Researchers appreciate the need to understand the social and cultural factors related to health and illness in diverse settings because of the direct and indirect impact of these factors on exposure and vulnerability to disease, risk-taking and health promotion behaviors, and access to quality healthcare . Ethnographic methods, including field observations, in-depth interviews, focus groups, and social mapping, provide an important set of tools for understanding the contextual features that elicit behaviors that are protective or damaging. Rapid ethnographic assessment is regularly applied in behavioral intervention research and programme evaluation [2–6]. In-depth interviews are particularly useful for exploring sensitive topics, such as sexual practices, sexual norms, and HIV-related risk behaviors, and providing explanations for quantitative findings [7,8].
To ensure that the intervention, the community popular opinion leader (C-POL), and assessment measures were culturally appropriate and acceptable, a rapid ethnographic assessment  was conducted before final protocol development for this five-country (China, India, Peru, Russia, and Zimbabwe) study. Using a standard set of methods, with operational definitions specified in the protocol, each research site collected data on the following core topic areas: (i) sexual health data; (ii) sexual practices and meanings; (iii) healthcare delivery and beliefs; (iv) social groups; (v) characteristics of C-POL (natural leaders in a community); (vi) appropriate incentives; (vii) prevention messages; and (viii) population profiles. See ‘Design and integration of ethnography within an international behavior change HIV/STD prevention trial’  for a description of the core topic areas, the required elements within each core topic area for this study, and the suggested probes. Whereas the research teams in all five countries were required to adhere to the protocol, the strength of the approach was the flexibility in how each country would meet these expectations. Each team collected data on all core topic areas and implemented the methods according to the stated procedures, but each team used an iteractive process in applying the methods and tailoring the appropriate questions based on the demands of each sociocultural context and in-vivo field experience about how best to access the required data.
The culture, language, setting, and risk populations were different across the five countries, and therefore it was especially important to adhere to the procedures meticulously described in the protocol. This is an unusual and challenging approach because most studies that use rapid ethnographic assessment work in a single site with a specific population.
In this paper, we briefly discuss the rapid ethnographic assessment conducted in the five countries, the preliminary selection and description of venues and high-risk populations in each country, and the findings on two core topic areas from the rapid ethnographic assessment to demonstrate the relevance of ethnographic findings to developing effective HIV/sexually transmitted disease (STD) prevention messages, selecting appropriate C-POLs, and tailoring assessment instruments.
Rapid ethnographic assessment
Although a full discussion of the ethnographic methods used in this multisite, multipopulation Trial is beyond the scope of this paper, we will summarize the methods used to collect the data upon which this paper is based. Ethnographers conducted field observations and social mapping within the venues. In-depth interviews and focus groups were conducted in the local language (Mandarin and Fuzhou dialect in China, Tamil in India, Spanish in Peru, Russian in Russia, and Shona or Ndebele in Zimbabwe), audio tape recorded, transcribed, and translated into English as necessary for analysis. Site-specific ethnography teams reviewed all data for emergent themes and patterns. To facilitate cross-site analysis, each site was required to code transcripts using the core topic areas and to sub-code using the required elements within each core topic area, organizing the data with a pre-approved qualitative analysis software package (e.g. NUDIST, nVivo, ATLAS.ti, EthnoNotes). Each site prepared an executive summary and a full report triangulating the data from field observations, in-depth interviews, focus groups, and social mapping by core topic area, and including specific quotations from in-depth interviews and focus groups. The scientific teams used these reports to tailor various aspects of the Trial at each site according to protocol guidelines. These reports provide the basis for this paper. For a more complete discussion of the protocol for the rapid ethnographic assessment, please see ‘Design and integration of ethnography within an international behavior change HIV/sexually transmitted disease prevention trial’ .
