted BSS research that examines the association between participants’ perceptions of efficacy and product use will alert trial administrators to any false understanding about efficacy and, thus, enable them to address problems with adherence.
|
CHALLENGE 3: TRIAL PARTICIPANTS DIFFER FROM EVENTUAL USERS |
|---|
Supporters of this argument believe that integration is inappropriate, because individuals involved in clinical trial research are generally not like those to whom the product will eventually be marketed. For example, individuals may be from lower-risk populations (if Phase I), higher-risk populations (if Phase II/III), or be more likely to use healthcare services than the population for whom microbicides might eventually be destined. Therefore, because generalization is not possible, the resources used to conduct BSS research are wasted.
Clearly, people have numerous reasons for participating or not participating in research.26 With microbicide research, motivating factors are likely to include perception of personal risk for HIV/ AIDS, desire to contribute to society, financial gain, concerns about the experimental nature of the research, concerns about potential side effects, and fear of being associated with a stigmatized disease. Clinical trial administrators rely on randomization to distribute the confounding influence of individual characteristics across study arms. However, from a BSS perspective, examining individual motivations and concerns, couple dynamics, and broader sociocultural factors has direct application to clinical trial recruitment, retention, and adherence and to the acceptability and the introduction of an eventual microbicide product.
Consider an example from West Africa, where one of the authors recently traveled to assist sites with preparing for a Phase III microbicide trial. The trial intends to recruit women who have sexual relations with multiple partners and who are at risk for HIV. The experiences of local researchers suggest that the population of high-risk women is quite heterogeneous. They include self-identified sex workers—many of whom originated from neighboring countries—and numerous categories of local women who engage in clandestine high-risk sexual behaviors but do not consider themselves to be sex workers. Past research at those sites suggests that formal sex workers will be easy to recruit,26 because they acknowledge their personal risk for HIV. Many have previously participated in HIV prevention trials, and many already use a vaginal gel to reduce the pain and discomfort of high-frequency sexual behavior and to prevent condom breakage. Because many of these women have experience with the dual use of condoms and gel, we can assume that adherence to both gel and condom use during the trial—and acceptance of an eventual gel product—will be good. However, past research also shows that these women are highly mobile and, therefore, difficult to retain during trials.27 In contrast, local women who engage in clandestine or informal sex work might be more difficult to identify and recruit, but once they are, they should be easier to retain.
Evidence of low levels of condom use among these groups suggests that they might benefit greatly from the development of an effective microbicide product. During the trip, a peer educator from the informal sex worker community identified numerous factors that could inhibit the willingness of these local women to participate in a microbicide trial. These factors included concerns about being associated with sex work, low perceived risk for HIV, fear of side effects, and negative attitudes toward a vaginal gel. BSS research could help clarify potential participants’ perceived benefits and concerns about the trial and develop messages and processes aimed at identifying and recruiting participants. Therefore, finding ways to include local women who have multiple sexual partners in the microbicide trial would not only benefit the trial but also inform strategies for reaching an important group of potential users postmarketing.
|
CHALLENGE 4: EFFICACY MUST BE PROVEN BEFORE RESEARCH IS CONDUCTED |
|---|
This challenge is about timing. Those who support a linear program of research may agree that additional research is needed to determine the most effective means for making the drug available, but they believe it should be conducted only after efficacy has been proven.28
However, experience in the field of family planning suggests that BSS research is often conducted only after a less-successful-than-expected introduction of the drug rather than as a way of guiding introduction. For example, the IUD was introduced in India without the benefit of empirical research to determine issues about introduction, and Norplant went through a stage of introductory trials that were fashioned on the clinical trials preceding them.29 As Simmons noted, a focus on the technology rather than the broader service delivery system and sociocultural context contributed to the failure of both these programs. Similarly, Potter said, "Historical events may result in contraceptive practices and service delivery systems that are far from optimal." 30(p732) Clinical trials are likely to construct administrative systems (e.g., types of information provided to clients, the number and types of required follow-up visits, and resupply regimens) that are often retained long after trials have ended, without fully understanding the influence of those systems on the acceptability and use of the product or method. Because there is an urgency to slow the HIV epidemic, the pressure to move from efficacy to wide-scale introduction will be great. Strategies for microbicide introduction should be based on an understanding of the psychosocial, cultural, and service delivery factors that will affect demand for and use of the products.7,31
|
CHALLENGE 5: BSS DATA ARE SOFT, SUBJECTIVE, AND CONSEQUENTLY UNRELIABLE |
|---|
This perspective stems in part from the idea that human behavior is messy and difficult to measure. Indeed, behavior is driven by the subjective cognitions and emotions of individuals who are in turn influenced by the social systems and cultures they inhabit. However, whereas the content of BSS research necessarily encompasses the subjective, its methods need not be wholly subjective. It is also relevant to note that social scientists do not always do a good job explaining, and sometimes adhering to, the rigorous standards that are possible with BSS research. We provide examples of behavioral measures and methods that respond to the needs of microbicide clinical trials in the following paragraphs.