We focused on the two core topic areas selected for discussion in this paper, sexual practices and meanings and the characteristics of C-POLs, because they were particularly critical to the process of designing and tailoring the intervention and assessment. The C-POL intervention consists of training natural leaders in the community, who are known as ‘community popular opinion leaders’ (C-POLs), to deliver HIV prevention messages to their peers and friends through regular conversations in the venues described below. The intervention model is defined by nine required core elements, which must be implemented at each site. The C-POLs are trained by facilitators, who are staff on the study, using a common manual developed by the study and tailored to the needs of each site. A critical part of the training is to help the C-POL develop and learn to deliver HIV-related prevention messages using language tailored to community sexual practices and meaning. Therefore, the success of the intervention relies on selecting appropriate C-POLs who are natural leaders, and on crafting salient and relevant HIV-related prevention messages that can be delivered to their friends and neighbors as part of daily conversation. The theory of diffusion , on which this intervention is based, posits that members of a community will adopt the recommendations and behaviors of their natural leaders, and this will result in a shift in the social norms regarding acceptable behavior. A baseline structured assessment, which included a personal interview on sexual risk taking and other health behaviors and tested for HIV/STDs, was administered to a randomly selected cohort from each venue. The intervention was then initiated in half of the venues in each country. Follow-up assessments were conducted 12 and 24 months after the baseline assessment. If the C-POLs have been effective in delivering the intervention, intervention venues should show a greater improvement in HIV-related risk behavior and the incidence of HIV/STD than comparison venues at the time of the 24-month assessment. The intervention is described elsewhere. See ‘Methodological overview of a five-country community-level HIV/sexually transmitted disease prevention trial’  for an overview of the methodology for all aspects of the study and ‘The community popular opinion leader HIV prevention programme: conceptual basis and intervention procedures’  to learn more about the intervention and how the intervention core elements were tailored across the five countries.
Venues and populations studied
One of the nine core elements of the C-POL intervention mentioned above is the selection of a venue that has clearly defined physical or social boundaries (e.g. housing project, bar, dormitory, etc.) with an identifiable target population where multiple informal conversations occur frequently . Within or near those venues, the target populations must engage in high-risk sexual behavior and have a high prevalence of sexually transmitted disease (STD). On the basis of the rapid ethnographic assessment in each country, venues and target populations that met the criteria specified in the protocol were selected to implement the National Institute for Mental Health (NIMH) Collaborative HIV/STD Prevention Trial (the Trial) . The appropriateness of the venues and populations were confirmed by epidemiological studies that collected both behavioral and biological data . We will briefly describe the venues and populations selected in the five countries to provide a context for the ethnographic data collected on the two core topic areas: (i) sexual practices and meaning; and (ii) characteristics of C-POLs.
In China, venues were food stalls in markets, each with between 50 and 150 independent owners and employees, who formed our study population. The ethnographic study included 43 in-depth interviews (22 women, 21 men) focused on broad aspects of risk behavior; another 52 semi-structured interviews (31 women, 21 men) focused on knowledge, attitudes, and beliefs about HIV/STDs; one focus group with market workers (four women, six men) covered a variety of intervention-specific issues; and six field observations of market activities. In-depth interview respondents included market workers, female sex workers employed near a market, STD clinic patients, one market manager, and one public hospital physician.
The original venues selected for study in India were slums (housing projects). ‘Slum’ is the local term for public housing units or a housing project in Chennai, India. Housing allotment, construction, and oversight are managed by a government agency called the Tamil Nadu Slum Clearance Board. The link to their web site is http://www.tn.gov.in/citizen/tnscb.htm. The name for these venues reflects the local term used by residents and government and is not intended to be pejorative. Because of the low HIV/STD risk in the overall population in these areas, the final venues selected for study were clusters of wine shops in and near these communities. The study populations in India are men who are regular customers of the selected wine shops and the high-risk women who frequent the area near these shops. Key informant interviews with community leaders and healthcare providers, field observations, and social mapping exercises were conducted to facilitate the recruitment and implementation of the ethnographic interviews and focus groups. A total of 62 in-depth interviews and 24 focus group discussions were conducted with married and unmarried men and women in several slum areas. In addition, staff made field observations to understand locations where communication and sexual liaisons occur.
The venues selected in three cities in Peru (Lima, Trujillo, and Chiclayo) were lower socioeconomic neighborhoods and the high-risk populations selected were young heterosexually identified men who do not work or study, young women with multiple male sexual partners, and self-identified homosexual men. The ethnographic study included field observation, social mapping, 108 in-depth interviews (59 heterosexually identified men, 40 women, and nine men self-identified as homosexual), 17 focus groups (eight groups of heterosexual men, seven of women, and two of homosexual men), and 70 semi-structured interviews with gatekeepers, stakeholders, and key informants who were respected leaders and members of the target population in the community.