At a minimum, microbicide effectiveness trials must be able to accurately assess delivery of the intervention (i.e., gel use) and disease exposure in order to attribute any differences between study arms in HIV or STD rates to the effect of the study product rather than to other reasons. Typically, clinical trial researchers attempt to obtain this information by requesting participants to report verbally or in diaries the number of sexual encounters in which they used the gel product alone, the gel and condoms together, condoms only, or neither. Such questions may require complex mental calculations, especially for participants who engage in high-frequency sexual relations or who have multiple partners. The use of cognitive testing or "think aloud" processes during the questionnaire design phase can both rationalize the sequencing and clarify the wording of questions so that they correspond more closely to the way participants derive their responses. For a woman engaged in sex work, mental calculations for determining the number of sexual acts engaged in during the last month for which they were protected by both gel and condoms might include (1) the number of clients during the last day of work, (2) whether that number was typical or not of the last week of work, (3) an estimation of a typical week times 4 weeks, (4) subtracting any period during the last month that she was sick or traveled, (5) whether there were any clients with whom she did not use a condom, and (6) whether there were any times she did not use the gel with a client. Unless prompted, she might not include the number of times she had sexual relations with a husband or romatic partners and whether condoms and/ or gel were used.
Rather than obtain use behavior on a per-act basis, it might be sufficient for trial purposes to group participants into more general categories of use and exposure. Usually, this approach entails asking participants to identify their behavior as always, sometimes, or never using condoms, gels, or both during sexual relations. Compared with this single-variable approach, psychometric scales could provide a more refined estimate of product compliance. A scale is composed of multiple items that, when combined, measure a single concept or variable.32–34 The redundancy of items increases the precision of an estimate in much the same way that the redundancy of participants does in survey research (DeVillis R, PhD, oral communication, May 2004). At present, we know of no standardized adherence scales that could be integrated into microbicide trials; however, the development of such a scale would be a great contribution to the field.
One disadvantage to using highly structured questions for assessing acceptability and use behavior is that such questions may lack relevance for participants. In contrast, semistructured interviewing techniques have the advantage of increasing salience, because participants are encouraged to describe a set of behaviors in context and discuss in-depth only those topics that are of personal concern.35,36 The process of recounting a behavioral event may both jog a participant’s memory (about whether she used a condom or gel with a particular partner or why she did not) and alert an interviewer to inconsistencies that can be further explored and resolved. When supplemented with more structured data, such qualitative information could be used to refine categories for gel adherence or HIV exposure. However, when using qualitative approaches, it is essential to identify and document the analytic steps taken to ensure that findings accurately represent the attitudes and the behaviors of participants rather than the researcher.36–38
|
CHALLENGE 6: BSS RESEARCH BURDENS CLINICAL TRIAL PARTICIPANTS AND STAFF |
|---|
There is no doubt that clinical trial research places tremendous demands on all who are involved. Participants are required to rearrange their lives in multiple ways and then undergo procedures that are, at times, unpleasant or even risky. Trial managers must develop and maintain integrated systems for tracking participants, products, and data that adhere to the highest ethical and quality standards. Thus, it is little wonder that some may perceive the addition of BSS as overly burdensome. Although such concerns are not unwarranted, we suggest that well-integrated and carefully targeted BSS research will not only assist the clinical trial team in implementing the trial but also help with the translation of research findings to practical application.
When evaluating the usefulness of a BSS component, the clinical trial team should consider how confident they are in their ability to identify and recruit appropriate participants, whether they anticipate ethical concerns arising from community groups or participants themselves, what kinds of adherence and use data are required, and how aspects of the clinical trial or the broader social context might influence adherence and use.
Once areas for collaboration have been identified, a multidisciplinary team should determine the appropriate timing of, and approach to, BSS data collection and analysis. To date, many BSS assessments appended to microbicide clinical trials have recruited individuals who were no longer participating in the clinical trial.9,10 Nonparticipants also may be recruited for BSS assessments during other phases of the clinical trial. For example, information collected from nonrecruited individuals during the screening phase can provide insight into the characteristics of potential acceptors. Both the nonrecruited and those who discontinue early may provide important information about barriers to microbicide use.