Dormitories at vocational trade schools in Russia were selected as the venues, and the student residents were the high-risk study population. A total of six focus groups and 20 in-depth interviews were conducted with residents in these venues. All participants were students living in the selected dormitories. Focus groups were conducted separately for male and female participants (three groups each). In addition, in-depth interviews were conducted with 30 key informants and providers, including administration and technical personnel of the dormitories, student committee members, doctors, and cafe and bar owners. Field observations and social mapping were also conducted.
Thirty-two of the 56 villages identified by the Zimbabwe government as growth points for economic development were selected as the venues and, specifically, within the growth point villages, the bottle stores, markets and general dealer stores were selected as the points of contact. In conjunction with field observations and social mapping, a total of 49 focus group discussions (25 with women) and 55 key informant interviews (24 with women) were conducted. In-depth, 2–3 hour interviews were held with 15 participants, with approximately equal numbers of men and women. A total of 176 rapid assessment interviews were also held with target community members.
In summary, the venues in each country support frequent informal conversations among friends, neighbors, and co-workers, which provide the C-POL with multiple opportunities to deliver HIV/STD prevention messages designed to shift the social norms towards HIV-related risk reduction. The core age range of the high-risk populations was between 18 and 30 years. In some countries, however, the range was adjusted to reflect the ages with the highest risk behaviors and prevalence of HIV/STD in the venue. For example, in China, older men in the markets had the resources to engage in rituals associated with sex outside of marriage. Therefore, on the basis of ethnographic and epidemiological findings, the upper age range of the population was extended to 40 years in India and Peru and to 49 years in China.
Both men and women were selected in all countries, although the populations were not always balanced. In India and Peru, only approximately 10% of each population was female. In India, only men drink in wine shops because of cultural mores, but the high-risk women who congregate near the shops and often meet with those men as they leave the shops were included in the population. In Peru, approximately 80% of the members of the population were high-risk heterosexual men, but young women with multiple partners and homosexual men each made up approximately 10% of the population because they engaged in high-risk sex within the venue. In China, all workers in the markets were vulnerable to HIV, so both male and female market stall owners and employees were included in the population. Similarly, in Russia, both men and women engaged in HIV risk behaviors in the dormitories and tended to have serially monogamous relationships without using condoms, so all dormitory residents over the age of 18 years who were not going to graduate within a year were included in the population. In Zimbabwe, primarily men engaged in sex with multiple partners, but women were vulnerable to HIV and STD because they did not tend to use condoms with their sexual partners, whom they were likely to see as steady. Therefore, both men and women had high HIV and STD rates, and all members of the venue who attended the selected bottle stores, general dealers, etc., at least twice a week were selected as members of the study population. The selected C-POLs interact with these high-risk populations and received training about how to deliver effective HIV/STD prevention messages. In the next section, we discuss how the rapid ethnographic assessment contributed to the identification of C-POLs and the basis for messages.
Discussion of two core topic areas
Across the core topic areas investigated, findings related to sexual practices and meanings, and C-POL characteristics were most instrumental in shaping the implementation of the C-POL intervention in the five countries. The collected ethnographic information was central to decisions in all sites about those aspects of the sociocultural context and HIV-related risk behaviors most open to influence and who would be most effective in delivering HIV/STD prevention messages. Examples from the findings in each country illustrate that despite the common C-POL intervention model being applied, economic and sociocultural variations required site-specific tailoring of each of the required core elements of the C-POL intervention.
Sexual practices and meanings
Across countries, a number of prevention-related themes emerged from field observations, focus groups and in-depth interviews related to sexual practices and meanings.
The overarching issue across sites was the power imbalance between men and women, which was reflected in their sexual practices. Reports from all five countries described power dynamics tilted toward men, most dramatically in China, India, and Peru, which rendered women particularly vulnerable to HIV and other STDs. Both women and men highlighted the role of sex inequities in shaping intracouple dynamics and sexual practices. Women's relative lack of power also manifested itself in sexual and physical violence against them in all sites except for China, especially when they were resisting having sex with a husband or partner. Similarly, in Peru, power dynamics between men who have sex with men were tilted towards men who self-identified as heterosexual, with limited ability for the men who self-identified as homosexual to negotiate safer sex.
Another manifestation of the power imbalance was the low condom use across sites despite the high prevalence of STDs or HIV in the populations selected in the venues . Women were frequently unable to refuse sex or to negotiate condom use. Most men and women explained that a wife is expected to respond to her husband's sexual urges in a way that satisfies him. In addition to the inability of women to protect themselves as a result of men's entitlement is the implied accusation of infidelity when a woman attempts to suggest condom use. Respondents across the sites did report that condoms are used for preventing HIV/STDs primarily when men have sex with prostitutes or others they consider ‘high’ risk. In Peru, condom use is limited and both men and women reported that it does not feel the same to use a condom; furthermore, condoms were believed potentially to break, be too small, or be ineffective. If a man suggested using condoms, a woman would be offended, believing that he was unfaithful or distrusted her. In Russia, steady couples refuse to use condoms and instead use contraceptive pills to prevent an unwanted pregnancy, even though one or both members of the couple may have had a similar relationship with someone else only a month or two earlier.
In addition to limiting women's choices in sexual practices, sex inequities were evident in the acceptability of premarital and extramarital affairs, which resulted in a double standard for men and women in four of the sites. Only Russia accepted premarital and extramarital affairs for both men and women. In the other four countries, these behaviors were only tolerated (and frequently encouraged) for men. In those sites, husbands reported that wives may be displeased with their extramarital affairs, but the women had no recourse. In Peru, extramarital sex is prohibited among ‘decent’ women, but men may release their sexual tensions with ‘promiscuous’ women, who have multiple male partners and are marginalized within the population. Among the low-income men and women interviewed in India, cultural norms strongly prohibit women from engaging in premarital or extramarital sex, whereas norms sanction a variety of sexual behaviors for men . Reports indicate that a strong sexual drive for young Indian men is natural and expected. This attitude justifies both premarital and extramarital affairs for married and unmarried low-income men in urban India, and contributes to their high risk of HIV/STD infection and, secondarily, to a high risk for their wives and girlfriends.
Types of partners and risk behaviors
Across the sites, ethnographers found similarities and differences in the types of sexual partners and associated sexual risk. Men and women cited trust and knowing a partner's behavior as the most common reasons for choosing a steady partner. Both men and women used a visual assessment that reassured them that the potential partner was healthy and therefore no condom was required during sex. Except for commercial sex workers, fewer women compared with men were described as engaging in sex with a variety of different partners. Distinctions were made between ‘regular’ partners or ‘good’ girls and ‘casual’ partners or ‘fast’ girls. Many unmarried men described their sexual relationships with their regular girlfriends in romantic terms because ‘love is involved’, whereas in casual relationships or with prostitutes or sex workers ‘it's only sex’. Having a steady partner was also described as a means by which to avoid HIV/STD infection and a justifiable reason for not using condoms. Culturally, premarital sex for men remains acceptable in all five countries, so men who desire sexual experiences with multiple partners before marriage have multiple opportunities.
In China, expressions of sexuality and media exposure to sex have, until recently, been culturally and politically suppressed . As these norms change, they also reflect changing types of sexual partners and opportunities for sexual risk and early sexual initiation. Both married and unmarried men frequent clubs, saunas, and dance halls or bars, where alcohol is consumed and sexual liaisons and interactions occur with waitresses, bar girls, massage girls, and sex workers. These casual sexual partners may be labeled as ‘fast women’. In contrast, sexual initiation for young men in India may result from sexual partners known as ‘aunties’ (an older, often married relative or family friend), with other men , or with a prostitute, which may involve a group of young men visiting a venue where sex workers are available. In both China and India, alcohol consumption in groups frequently precedes sexual experiences to help the men gain the ‘courage’ needed for the visit but does not result in steady partnerships.
In Peru, men distinguished between ‘fast girls’ (movidas) and ‘decent girls’ (tranquilas), and described different kinds of sex acts that are permissible with them. Gang rape, associated with getting a fast girl drunk, was a practice reported often. Young, single, unemployed Peruvian men reported infrequent sex with prostitutes because of the cost. Instead, they reported going to parties and chicken cookouts to look for fast girls. Another strategy to have sex was to approach homosexual men from the same neighborhoods, because young, inexperienced heterosexual men think it is easier than having sex with a woman; it is free, and they perceived that the homosexual men are always available for sex. In those situations, heterosexual men mostly performed the insertive role with homosexual men. Although knowledge of HIV/STD among homosexual men in Peru was relatively higher than among heterosexual men and women of that country, their capacity to negotiate safer sex and condom use was limited.
Data collected from young Russian male and female students residing in dormitories also indicated different types of sexual partners, although sex influenced the interpretation of the types. ‘Permanent’ and ‘incidental’ partners reflect the categories of ‘regular’ and ‘casual’ that were evident in the other countries. Respondents reported that ‘incidental’ acquaintances and liaisons are usually found in clubs, discotheques, and bars, and are accompanied by high levels of alcohol consumption. These sexual relationships are for the most part unprotected because of the low use of condoms among the students. Participants reported that the majority of students have little or no sexual experience before moving into the Russian dormitories. Many students considered partnerships lasting one month or more to be steady, stable, and permanent. Female students, however, often interpreted this steady relationship (frequently the only relationship for the female during a defined period of time) as a civil marriage (usually not registered or formalized, or even verbalized between the partners). By contrast, young men consider themselves free even in the presence of a steady partner (of whom there could be several at one time), until they declare their intention to marry.
In Zimbabwe, several different types of sexual partners are recognized. A man may have multiple wives as part of the practice of polygamy. A man who moves to another place for employment often has a wife in the rural area and a regular girlfriend for whom he is paying the ‘lobola’ (bride price) at the site where he found work. Men described having sex with concurrent relationships, usually a steady partner or wife and also other partners. If men had sex with casual partners or commercial sex workers, they often declared that they used condoms. Both men and women, however, mentioned that condom use in steady or martial relationships was ‘not the done thing’. Women were much less likely to have multiple partners, and although acknowledging that their steady partner often had other partners, they felt that they could not discuss or expect a commitment or monogamy from their husbands or steady partners.
This more complete understanding of sexual practices and meanings was critical to the development of HIV/STD prevention messages delivered by the C-POL, particularly messages that recognize sex inequities in practising safer sex with different sexual partners.
Characteristics of community popular opinion leaders
Across sites, members of study populations reported many similarities regarding the characteristics of natural leaders who would be able to deliver effective health messages to their social networks. Regardless of the subpopulation within the country (e.g. married men, married women, men who have sex with men), informants converged on the following characteristics of an ideal C-POL: respectable, credible, experienced (life and sexual), trustworthy, empathetic, well-spoken, and self-confident. Within study populations in all countries, however, the characteristics that made for a natural leader varied by subpopulation. Again, country-specific descriptions illustrate the variety of characteristics that emerged in the data regarding the selection of C-POL candidates that would best fit the unique sociocultural circumstances.
Recognizing the need to emphasize female-focused HIV/STD prevention messages, married and unmarried Chinese and Indian women identified older, married, sexually experienced, empathetic, persuasive, non-judgmental, and bold women who were uninhibited about discussing sensitive topics as ideal C-POLs. Both married and unmarried women at the two sites agreed that an unmarried woman may be too shy and inexperienced to be an effective C-POL.
In Peru and Russia, somewhat younger populations were targeted, and C-POL characteristics included skills in getting along with everyone, having social prestige, and being a little older and therefore having more life experience than the youngest members of the population. Targeting a student population, the Russian participants described optimal C-POLs as individuals with a depth of knowledge on relevant topics, who possessed good communication skills, and who had an interesting personality. According to Russian students, C-POLs should be peers who are extroverts, are involved in a variety of school activities, frequently offer to help, provide good advice to fellow students, and have lived in the dormitory for at least 2 years. Some groups of male students completely rejected the concept of a leader, as leadership was equated with subordination of the masses.
In Zimbabwe, men and women tended to identify more formal leaders as potential C–POLs, although, in addition, popular individuals who were seen as leaders were also identified. Village chiefs, headmen, kraalheads (leaders of a rural village, typically consisting of huts surrounded by a stockade), village elders, and health workers were widely considered the most appropriate people to deliver health messages, given their experience and training and the respect that they hold in the community. Both popular men and women who were not necessarily part of the formal leadership network were also identified as C-POLs. Both men and women in Zimbabwe agreed that desirable C-POL characteristics included honesty, sensibility, eloquence in speech, ability to provide good advice and keep confidences, and being a behavioral role model. Being a hard worker, religiously faithful, easily approachable, intelligent, and active in the community were offered as admirable characteristics.
A major challenge for this study was to adapt a single community-based C-POL intervention to prevent HIV/STD transmission to five diverse international settings (China, India, Peru, Russia, and Zimbabwe). A rapid ethnographic assessment was conducted to tailor the assessment and intervention design and implementation. Although a detailed protocol was developed to standardize methods and assessment tools, the teams had latitude in matching the method used to elicit the required data to the sociocultural context and high-risk population. During the conduct of the Trial, the ethnographers communicated across the sites frequently through meetings, conference calls, and E-mail in order to prevent protocol drift. The success of the implementation of the C-POL intervention in such diverse cultural situations with different high-risk populations was a result of the important data collected by the ethnographers.
Several patterns emerged when examining data across sites for two core topic areas: sexual practices and meanings and C-POL characteristics. First, cultural norms of sex inequity, sex relations that favor men, and sanctioning of men's premarital and extramarital sexual behavior played an important role in increasing sexual risk for both men and women. Women, whether married or unmarried, younger or older, understood their vulnerability to HIV/STD infection related to a partner's sexual behavior and their limited ability to protect themselves by negotiating condom use or insisting on a partner's fidelity.
In all sites except China, the narrative data contained explicit discussion by both men and women about the use of violence in sexual relationships. This violence took the form of forced sex, often between married partners, in addition to other kinds of sexual relationships, beatings, and individual or gang rape. The prevailing viewpoint in Zimbabwe was that a man has a right to have sex with his wife, and the wife must acquiesce; thus pressure or coercion exists, although physical violence was only rarely mentioned. In India, where violence toward women is prevalent in daily life, women are likely to submit to sex to avoid harm to themselves [16,19–21]. In all of the sites, alcohol played a key role in sexual risk-taking because alcohol consumption increased the likelihood of unprotected sex with partners who may be at high risk of the transmission of HIV/STD.
In China and India, where the HIV/AIDS epidemic was at a relatively early yet rapidly growing stage, stigma associated with disclosure of infection was evident. For example, in China, having an STD marked an individual's failure to adhere to sexuality norms, prompted negative emotions (e.g. shame, fear, and embarrassment), devalued the infected individual's social roles, and reduced his or her status. Strategies that informants reported using to avoid stigmatization included avoiding HIV/STD knowledge, avoiding seeking healthcare professionals, particularly in public settings, and conforming to community norms of shunning those suspected of risky behaviors . In Peru, however, having an STD was a sign of ‘manliness’ and proof that the man had multiple partners and sexual prowess.
This formative multisite ethnographic study was instrumental in identifying the social and cultural factors related to health behaviors important to HIV and STD transmission that were to be considered during the design and development of the assessment and the HIV prevention intervention. For example, the behavioral assessment was designed with core questions that were asked at all sites and with optional and unique questions that were used to tailor the assessment at individual sites, a necessarily flexible application of a systematic approach. In China, some of the optional questions were used to introduce the assessment as a broader health questionnaire rather than a sexual behavior questionnaire to increase its acceptance by the target population. In other countries, additional questions focusing on mediators identified during the ethnographic activities, such as injection drug use or violence, were included as unique questions in the assessment.
Each site culturally tailored the HIV/STD prevention messages to address the local social and cultural context and sexual practices and meanings. For example, Russian students tend to be single and have serial relationships. Accordingly, messages were developed to encourage students to use a condom every time they have sex. In China, many of the study participants were married. Therefore, messages encouraged all members of the population to have routine checkups to detect asymptomatic HIV/STDs and receive treatment when needed. Messages also encouraged married people to use condoms outside of the marriage and encouraged single people to use condoms for every sexual act. Across the sites, C-POLs encouraged their friends and neighbors to initiate conversations about sexual health with all sexual partners.
This paper illustrates how ethnographic formative research provided critical information that enabled the study teams to adapt components of the Trial in culturally appropriate ways in diverse international settings. The attention paid to both standardization and tailoring of the C-POL intervention and assessment based on a comprehensive rapid ethnographic assessment strengthened the ability of this Trial to evaluate the efficacy of this community-based intervention.
1. Bachrach CA, Abeles RP. Social science and health research: growth at the National Institutes of Health. Am J Public Health 2004; 94:22–28.
2. Scrimshaw SC, Hurtado E. Rapid assessment procedures for nutrition and primary health care: anthropological approaches to improving programme effectiveness
. Reference Series, Vol. 11. Los Angeles, CA: UCLA Latin American Center; 1987.
3. Bentley ME, Gittelson JG, Nag M, Pelto PJ, Russ J. Use of qualitative research methodologies for women's reproductive health data in India
. In: Scrimshaw N, Gleason G, editors. Rapid assessment procedures qualitative methodologies for planning and evaluation of health related programmes
. Boston, MA: International Nutrition Foundation for Developing Countries (INFDC); 1992.
4. Bentley ME, Fullem AM, Tolley EE, Kelly CW, Jogelkar N, Srirak N, 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.
5. Schensul JJ, LeCompte MD, Hess GA Jr, Nastasi BK, Berg MJ, Williamson L, et al
. Using ethnographic data: interventions, public programming, and public policy
. Ethnographer's Toolkit, Vol. 7. Lanham, MD: Altamira Press; 1999.
6. Steckler AL, Linnan L, editors. Process evaluation for public health interventions and research
. San Francisco, CA: Jossey-Bass; 2002.
7. Verma RK, Pelto PJ, Schensul SL, Joshi A, editors. Sexuality in the time of AIDS: contemporary perspectives from communities in India
. Thousand Oaks, CA: Sage Publications Pvt. Ltd.; 2004.
8. Ulin PR, Robinson ET, Tolley EE. Qualitative methods in public health: a field guide for applied research
. San Francisco, CA: Jossey-Bass; 2004.
9. NIMH Collaborative HIV/STD Prevention Trial Group. Design and integration of ethnography within an international behavior change HIV/sexually transmitted disease prevention trial.AIDS
2007; 21 (Suppl. 2)
10. Rogers EM. Diffusion of innovations
, 4th ed. New York: Free Press; 1995.
11. NIMH Collaborative HIV/STD Prevention Trial Group. Methodological overview of a five-country community-level HIV/sexually transmitted disease prevention trial
2007; 21 (Suppl. 2)
12. NIMH Collaborative HIV/STD Prevention Trial Group. The community popular opinion leader HIV prevention programme: conceptual basis and intervention procedures
2007; 21 (Suppl. 2)
13. Kelly JA. Popular opinion leaders and HIV prevention peer education: resolving discrepant findings, and implications for the development of effective community programmes. AIDS Care 2004; 16:139–150.
14. NIMH Collaborative HIV/STD Prevention Trial Group. Selection of populations represented in the NIMH Collaborative HIV/STD Prevention Trial
2007; 21 (Suppl. 2)
15. NIMH Collaborative HIV/STD Prevention Trial Group. Sexually transmitted disease and HIV prevalence and risk factors in concentrated and generalized HIV epidemic settings
2007; 21 (Suppl. 2)
16. Go VF, Sethulakshmi CJ, Bentley ME, Sivaram S, Srikrishnan AK, Solomon S, et al
. When HIV-prevention messages and gender norms clash: the impact of domestic violence on women's HIV risk in slums of Chennai, India. AIDS Behav 2003; 7:263–272.
17. Lieber E, Li L, Wu Z, Rotheram-Borus MJ, Guan J, the National Institute of Mental Health (NIMH) Collaborative HIV Prevention Trial Group. HIV/STD stigmatization fears as health-seeking barriers in China. AIDS Behav 2005; 23:1–9.
18. Go VF, Srikrishnan AK, Sivaram S, Murugavel GK, Galai N, Johnson SC, et al
. High HIV prevalence and risk behaviors in men who have sex with men in Chennai, India. J Acquir Immune Defic Syndr 2004; 35:314–319.
19. Go VF, Johnson SC, Bentley ME, Sivaram S, Srikrishnan AK, Celentano DD, et al
. Crossing the threshold: engendered definitions of socially acceptable domestic violence in Chennai, India. Cult Health Sex 2003; 5:393–408.
20. Sivaram S, Lakshimi CS, Go VF, Srikrishnan AK, Celentano DD, Solomon S, et al
. Telling stories: narrative scenarios to assess sexual norms for an HIV prevention study in Chennai, India. In: Verma RK, Pelto PJ, Schensul SL, Joshi A, editors. Sexuality in the time of AIDS: contemporary perspectives from communities in India. Thousand Oaks, CA: Sage Publications Pvt. Ltd; 2004.
21. Panchanadeswaran S, Johnson SC, Go VF, Srikrishnan AK, Solomon S, Bentley ME, Celentano DD. Using the theory of gender and power to examine the experiences of partner violence, sexual negotiation, and risk of HIV/AIDS among economically disadvantaged women in Southern India
. In: McCloskey K, Sitaker M, editors. Backs against the wall: battered women's use of violence
. Binghamton, NY: the Haworth Press Inc.; in press.