Nevertheless, some BSS questions are best assessed when the individual is actually using the product, i.e., during the clinical trial itself. We believe it is possible to add such questions without negatively affecting the trial. First, the BSS should consider clinical trial demands on participants’ time when designing its protocol. For example, it is possible to schedule BSS interviews so that they do not coincide with particularly intensive phases of the clinical trial. During an upcoming microbicide clinical trial, limited baseline BSS data will be collected at enrollment; BSS follow-up will not occur until trial participants have concluded their third month of trial participation, before which time they are required to make monthly rather than quarterly visits to undergo a number of physical examinations. Once BSS follow-up begins, data collectors will work with trial participants to identify convenient dates, times, and locations for future follow-up interviews.
Second, coordination between the clinical trial and BSS assessment will ensure that there is no duplication of key questions associated with use behavior. For example, if the clinical trial measures variables associated with a BSS outcome (i.e., number of sexual acts protected by microbicide use), the BSS assessment will limit its assessment to measuring covariates of microbicide use (i.e., scaled measures of HIV risk perception, attitudes toward product attributes, and trial adherence motivation) or to qualitatively investigating participants’ experiences with microbicides during the trial.
Finally, the choice of data collection method may influence participant burden. In general, structured questions, including the use of psychometric scales, tend to be less time-consuming. However, trial participants may welcome opportunities to talk about their experiences in a less structured format rather than in a formal format used by qualitative researchers. Participants in an HIV vaccine efficacy trial who took part in a qualitative study of their experiences with risk reduction counseling associated feelings about trial participation with counselor qualities. The kind of personal rapport and conversational quality adopted by trained qualitative interviewers was found to create overall positive experiences with trial participation.39 Nevertheless, such techniques do demand special expertise in data collection and analysis.
|
CONCLUSIONS |
|---|
There are many challenges to expanding the role of BSS research within the clinical trial context, but there are highly warranted grounds for doing so. As the size and the pace of the epidemic increase in some regions of the world, the character and the context of those at risk for HIV continue to change. The success of risk reduction products or strategies, including microbicides, will depend in part on our ability to incorporate them into public health programs that reach these different groups and empower them to use products effectively. We are not suggesting that every clinical trial of a health promotion product or strategy be accompanied by a BSS research component. However, we hope that researchers will conduct their own cost–benefit analyses and weigh the burden that might ensue against the additional information derived from expanding clinical trial research to incorporate unanswered questions associated with the contexts and determinants of intervention acceptability or use.
|
Acknowledgments |
|---|
We were supported by the US Agency for International Development. We thank Janet Robinson for her insightful comments and suggestions about the impact of integration on regulatory submissions to the Food and Drug Agency. We also thank Theresa Hatzell and Vera Grigorieva for their thoughtful reviews of earlier drafts. Finally, we appreciate the careful review and comments of the anonymous reviewers and Mary Northridge.
Human Participation Protection
No protocol approval was needed for this study.
|
Footnotes |
|---|
Peer Reviewed
Contributors
Both authors originated the article and determined its content. E. E. Tolley had primary responsibility for writing the article.
Accepted for publication February 25, 2005.
|
References |
|---|
1. Glasgow RE, Lightenstein E, Marcus AC. Why don’t we see more translation of health promotion research to practice? Rethinking the efficacy-to-effectiveness transition. Am J Public Health. 2003;93: 1261–1267.
2. UNAIDS. Women, Girls, HIV and AIDS—World AIDS Campaign 2004. Available at: http://www.unaids.org/wac_2004/index_en.htm. Accessed October 6, 2005.
3. Boonstra H. Campaign to accelerate microbicide development for STD prevention gets under way. Guttmacher Rep Public Pol. 2000;Feb:3–5.
4. Kresge K. Agreeing to disagree: the future of microbicides. HIV/AIDS Treatment Insider. 2003;4:1–6.
5. Speiler J. Behavior Issues in Dual Protection. Washington, DC: Population Association of America; 1997.
6. Stone A. Clinical Trials of Microbicides. Microbicide Q. 2003;1:13–18.
7. Elias C, Coggins C. Acceptability research on female-controlled barrier methods to prevent heterosexual transmission of HIV: Where have we been? Where are we going? J Women’s Health Gender-Based Med. 2001;10:163–173.
8. Bentley ME, Fullem AM, Tolley EE, et al. Acceptability of a microbicide among women and their partners in a 4-country Phase I trial. Am J Public Health. 2004;94:1159–1164.
9. Bentley ME, Morrow KM, Fullem A, et al. Acceptability of a novel vaginal microbicide during a safety trial among low-risk women. Fam Plann Perspect. 2000;32:184–188.
10. Morrow K, Rosen R, Richter L, et al. The acceptability of an investigational vaginal microbicide, PRO 2000 Gel, among women in a phase I clinical trial. J Women’s Health. 2003;12: 655–666.
11. Auerbach JD, Coates TJ. HIV prevention research: accomplishments and challenges for the third decade of AIDS. Am J Public Health. 2000;90: 1029–1032.
12. Bachrach CA, Abeles RP. Social science and health research: growth at the National Institutes of Health. Am J Public Health. 2004;94:22–28.
13. Folkers GK, Fauci AS. The AIDS research model: implications for other infectious diseases of global health importance. JAMA. 2001;286:458–461.
14. Warner DL, Hatcher RA. Male condoms. In: Hatcher RA, Trussell J, Stewart F, et al., eds. Contraceptive Technology. 17th ed. New York, NY: Ardent Media Inc; 1998:325–355.
15. Steiner MJ, Cates WJr, Warner L. The real problem with male condoms is nonuse. Sex Transm Dis. 1999;26: 459–462.
16. Gardner R, Blackburn RD, Upadhyay UD. Closing the Condom Gap. Baltimore, Md: Johns Hopkins University School of Public Health, Population Information Program; 1999.
17. Misovich SJ, Fisher JD, Fisher WA. Close relationships and elevated HIV risk behavior: evidence and possible underlying psychological processes. Rev General Psychol. 1997;1:72–107.
18. Castaneda D. The close relationship context and HIV/AIDS risk reduction among Mexican Americans. Sex Roles. 2000;42:551–579.
19. Worth D. Sexual decision-making and AIDS: why condom promotion among vulnerable women is likely to fail. Stud Fam Plann. 1989;20:297–307.
20. de Zoysa I, Sweat MD, Denison JA. Faithful but fearful: reducing HIV transmission in stable relationships. AIDS. 1996;10(suppl A):S197–S203.
21. Heise LL, Elias C. Transforming AIDS prevention to meet women’s needs: a focus on developing countries. Soc Sci Med. 1995;40:931–943.
22. Varga C. The condom conundrum: barriers to condom use among commercial sex workers in Durban, South Africa. Afr J Reprod Health. 1997;1:74–88.
23. Ortiz-Torres B, Williams SP, Ehrhardt A. Urban women’s gender scripts: implications for HIV prevention. Culture, Health Sexual. 2003;5:1–17.
24. Beckman LJ, Harvey M. Factors affecting the consistent use of barrier methods of contraception. Obstet Gynecol. 1996;88:65S–71S.
25. Stein ZA, Myer L, Susser M. The design of prophylactic trials for HIV: the case of microbicides. Epidemiology. 2002;14:80–83.
26. Pedhazur EJ, Schmelkin LP. Artifacts and pitfalls in research. In: Pedhazur ED, Schmelkin LP, eds. Measurement, Design, and analysis: An Integrated Approach. Hillsdale, NJ: Lawrence Erlbaum Associates; 1991:234–241.
27. Alary M, Mukenge-Tshibaka L, Bernier F, et al. Decline in the prevalence of HIV and sexually transmitted diseases among female sex workers in Cotonou, Benin, 1993–1999. AIDS. 2002;16:463–470.
28. Folkers GK, Fauci AS. AIDS Agenda Still Daunting. Available at: http://www.nap.edu/issues/19.4/folkers.html. Accessed August 26, 2003.
29. Simmons R, Hall P, Diaz J, Diaz M, Fajans P, Satia J. The strategic approach to contraceptive introduction. Stud Fam Plann. 1997;28:79–94.
30. Potter JE. The persistence of outmoded contraceptive regimes: the cases of Mexico and Brazil. Popul Dev Rev. 1999;25:703–739.
31. Rao Gupta G, Weiss E. Women’s lives and sex: implications for AIDS prevention. Cult Med Psychiatry. 1993;17: 399–412.
32. Bernard HR; Social Research Methods Qualitative and Quantitative Approaches. Thousand Oaks, Calif: Sage; 2000.
33. Mahoney CA, Thombs DL, Howe CZ. The art and science of scale development in health education research. Health Educ Res. 1995;10:1–10.
34. DeVellis R. Scale Development Theory and Applications. Newbury Park, Calif: Sage; 1991.
35. Fitzpatrick R, Davey C, Buxton MJ, Jones DR. Evaluating patient-based outcome measures for use in clinical trials. Health Technol Assess. 1998;2: 1–73.
36. Silverman D. The quality of qualitative health research: the open-ended interview and its alternatives. Soc Sci Health. 1998;4:104–118.
37. Ulin PR, Robinson ET, Tolley EE; Qualitative Methods in Public Health: A Field Guide for Applied Research. Indianapolis, Ind: Jossey-Bass; 2005.
38. Lincoln Y, Guba E. Establishing trustworthiness. In: Lincoln Y, Guba E, eds. Naturalistic Inquiry. Beverly Hills, Calif: Sage; 1985:397–444.
39. Guest G, McLellan E, Matia D, et al. HIV vaccine efficacy trial participation: men-who-have-sex-with-men’s experiences of risk reduction counseling and perceptions of risk behavior change. AIDS Care. 2005;17:46–57.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |