Article
The Impact of Key HIV Intervention Components as Predictors of Sexual Barrier Use: The Zambia Partner Project
Journal of the International Association of Providers of AIDS Care 1-8 ª The Author(s) 2014 Reprints and permission: sagepub.com/journalsPermissions.nav DOI: 10.1177/2325957414520980 jiapac.sagepub.com
Ndashi Chitalu, MD1, Mirriam Mumbi, CNM1, Ryan Cook, BA2, Stephen M. Weiss, PhD, MPH2, and Deborah Jones, PhD2
Abstract Behavioral interventions have utilized a variety of strategies and components to reduce HIV risk. This article describes the partner intervention, a couple-based group HIV risk reduction intervention implemented in 6 urban community health clinics in Lusaka, Zambia, and examines the components of the intervention and their relationship with condom use. Couple members completed assessments on condom use, acceptability, willingness to use condoms, communication, intimate partner violence (IPV), self-efficacy, and HIV information at baseline and 6 months’ follow-up. This study examined the relative impact of elements of the intervention as predictors of condom use. Changes in acceptability had the greatest overall influence on condom use, followed by social support, relationship consensus, and willingness to use condoms. Changes in self-efficacy, IPV, negotiation, and information had no influence. Results support the use of multidimensional approaches in behavioral interventions and highlight the importance of identifying critical elements of interventions to maximize risk reduction outcomes. Keywords HIV, intervention, Africa, couples
Introduction 1
The HIV epidemic has begun to decline, which is likely due, in part, to the efficacy of programs developed and implemented globally to reduce sexual risk behavior.2 These programs rely on a variety of methods and elements to influence behavior and reduce the risk of HIV transmission and/or infection.2 Program interventions are tailored to the cultural context of the target population and focus on elements such as information about HIV and testing, transmission of the virus, and prevention strategies, such as condom use.3,4 Interventions may also target communication skills,5 sexual negotiation,6 intimate partner violence (IPV),7 self-efficacy,8 perceptions of sexual barrier use,9 and skill training10,11 and adapt these techniques to the intended audience (eg, people living with HIV).12,13 However, the combination of diverse elements in behavioral interventions within specific settings makes identification of the critical intervention elements’ underlying risk reduction challenging. The partner intervention is a couple-based HIV risk reduction intervention that was culturally tailored to the local population in the capital province of Lusaka, Zambia, adapted for use in the community health clinic setting14,15 and implemented in 6 community health clinics.16 Nationwide, Zambia has a generalized HIV epidemic, with over 1 million persons living with the virus and high rates of HIV in urban Lusaka (20.8%).17 The elements of the partner intervention, that is, information,
skill building, and cognitive behavioral (CB) strategies targeting self-efficacy, sexual communication, and sexual barrier acceptability, were hypothesized to enhance the uptake of sexual barrier products. The intervention format, gender concordant groups, has been found to be effective in reducing risk behavior18,19 and shown to enhance the uptake of sexual barrier products in this setting,20 in comparison with traditional individual health education21 and with the standard of care.16 This article reviews the components of the Zambian intervention and examines the relative impact of intervention elements, that is, information, social support, self-efficacy, communication, IPV reduction, and sexual barrier product acceptability, on sexual barrier product use.
1
Department of Pediatrics, University Teaching Hospital, University of Zambia School of Medicine, Lusaka, Zambia 2 Department of Psychiatry and Behavioral Sciences, University of Miami Miller School of Medicine, Miami, FL, USA Corresponding Author: Deborah Jones, Department of Psychiatry and Behavioral Sciences, University of Miami Miller School of Medicine, 1400 NW 10th Ave, Suite 404A, Miami, FL 33136, USA. Email: [email protected]
Downloaded from jia.sagepub.com by guest on November 14, 2015
2
Journal of the International Association of Providers of AIDS Care
Methods
Measures
This study was conducted from March 2009 to March 2012; ethical review and approval were obtained prior to study onset from the University of Zambia Research Ethics Committee and the University of Miami Miller School of Medicine Institutional Review Board, in accordance with the provisions of the US Department of Health and Human Services and the Zambian government. Study sites (n ¼ 6 urban community health centers [CHCs]) enrolled their first cohort (a group of 8 couples) into the control condition and all subsequent cohorts into the intervention condition (a total of 5 cohorts were enrolled per site).
Demographics. Participants’ age, educational level, employment status, religion, marital status, number of children, and antiretroviral therapy status (if HIV positive) were assessed.
Recruitment and Enrollment Recruitment of participants was conducted by community mobilizers at the 6 CHCs. Mobilizers referred potential participants to assessors at their local CHC for screening and enrollment. Participants were 18 years of age or older, sexually active within the past month, in a couple relationship for at least 6 months, and 1 or both members of the couple were seropositive. HIV serostatus was verified for the sero-positive status of one partner or both the partners; prior to enrollment, seronegative participants were referred for HIV testing with their partner. Approximately 30% of couples screened for enrollment were not eligible due to the lack of sexual activity in the previous month or failure to demonstrate couple status. All informed consent, assessment, and intervention materials were translated into the major Lusakan languages (Bemba and Nyanja) in accordance with a ‘‘cultural brokerage’’ process to ensure comprehension and validity of data. Cultural brokerage is a process of translation and back translation, followed by discussion of items of disagreement, to arrive at the most accurate and meaningful translation. Following the provision of informed consent, enrolled participants (n ¼ 240 couples) were invited to complete a baseline assessment. All study assessments were completed using an audio computerassisted self-interview system that collected information with minimal assistance from the assessors. The participants were left alone to listen, read, and respond to recorded questions presented on the computer screen and through head phones; the participants recorded their answers using a touch screen connected to the computer. All responses were stored by the computer, and the information was uploaded and transformed to SPSS. Participants were compensated (K50 000 Zambian Kwacha *US$10) for each assessment but were not compensated for intervention attendance. Participants enrolled in the intervention condition were provided with an appointment for the intervention; those in the control condition were in a delayed start condition and were offered the intervention after 12 months, following their final assessment. During the waiting period, only the control condition participants completed the assessments. This article focuses on the elements of the intervention, which influenced sexual barrier use; therefore, the 40 couples in the control condition were not included in the analytic data set (included n ¼ 200 couples).
Sexual barrier use. Participants reported each incidence of sexual intercourse each day of the preceding week and whether or not a condom was used, in a diary format. A pictorial representation of condoms was provided to enhance comprehension. Condom use in the past week was calculated by dividing the total number of condoms used during the week by the total number of sex acts. Because not all participants had sex within the week before assessments, they also provided an assessment of their typical condom use (in general) using a 5-point Likert-type scale (5 ¼ every time; 4 ¼ almost every time; 3 ¼ sometimes; 2 ¼ almost never; and 1 ¼ never). Willingness to use condoms and acceptability of condoms. This measure was adapted from the University of California at San Francisco Center for AIDS Prevention Studies Barrier Questionnaire, and the current and previous use of sexual barriers and willingness to use sexual barriers using Likert-type scales were measured. Acceptability, here defined as ‘‘liking’’ a product, was measured by asking each individual how much he or she liked using male condoms. Participants responded using the scale ‘‘like very much ¼ 5; like somewhat ¼ 4; neutral ¼ 3; dislike somewhat ¼ 2; strongly dislike ¼ 1, and never used ¼ 0.’’ Participants also reported their ‘‘willingness to use’’ male condoms. This item was rated as ‘‘very willing to use ¼ 4; moderately willing ¼ 3; slightly willing ¼ 2; and not at all willing ¼ 1.’’ Communication—relationship consensus. The relationship consensus subscale of the Dyadic Adjustment scales (DAS)22 assessed achievement of consensus in couples’ relationships. Higher scores on the DAS consensus subscale indicate higher consensus on difficult topics such as financial, religious, and sexual issues. The consensus subscale demonstrated high internal consistency in this sample (a ¼ .86). Communication—negotiation and IPV. A modified version of the 17-item Conflict Tactics scale23 was used to measure negotiation (eg, discussion and problem solving) and IPV (eg, pushing, hitting, and slapping). Participants indicated the number of times their partners had engaged in these behaviors during the preceding month. Responses were scored using a Likert-type scale of 0 (never) to 6 (more than 20 times). Information. HIV knowledge was assessed using an adaptation of the AIDS-Related Knowledge Test.24 This adaptation from the 10-item scale assesses HIV risk and prevention-related knowledge. Items reflect information about HIV transmission, condom use, and AIDS-related knowledge and are responded to as yes, no, or don’t know. The AIDS-Related Knowledge Test was scored for the number of correct responses and has acceptable internal consistency (a ¼ .73).
Downloaded from jia.sagepub.com by guest on November 14, 2015
Chitalu et al
3
Self-efficacy. Sexual self-efficacy was measured using the Sexual Self Efficacy Questionnaire.25 This 7-item measure assesses the level of sexual self-efficacy by asking participants to rate their level of comfort with specific sexual risk reduction behaviors using a Likert-type scale, including purchasing and using male or female condoms, refusing unprotected sex, and negotiating condom use. The self-efficacy score has demonstrated acceptable internal consistence (a ¼ .77). Social support. Social support was measured using the Brief COPE-Revised.26 The 38-item revised version of the Brief COPE measures the extent to which an individual engages in various forms of coping strategies in response to life stressors; the instrumental and emotional support subscales of the Brief COPE-revised were combined into a 4-item social support measure. Items are rated from 0 (I haven’t been doing this at all) to 3 (I’ve been doing this a lot). Higher scores are associated with greater use of the coping strategy. The 2 subscales used to create the social support measure were highly correlated (r ¼ .61, P < .001), and the social support measure demonstrated acceptable internal consistency in this sample (a ¼ .76).
Intervention Topics The partner intervention began as a cross-cultural adaptation of an evidence-based HIV prevention intervention, the New Opportunities for Women Project.27 Key elements of the intervention were identified by focus group and pilot research conducted from 1999 to 2003.20,28 From 2003 to 2008, the partner project was translated, tested, and implemented into the CHC setting.15 The theory-based intervention14 consisted of 4 sessions presented in same-gender groups of 8 participants delivered over 4 weeks; sessions were led by trained facilitators certified in HIV counseling and/or nursing. Sessions began by providing participants with an assurance that confidentiality is a cardinal factor during the groups to alleviate fears and enhance trust within the groups and toward the facilitators leading the intervention. Participants were informed that the sessions would take place at an appointed time for roughly 2 to 2.5 hours. Audio taping was used to record comments participants shared among each other, to review what was discussed, and to provide information for quality control by the investigators. Participants were also informed that the tapes used are kept under lock and key. Cognitive behavioral strategies: acceptability and willingness to use sexual barriers. This component was included in all of the sessions and often elicited excitement and laughter from participants. Discussion of the techniques stimulated extensive conversation, and, if not mindful of the time, facilitators could find themselves exceeding the amount of time planned. Cognitive behavioral strategies address how a person reacts to a particular situation before any behavior occurs, describing the process of reacting first with the mind, the emotions, and later with behavior. This topic was both difficult and interesting for participants as it made them aware of the source from which
their actions arose. In presenting CB, it was important to provide a number of examples, such as first impressions of a female condom might elicit thoughts such as ‘‘it is big, it is ugly’’ or ‘‘I am scared to use it’’ or ‘‘I am excited about this,’’ leading to positive or negative behavioral outcomes. This section also addressed coping, developing problem-solving skills, and applying stress reduction strategies, using CB strategies such as linking thoughts to emotions, breaking down the process of reacting with thoughts and feelings, identifying automatic thoughts, rational thought replacement, and cognitive restructuring, and changing how we think. Participants were shown that changing how they think can lead to changing their behavior, which may result in a more positive perception of condom use, and thus increase the likelihood of practicing safer sex. Communication. Communication about sex between men and women is an essential component of HIV prevention, and the topic was incorporated into most sections of the intervention. Sexual negotiation was conceptualized as having the goal of reaching an agreement on sexual topics such as the use of condoms. However, communication training did not only address sex but also included strategies that can render all types of communication between partners more positive and effective. This study encouraged participants to talk about things that are not normally discussed by couples in Zambia, such as HIV reinfection, sexually transmitted infections (STIs), unwanted pregnancy, and many other sensitive issues. In order to illustrate effective communication, role-plays were performed within the groups, in which one participant acted as a man and another as a woman. During the role-play, the ‘‘couple’’ focused on the importance of achieving a level of understanding with their partner while ensuring no conflict would ensue between the two people. Both men and women presented their partners using their typical manner of speech, as in the tone of voice and the way they might ‘‘handle’’ their partner, explaining the use of sexual barriers and why. As most participants were shy in the beginning, facilitators illustrated role-playing before asking participants to perform for the group; most appeared to enjoy role-playing during the sessions. Intimate partner violence. It was recognized that some couples do not continuously have a rosy relationship but may also have some degree of violence in their relationships. The IPV was integrated into communication topics, and some participants shared psychological abuse, as in not communicating with the partner or screaming and shouting at their partners. Some also shared experiences of being beaten with objects and described having resorted to running away from home, reporting their situation to close relatives or friends, and also turning to IPV Victim Support Units. Participants were encouraged to share concerns about the potential for IPV and to establish safety plans in the event that IPV became an issue. Those women who became estranged from their partners were provided with counseling support and scheduled for sessions that did not coincide with their partner’s sessions.
Downloaded from jia.sagepub.com by guest on November 14, 2015
4
Journal of the International Association of Providers of AIDS Care
Information: STDs and HIV. As there are many forms of STIs, including HIV, this topic utilized a poster showing different types of STIs, including chlamydia, syphilis, human papilloma virus, genital herpes, hepatitis B, and HIV. Participants were encouraged to develop a clear picture of what infectious diseases are and look like, enabling them to understand the ways these infections could be avoided or the transmissions could be prevented. Through discussion, participants were queried on their level of understanding about STIs and HIV. Information: safer sex and reproductive issues. This discussion of relative safety of different types of sexual behavior was an adapted hierarchy of safer sex; facilitators began by splitting the term ‘‘safer sex,’’ safer and sex, to facilitate understanding. Group members generally understood it to mean the prevention of the transmission of HIV to partners, and most were aware that condoms are the best safer sex method. Abstinence was controversial as most were married or cohabiting couples, but the majority had experienced abstinence as a period during which the other partner was away, sick, or sexually fasting, related to prayer. The discussion included nonpenetrative safer sex, such as sexual friction between the thighs or breasts of a woman, and sexual practices such as the use of vaseline or lubricants in the vagina or coitus interruptus. The discussion provided the opportunity to correct misunderstandings among the participants regarding safer sex. Very few participants knew that medical male circumcision also makes sex safer, and very few mentioned masturbation as safer sex, perhaps due to shyness to discuss this practice in public. However, participants, especially women, opened up to discuss masturbation when the facilitator initiated the discussion topic. Discussion of reproductive issues began by exploring desires to have children, addressing participants’ awareness of issues surrounding safer conception, and focusing on times during the menstrual cycle when a woman is most likely to conceive. Participants who wished to conceive were guided on how to chart their cycles to enable them to restrict unprotected sex to fertile periods, thereby reducing the potential for transmission to partners. The female reproductive cycle was explained in detail with a poster diagram and a female anatomic model, to provide participants with a full understanding of the process of conception, from sexual intercourse to delivery. Knowing the days when the woman is fertile was used to help participants know when the woman cannot become pregnant, enabling her to use dual methods for protection against HIV during at least two-thirds of her cycle (eg, condoms plus contraceptives, diaphragms, or caps). Very few participants considered the increased potential for transmission during conception and were unaware of strategies to plan safer conception. In fact, most women did not engage in any communication with their spouses concerning getting pregnant. Discussion also addressed protecting mother, father, and baby from STIs and HIV during pregnancy using safer sex practices. Participants were reminded of the benefits of creating and maintaining a good relationship with their doctors preconception, during pregnancy, and postpartum and of the need
to discuss having a baby with their doctor as the first step in planning conception. Women discussed the tests typically performed preconception, that is, pap smear, HIV, and CD4 count to determine the level of immunity in the body. Skill training. Sessions included opportunity for skill training in (1) the use of male and female condoms and vaginal lubricants, (2) positive communication and sexual negotiation strategies, and (3) CB strategies such as reframing perceptions to change behavior. Male and female condoms and lubricants were supplied to each and every individual upon completion of each session. Skill training and group sessions were used to enhance self-efficacy and social support for engaging in new behaviors.
Statistical Analyses In order to examine the relationship between changes in elements targeted by the intervention and changes in condom use, change scores between baseline and follow-up were computed. Follow-up was 6 months postbaseline, which was approximately 4 to 5 months after participants completed the intervention. Then, changes in condom use were regressed on changes in each element of the intervention separately. Differences between genders were tested by testing interactions with gender, and gender effects were dropped if nonsignificant. Subsequently, a full-factorial combination of elements that were found to be significant was simultaneously entered into a multivariable model. Nonsignificant interactions and main effects were removed, beginning with the least significant, until a final model was determined. Changes in condom use in the week preceding assessment was the primary outcome; however, due to the small sample of participants reporting sex at both time points, a measure of condom use ‘‘in general’’ was also analyzed. Because this study assessed couples, models were estimated using generalized estimating equations including a residual correlation for person within couple. All analyses were conducted using SAS PROC GENMOD (SAS 9.3; SAS Corporation, Cary, North Carolina) at a 2-tailed level of significance of P ¼ .05.
Results Demographics Participants (N ¼ 400; 200 couples) averaged 38 years old with 8 years of education and most were unemployed (n ¼ 247, 62%). The majority of participants identified as Protestant (n ¼ 207, 52%) or Catholic (n ¼ 138, 35%). Nearly all were married (n ¼ 388, 97%) and had children (n ¼ 356, 89%); the mean number of children was 3. In all, 371 (93%) participants were HIV sero-positive and 29 (7%) were sero-negative, resulting in 171 seroconcordant and 29 serodiscordant couples. Of the 371 HIV-positive participants, 236 (64%) were taking antiretroviral medications. Table 1 presents further descriptions of participant demographics.
Downloaded from jia.sagepub.com by guest on November 14, 2015
Chitalu et al
5
Table 1. Demographics. Characteristic (N ¼ 400) Age, y Years of education Employment status Employed Unemployed Religion Protestant Catholic Other/not religious Marital status Married Not married At least 1 child No children Number of children HIV status HIV positive HIV negative ARV status (n ¼ 371 HIV positive) On ART Not on ART
M (SD)/n (%) 38.2 (8.2) 8.2 (3.3) 153 (38%) 247 (62%) 207 (52%) 138 (35%) 55 (14%) 388 (97%) 12 (3%) 356 (89%) 44 (11%) 3 (2) 371 (93%) 29 (7%) 236 (64%) 135 (36%)
Abbreviations: ART, antiretroviral therapy; ARV, antiretroviral; M, mean; SD, standard deviation.
Of the 400 participants, 138 reported having sex at least once during both the week before the baseline assessment and the week before the follow-up assessment. The proportion of condom-protected sexual acts in the past week at baseline was 69.0% (standard deviation [SD] ¼ .41), increasing to 83.0% at follow-up (SD ¼ .31). The mean increase in weekly condom use was 14.0% (95% confidence interval [CI] ¼ 0.07-0.21), which did not differ between HIV-negative and -positive participants (t ¼ .73, P ¼ .467) or between men and women (t ¼ .11, P ¼ .910). Similarly, 318 participants contributed both baseline and follow-up general condom use scores. General condom use, which ranged from 1 ¼ never used condoms to 5 ¼ always used condoms, was 4.19 (SD ¼ 1.19) at baseline and increased to 4.44 at follow-up (SD ¼ 1.03). Mean change in general condom use was 0.24 (95% CI ¼ 0.10-0.39), which also did not differ by HIV serostatus (t ¼ .64, P ¼ .522) or gender (t ¼ .64, P ¼ .521). Changes in weekly and general condom use were correlated (r ¼ .46, P < .001).
Impact of Elements of the Intervention on Condom Use In order to determine the elements of the intervention that were associated with changes in sexual barrier use, changes in condom use were regressed on changes in condom acceptability, communication, IPV, information, self-efficacy, and social support. The first series of analyses considered each element separately and were conducted for both condom use in the past week and condom use in general. Results of these analyses, including b coefficients, standard errors, and 95% CIs, are presented in Table 2. In summary, changes in sexual barrier acceptability and willingness to use sexual barriers as well as
relationship consensus, self-efficacy, and social support impacted sexual barrier use. Changes in information, IPV, and negotiation had no impact. Significant predictors and interactions were entered into multivariable models predicting weekly condom use and condom use in general. The results of these analyses are presented in Table 3, including both unstandardized and standardized b coefficients along with standard errors and 95% CIs. As each predictive variable was measured using a different scale, the percentages presented are not directly comparable. The standardized b coefficients are more comparable as they represent the effect of 1 SD change in the predictor on 1 SD of the outcome. In summary, a significant positive relationship between changes in sexual barrier acceptability and weekly condom use was found. A 1-point increase in the acceptability of male condoms was associated with a 5% increase in condom use (standardized b ¼ .228). Social support positively impacted weekly condom use, such that a 1-unit increase in social support was associated with a 2% increase in condom use (b ¼ .211). Finally, willingness to use male condoms also impacted condom use, such that a 1 point increase in willingness was associated with an 7% increase in use (b ¼ .146). No interactions were significant, and self-efficacy did not impact condom use in multivariable analysis. Sexual barrier acceptability also impacted condom use in general; increases in the acceptability of male condoms were associated with significant increases in general condom use (b ¼ .10; standardized b ¼ .159). Additionally, relationship consensus positively impacted general condom use (b ¼ .02; b ¼ .182). No interaction between acceptability and relationship consensus was found.
Discussion This study examined the relative impact of elements of a couple-based group sexual risk reduction intervention on sexual barrier use. Changes in acceptability had the greatest overall influence on sexual barrier use, followed by social support, relationship consensus, and willingness to use sexual barriers. Self-efficacy was associated with the use of sexual barriers, but increased self-efficacy did not contribute to increased barrier use in multivariable analysis. Finally, negotiation, IPV, and information had no apparent influence on sexual barrier use. Study results provide further support for the predictive relationship between acceptability and self-reported willingness to use barriers and sexual barrier use, in line with earlier findings by this team using individual rather than dyadic analyses.28 One of the key elements of CB therapy is the premise that attitudes influence behavior, for example, acceptability of condoms will predict their use. In this study, acceptability may have been increased by CB strategies designed to promote positive attitudes and reduce negative opinions regarding condom use. Condoms are highly effective in the prevention of HIV transmission, but their impact as a prevention strategy has been limited by low acceptability and inconsistent use. Metaanalyses of positive prevention studies in developing countries
Downloaded from jia.sagepub.com by guest on November 14, 2015
6
Journal of the International Association of Providers of AIDS Care
Table 2. Bivariate Analyses of Changes in Condom Use Regressed on Changes in Condom Acceptability, Communication, IPV, Information, Self-Efficacy, and Social Support.a
Condom use in past week Male condom acceptability Willingness to use male condoms Relationship consensus Negotiation Intimate partner violence HIV/AIDS-related knowledge Self-efficacy Social support Condom use in general Male condom acceptability Willingness to use male condoms Relationship consensus Negotiation Intimate partner violence HIV/AIDS-related knowledge Self-efficacy Social support
b (SE)
P
95% CI (b)
.060 (.019) .113 (.036) .005 (.003) .001 (.007) .004 (.004) .030 (.021) .010 (.004) .026 (.009)
.002 .002 .100 .894 .404 .153 .018 .004
0.023-0.096 0.043-0.183 0.011-0.001 0.012-0.014 0.005-0.012 0.011-0.072 0.002-0.019 0.008-0.043
.101 (.038) .046 (.096) .021 (.007) .007 (.015) .019 (.012) .065 (.049) .001 (.010) .029 (.020)
.008 .636 .005 .646 .134 .188 .881 .146
0.026-0.176 0.143-0.234 0.006-0.035 0.036-0.022 0.043-0.006 0.032-0.161 0.018-0.020 0.010-0.068
Abbreviations: CI, confidence interval; IPV, intimate partner violence; SE, standard error. a Statistically significant variables are indicated in boldface.
Table 3. Multivariable Analysis of Changes in Condom Use Regressed on Significant Bivariate Predictors.a
Weekly condom use Male condom acceptability Willingness to use male condoms Social support Condom use in general Male condom acceptability Relationship consensus
b (SE)
P
95% CI (b)
b (SE)
95% CI (b)
.049 (.016) .071 (.032) .022 (.009)
.006 .019 .012
0.018-0.081 0.007-0.134 0.005-0.039
.228 (.084) .146 (.067) .211 (.084)
0.063-0.393 0.014-0.277 0.047-0.375
.102 (.040) .021 (.007)
.010 .004
0.025-0.180 0.007-0.035
.159 (.061) .182 (.063)
0.038-0.279 0.059-0.305
Abbreviations: CI, confidence interval; IPV, intimate partner violence; SE, standard error. a Changes in condom use were regressed on changes in condom acceptability, communication, IPV, information, self-efficacy, and social support individually. Significant predictors were entered simultaneously into a multivariable model; the results are shown here.
have reported that interventions targeting HIV sero-positive individuals are successful in increasing condom use, in both seroconcordant and discordant relationships.29 Recent studies, for example, microbicides and preexposure prophylaxis (PreP), have highlighted the importance of assessing the acceptability of prevention strategies as a predictor of uptake. In fact, low product acceptability may be the Achilles’ heel of HIV prevention. Enhancement of product acceptability is likely a critical intervention element and has important implications for the uptake of existing prevention strategies as well as new strategies such as PreP. Social support and relationship consensus also played a key role in the uptake of sexual barrier products. In psychotherapy, the group format is often used to increase the overall impact of the intervention by providing a variety of viewpoints on interpersonal issues, but the group format may also increase the perception of the availability of supportive like-minded peers. In addition, the supportive nature of the group format may be more comfortable for members of collective communities and
cultures, although results of previous research regarding the impact of the group format on condom use are mixed.2 Although no interaction was discovered, future studies should continue to explore the association between social support and relationship consensus. Both the elements of social support and the focus on achieving consensus appear to be critical intervention elements. Changes in information, IPV, self-efficacy, and negotiation within the relationship did not influence sexual barrier use. This may be due to a variety of factors. Widespread information on HIV is available in Zambia; most participants displayed high levels of knowledge about HIV, and there may have been a ceiling effect for the measure. Additionally, in Zambia, men are the primary sexual decision makers,30 and negotiation within sexual relationships may be limited. However, it is clear that relationship dynamics influence HIV transmission risk,2,31 but exactly how these dynamics contribute or can be influenced to enhance prevention is unclear. Future studies should continue this exploration.
Downloaded from jia.sagepub.com by guest on November 14, 2015
Chitalu et al
7
This study had several limitations. Serodiscordancy32 and the length of relationship have previously been related to increases in condom use. This study did not have sufficient numbers of serodiscordant couples to examine the potential influence of serostatus and did not assess the length of participant relationships. Additionally, although this study did not find changes in IPV to influence sexual barrier uptake, future studies should continue to assess gender power dynamics as potentially influential determinants of study outcomes. This team previously identified that a couple-based intervention conducted in the CHC setting in Lusaka, Zambia, was effective in increasing sexual barrier use in comparison with the standard of care.16 Although it cannot be concluded that the intervention increased sexual barrier use through enhancing these elements, it does appear that these components were associated with increased use. Translation of interventions, while including adaptation to local settings, must also emphasize the retention of evidence-based components to ensure optimal outcomes. Declaration of Conflicting Interests The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by a grant from NIH/NICHD RO1HD058481.
References 1. UNAIDS. Report on the global AIDS epidemic 2013; 2013. http://www.unaids.org/en/media/unaids/contentassets/documents/ epidemiology/2013/gr2013/UNAIDS_Global_Report_2013_en. pdf. Accessed November 5, 2013. 2. Noar SM. Behavioral interventions to reduce HIV-related sexual risk behavior: review and synthesis of meta-analytic evidence. AIDS Behav. 2008;12(3):335-353. 3. Johnson BT, Scott-Sheldon LA, Huedo-Medina TB, Carey MP. Interventions to reduce sexual risk for human immunodeficiency virus in adolescents: a meta-analysis of trials, 1985-2008. Arch Pediatr Adolesc Med. 2011;165(1):77-84. 4. Earl A, Albarracin D. Nature, decay, and spiraling of the effects of fear-inducing arguments and HIV counseling and testing: a metaanalysis of the short- and long-term outcomes of HIV-prevention interventions. Health Psychol. 2007;26(4):496-506. 5. Lacroix JM, Pellowski JA, Lennon CA, Johnson BT. Behavioural interventions to reduce sexual risk for HIV in heterosexual couples: a meta-analysis. Sex Transm Infect. 2013;89(8):620-627. 6. Albarracin D, McNatt PS, Klein CT, Ho RM, Mitchell AL, Kumkale GT. Persuasive communications to change actions: an analysis of behavioral and cognitive impact in HIV prevention. Health Psychol. 2003;22(2):166-177. 7. Prowse KM, Logue CE, Fantasia HC, Sutherland MA. Intimate partner violence and the CDC’s Best-evidence HIV risk reduction interventions [published online September 30, 2013]. Public Health Nurs. 2013.
8. Crepaz N, Horn AK, Rama SM, et al. The efficacy of behavioral interventions in reducing HIV risk sex behaviors and incident sexually transmitted disease in black and Hispanic sexually transmitted disease clinic patients in the United States: a meta-analytic review. Sex Transm Dis. 2007;34(6):319-332. 9. Scott-Sheldon LA, Johnson BT. Eroticizing creates safer sex: a research synthesis. J Prim Prev. 2006;27(6):619-640. 10. Edgar T, Noar SM, Murphy B. Communication Skills Training in HIV Prevention Interventions. New York: Lawrence Erlbaum; 2008. 11. Noar SM. An interventionist’s guide to AIDS behavioral theories. AIDS Care. 2007;19(3):392-402. 12. Crepaz N, Lyles CM, Wolitski RJ, et al. Do prevention interventions reduce HIV risk behaviours among people living with HIV? A meta-analytic review of controlled trials. AIDS. 2006;20(2): 143-157. 13. Johnson BT, Carey MP, Chaudoir SR, Reid AE. Sexual risk reduction for persons living with HIV: research synthesis of randomized controlled trials, 1993 to 2004. J Acquir Immune Defic Syndr. 2006;41(5):642-650. 14. Jones DJ, Chitalu N, Ndubani P, et al. Sexual risk reduction among Zambian couples. SAHARA J. 2009;6(2):69-75. 15. Jones DL, Weiss SM, Waldrop-Valverde D, Chitalu N, Mumbi M, Vamos S. Community-based risk reduction in Zambia. Open Health Serv Pol J. 2010;1:38-44. 16. Jones D, Weiss SM, Arheart K, Cook R, Chitalu N. Implementation of HIV prevention interventions in resource limited settings: the partner project [published online August 21, 2013]. J Community Health. 2013. 17. Lusaka ranks highest in HIV prevalence rate. Lusaka Times. June 26, 2013. http://www.lusakatimes.com/2013/06/26/lusaka-rankshighest-in-hiv-prevalence-rate/. Accessed January 17, 2014. 18. Neumann MS, Johnson WD, Semaan S, et al. Review and metaanalysis of HIV prevention intervention research for heterosexual adult populations in the United States. J Acquir Immune Defic Syndr. 2002;30(Suppl 1):S106-S117. 19. Herbst JH, Beeker C, Mathew A, et al. The effectiveness of individual-, group-, and community-level HIV behavioral riskreduction interventions for adult men who have sex with men: a systematic review. Am J Prev Med. 2007;32(Suppl 4):S38-S67. 20. Jones DL, Weiss SM, Bhat GJ, Bwalya V. Influencing sexual practices among HIV-positive Zambian women. AIDS Care. 2006;18(6):629-634. 21. Jones DL, Weiss SM, Bhat GJ, Feldman D, Bwalya V, Budash D. A sexual barrier intervention for HIVþ/ Zambian women: acceptability and use of vaginal chemical barriers. J Multicult Nurs Health. 2004;10(1):24-31. 22. Spanier GB. Measuring dyadic adjustment: new scales for assessing the quality of marriage and similar dyads. J Marriage Fam. 1976;38(1):15-28. 23. Straus MA. Measuring intrafamily conflict and violence: the conflict tactics scales. J Marriage Fam. 1979;41(1):75-88. 24. Fisher JD, Fisher WA, Misovich SJ, Kimble DL, Malloy TE. Changing AIDS risk behavior: effects of an intervention emphasizing AIDS risk reduction information, motivation, and behavioral skills in a college student population. Health Psychol. 1996;15(2):114-123.
Downloaded from jia.sagepub.com by guest on November 14, 2015
8
Journal of the International Association of Providers of AIDS Care
25. Smith KW, McGraw SA, Costa LA, McKinlay JB. A self-efficacy scale for HIV risk behaviors: development and evaluation. AIDS Educ Prev. 1996;8(2):97-105. 26. Carver CS. You want to measure coping but your protocol’s too long: consider the brief COPE. Int J Behav Med. 1997;4(1): 92-100. 27. Jones DL, Weiss SM, Malow R, et al. A brief sexual barrier intervention for women living with AIDS: acceptability, use, and ethnicity. J Urban Health. 2001;78(4):593-604. 28. Jones DL, Weiss SM, Chitalu N, et al. Acceptability and use of sexual barrier products and lubricants among HIV-seropositive Zambian men. AIDS Patient Care STDS. 2008;22(12):1015-1020. 29. Kennedy CE, Medley AM, Sweat MD, O’Reilly KR. Behavioural interventions for HIV positive prevention in developing countries:
a systematic review and meta-analysis. Bull World Health Organ. 2010;88(8):615-623. 30. Jones DL, Ross D, Weiss SM, Bhat G, Chitalu N. Influence of partner participation on sexual risk behavior reduction among HIV-positive Zambian women. J Urban Health. 2005;82(3 suppl 4):iv92-iv100. 31. Vamos S, Cook R, Chitalu N, Mumbi M, Weiss SM, Jones D. Quality of relationship and sexual risk behaviors among HIV couples in Lusaka, Zambia. AIDS Care. 2013;25(9):1102-1108. 32. Bouhnik AD, Preau M, Schiltz MA, Lert F, Obadia Y, Spire B. Unprotected sex in regular partnerships among homosexual men living with HIV: a comparison between sero-nonconcordant and seroconcordant couples (ANRS-EN12-VESPA Study). AIDS. 2007;21(Suppl 1):S43-S48.
Downloaded from jia.sagepub.com by guest on November 14, 2015
The Impact of Key HIV Intervention Components as Predictors of Sexual Barrier Use: The Zambia Partner Project
Journal of the International Association of Providers of AIDS Care 1-8 ª The Author(s) 2014 Reprints and permission: sagepub.com/journalsPermissions.nav DOI: 10.1177/2325957414520980 jiapac.sagepub.com
Ndashi Chitalu, MD1, Mirriam Mumbi, CNM1, Ryan Cook, BA2, Stephen M. Weiss, PhD, MPH2, and Deborah Jones, PhD2
Abstract Behavioral interventions have utilized a variety of strategies and components to reduce HIV risk. This article describes the partner intervention, a couple-based group HIV risk reduction intervention implemented in 6 urban community health clinics in Lusaka, Zambia, and examines the components of the intervention and their relationship with condom use. Couple members completed assessments on condom use, acceptability, willingness to use condoms, communication, intimate partner violence (IPV), self-efficacy, and HIV information at baseline and 6 months’ follow-up. This study examined the relative impact of elements of the intervention as predictors of condom use. Changes in acceptability had the greatest overall influence on condom use, followed by social support, relationship consensus, and willingness to use condoms. Changes in self-efficacy, IPV, negotiation, and information had no influence. Results support the use of multidimensional approaches in behavioral interventions and highlight the importance of identifying critical elements of interventions to maximize risk reduction outcomes. Keywords HIV, intervention, Africa, couples
Introduction 1
The HIV epidemic has begun to decline, which is likely due, in part, to the efficacy of programs developed and implemented globally to reduce sexual risk behavior.2 These programs rely on a variety of methods and elements to influence behavior and reduce the risk of HIV transmission and/or infection.2 Program interventions are tailored to the cultural context of the target population and focus on elements such as information about HIV and testing, transmission of the virus, and prevention strategies, such as condom use.3,4 Interventions may also target communication skills,5 sexual negotiation,6 intimate partner violence (IPV),7 self-efficacy,8 perceptions of sexual barrier use,9 and skill training10,11 and adapt these techniques to the intended audience (eg, people living with HIV).12,13 However, the combination of diverse elements in behavioral interventions within specific settings makes identification of the critical intervention elements’ underlying risk reduction challenging. The partner intervention is a couple-based HIV risk reduction intervention that was culturally tailored to the local population in the capital province of Lusaka, Zambia, adapted for use in the community health clinic setting14,15 and implemented in 6 community health clinics.16 Nationwide, Zambia has a generalized HIV epidemic, with over 1 million persons living with the virus and high rates of HIV in urban Lusaka (20.8%).17 The elements of the partner intervention, that is, information,
skill building, and cognitive behavioral (CB) strategies targeting self-efficacy, sexual communication, and sexual barrier acceptability, were hypothesized to enhance the uptake of sexual barrier products. The intervention format, gender concordant groups, has been found to be effective in reducing risk behavior18,19 and shown to enhance the uptake of sexual barrier products in this setting,20 in comparison with traditional individual health education21 and with the standard of care.16 This article reviews the components of the Zambian intervention and examines the relative impact of intervention elements, that is, information, social support, self-efficacy, communication, IPV reduction, and sexual barrier product acceptability, on sexual barrier product use.
1
Department of Pediatrics, University Teaching Hospital, University of Zambia School of Medicine, Lusaka, Zambia 2 Department of Psychiatry and Behavioral Sciences, University of Miami Miller School of Medicine, Miami, FL, USA Corresponding Author: Deborah Jones, Department of Psychiatry and Behavioral Sciences, University of Miami Miller School of Medicine, 1400 NW 10th Ave, Suite 404A, Miami, FL 33136, USA. Email: [email protected]
Downloaded from jia.sagepub.com by guest on November 14, 2015
2
Journal of the International Association of Providers of AIDS Care
Methods
Measures
This study was conducted from March 2009 to March 2012; ethical review and approval were obtained prior to study onset from the University of Zambia Research Ethics Committee and the University of Miami Miller School of Medicine Institutional Review Board, in accordance with the provisions of the US Department of Health and Human Services and the Zambian government. Study sites (n ¼ 6 urban community health centers [CHCs]) enrolled their first cohort (a group of 8 couples) into the control condition and all subsequent cohorts into the intervention condition (a total of 5 cohorts were enrolled per site).
Demographics. Participants’ age, educational level, employment status, religion, marital status, number of children, and antiretroviral therapy status (if HIV positive) were assessed.
Recruitment and Enrollment Recruitment of participants was conducted by community mobilizers at the 6 CHCs. Mobilizers referred potential participants to assessors at their local CHC for screening and enrollment. Participants were 18 years of age or older, sexually active within the past month, in a couple relationship for at least 6 months, and 1 or both members of the couple were seropositive. HIV serostatus was verified for the sero-positive status of one partner or both the partners; prior to enrollment, seronegative participants were referred for HIV testing with their partner. Approximately 30% of couples screened for enrollment were not eligible due to the lack of sexual activity in the previous month or failure to demonstrate couple status. All informed consent, assessment, and intervention materials were translated into the major Lusakan languages (Bemba and Nyanja) in accordance with a ‘‘cultural brokerage’’ process to ensure comprehension and validity of data. Cultural brokerage is a process of translation and back translation, followed by discussion of items of disagreement, to arrive at the most accurate and meaningful translation. Following the provision of informed consent, enrolled participants (n ¼ 240 couples) were invited to complete a baseline assessment. All study assessments were completed using an audio computerassisted self-interview system that collected information with minimal assistance from the assessors. The participants were left alone to listen, read, and respond to recorded questions presented on the computer screen and through head phones; the participants recorded their answers using a touch screen connected to the computer. All responses were stored by the computer, and the information was uploaded and transformed to SPSS. Participants were compensated (K50 000 Zambian Kwacha *US$10) for each assessment but were not compensated for intervention attendance. Participants enrolled in the intervention condition were provided with an appointment for the intervention; those in the control condition were in a delayed start condition and were offered the intervention after 12 months, following their final assessment. During the waiting period, only the control condition participants completed the assessments. This article focuses on the elements of the intervention, which influenced sexual barrier use; therefore, the 40 couples in the control condition were not included in the analytic data set (included n ¼ 200 couples).
Sexual barrier use. Participants reported each incidence of sexual intercourse each day of the preceding week and whether or not a condom was used, in a diary format. A pictorial representation of condoms was provided to enhance comprehension. Condom use in the past week was calculated by dividing the total number of condoms used during the week by the total number of sex acts. Because not all participants had sex within the week before assessments, they also provided an assessment of their typical condom use (in general) using a 5-point Likert-type scale (5 ¼ every time; 4 ¼ almost every time; 3 ¼ sometimes; 2 ¼ almost never; and 1 ¼ never). Willingness to use condoms and acceptability of condoms. This measure was adapted from the University of California at San Francisco Center for AIDS Prevention Studies Barrier Questionnaire, and the current and previous use of sexual barriers and willingness to use sexual barriers using Likert-type scales were measured. Acceptability, here defined as ‘‘liking’’ a product, was measured by asking each individual how much he or she liked using male condoms. Participants responded using the scale ‘‘like very much ¼ 5; like somewhat ¼ 4; neutral ¼ 3; dislike somewhat ¼ 2; strongly dislike ¼ 1, and never used ¼ 0.’’ Participants also reported their ‘‘willingness to use’’ male condoms. This item was rated as ‘‘very willing to use ¼ 4; moderately willing ¼ 3; slightly willing ¼ 2; and not at all willing ¼ 1.’’ Communication—relationship consensus. The relationship consensus subscale of the Dyadic Adjustment scales (DAS)22 assessed achievement of consensus in couples’ relationships. Higher scores on the DAS consensus subscale indicate higher consensus on difficult topics such as financial, religious, and sexual issues. The consensus subscale demonstrated high internal consistency in this sample (a ¼ .86). Communication—negotiation and IPV. A modified version of the 17-item Conflict Tactics scale23 was used to measure negotiation (eg, discussion and problem solving) and IPV (eg, pushing, hitting, and slapping). Participants indicated the number of times their partners had engaged in these behaviors during the preceding month. Responses were scored using a Likert-type scale of 0 (never) to 6 (more than 20 times). Information. HIV knowledge was assessed using an adaptation of the AIDS-Related Knowledge Test.24 This adaptation from the 10-item scale assesses HIV risk and prevention-related knowledge. Items reflect information about HIV transmission, condom use, and AIDS-related knowledge and are responded to as yes, no, or don’t know. The AIDS-Related Knowledge Test was scored for the number of correct responses and has acceptable internal consistency (a ¼ .73).
Downloaded from jia.sagepub.com by guest on November 14, 2015
Chitalu et al
3
Self-efficacy. Sexual self-efficacy was measured using the Sexual Self Efficacy Questionnaire.25 This 7-item measure assesses the level of sexual self-efficacy by asking participants to rate their level of comfort with specific sexual risk reduction behaviors using a Likert-type scale, including purchasing and using male or female condoms, refusing unprotected sex, and negotiating condom use. The self-efficacy score has demonstrated acceptable internal consistence (a ¼ .77). Social support. Social support was measured using the Brief COPE-Revised.26 The 38-item revised version of the Brief COPE measures the extent to which an individual engages in various forms of coping strategies in response to life stressors; the instrumental and emotional support subscales of the Brief COPE-revised were combined into a 4-item social support measure. Items are rated from 0 (I haven’t been doing this at all) to 3 (I’ve been doing this a lot). Higher scores are associated with greater use of the coping strategy. The 2 subscales used to create the social support measure were highly correlated (r ¼ .61, P < .001), and the social support measure demonstrated acceptable internal consistency in this sample (a ¼ .76).
Intervention Topics The partner intervention began as a cross-cultural adaptation of an evidence-based HIV prevention intervention, the New Opportunities for Women Project.27 Key elements of the intervention were identified by focus group and pilot research conducted from 1999 to 2003.20,28 From 2003 to 2008, the partner project was translated, tested, and implemented into the CHC setting.15 The theory-based intervention14 consisted of 4 sessions presented in same-gender groups of 8 participants delivered over 4 weeks; sessions were led by trained facilitators certified in HIV counseling and/or nursing. Sessions began by providing participants with an assurance that confidentiality is a cardinal factor during the groups to alleviate fears and enhance trust within the groups and toward the facilitators leading the intervention. Participants were informed that the sessions would take place at an appointed time for roughly 2 to 2.5 hours. Audio taping was used to record comments participants shared among each other, to review what was discussed, and to provide information for quality control by the investigators. Participants were also informed that the tapes used are kept under lock and key. Cognitive behavioral strategies: acceptability and willingness to use sexual barriers. This component was included in all of the sessions and often elicited excitement and laughter from participants. Discussion of the techniques stimulated extensive conversation, and, if not mindful of the time, facilitators could find themselves exceeding the amount of time planned. Cognitive behavioral strategies address how a person reacts to a particular situation before any behavior occurs, describing the process of reacting first with the mind, the emotions, and later with behavior. This topic was both difficult and interesting for participants as it made them aware of the source from which
their actions arose. In presenting CB, it was important to provide a number of examples, such as first impressions of a female condom might elicit thoughts such as ‘‘it is big, it is ugly’’ or ‘‘I am scared to use it’’ or ‘‘I am excited about this,’’ leading to positive or negative behavioral outcomes. This section also addressed coping, developing problem-solving skills, and applying stress reduction strategies, using CB strategies such as linking thoughts to emotions, breaking down the process of reacting with thoughts and feelings, identifying automatic thoughts, rational thought replacement, and cognitive restructuring, and changing how we think. Participants were shown that changing how they think can lead to changing their behavior, which may result in a more positive perception of condom use, and thus increase the likelihood of practicing safer sex. Communication. Communication about sex between men and women is an essential component of HIV prevention, and the topic was incorporated into most sections of the intervention. Sexual negotiation was conceptualized as having the goal of reaching an agreement on sexual topics such as the use of condoms. However, communication training did not only address sex but also included strategies that can render all types of communication between partners more positive and effective. This study encouraged participants to talk about things that are not normally discussed by couples in Zambia, such as HIV reinfection, sexually transmitted infections (STIs), unwanted pregnancy, and many other sensitive issues. In order to illustrate effective communication, role-plays were performed within the groups, in which one participant acted as a man and another as a woman. During the role-play, the ‘‘couple’’ focused on the importance of achieving a level of understanding with their partner while ensuring no conflict would ensue between the two people. Both men and women presented their partners using their typical manner of speech, as in the tone of voice and the way they might ‘‘handle’’ their partner, explaining the use of sexual barriers and why. As most participants were shy in the beginning, facilitators illustrated role-playing before asking participants to perform for the group; most appeared to enjoy role-playing during the sessions. Intimate partner violence. It was recognized that some couples do not continuously have a rosy relationship but may also have some degree of violence in their relationships. The IPV was integrated into communication topics, and some participants shared psychological abuse, as in not communicating with the partner or screaming and shouting at their partners. Some also shared experiences of being beaten with objects and described having resorted to running away from home, reporting their situation to close relatives or friends, and also turning to IPV Victim Support Units. Participants were encouraged to share concerns about the potential for IPV and to establish safety plans in the event that IPV became an issue. Those women who became estranged from their partners were provided with counseling support and scheduled for sessions that did not coincide with their partner’s sessions.
Downloaded from jia.sagepub.com by guest on November 14, 2015
4
Journal of the International Association of Providers of AIDS Care
Information: STDs and HIV. As there are many forms of STIs, including HIV, this topic utilized a poster showing different types of STIs, including chlamydia, syphilis, human papilloma virus, genital herpes, hepatitis B, and HIV. Participants were encouraged to develop a clear picture of what infectious diseases are and look like, enabling them to understand the ways these infections could be avoided or the transmissions could be prevented. Through discussion, participants were queried on their level of understanding about STIs and HIV. Information: safer sex and reproductive issues. This discussion of relative safety of different types of sexual behavior was an adapted hierarchy of safer sex; facilitators began by splitting the term ‘‘safer sex,’’ safer and sex, to facilitate understanding. Group members generally understood it to mean the prevention of the transmission of HIV to partners, and most were aware that condoms are the best safer sex method. Abstinence was controversial as most were married or cohabiting couples, but the majority had experienced abstinence as a period during which the other partner was away, sick, or sexually fasting, related to prayer. The discussion included nonpenetrative safer sex, such as sexual friction between the thighs or breasts of a woman, and sexual practices such as the use of vaseline or lubricants in the vagina or coitus interruptus. The discussion provided the opportunity to correct misunderstandings among the participants regarding safer sex. Very few participants knew that medical male circumcision also makes sex safer, and very few mentioned masturbation as safer sex, perhaps due to shyness to discuss this practice in public. However, participants, especially women, opened up to discuss masturbation when the facilitator initiated the discussion topic. Discussion of reproductive issues began by exploring desires to have children, addressing participants’ awareness of issues surrounding safer conception, and focusing on times during the menstrual cycle when a woman is most likely to conceive. Participants who wished to conceive were guided on how to chart their cycles to enable them to restrict unprotected sex to fertile periods, thereby reducing the potential for transmission to partners. The female reproductive cycle was explained in detail with a poster diagram and a female anatomic model, to provide participants with a full understanding of the process of conception, from sexual intercourse to delivery. Knowing the days when the woman is fertile was used to help participants know when the woman cannot become pregnant, enabling her to use dual methods for protection against HIV during at least two-thirds of her cycle (eg, condoms plus contraceptives, diaphragms, or caps). Very few participants considered the increased potential for transmission during conception and were unaware of strategies to plan safer conception. In fact, most women did not engage in any communication with their spouses concerning getting pregnant. Discussion also addressed protecting mother, father, and baby from STIs and HIV during pregnancy using safer sex practices. Participants were reminded of the benefits of creating and maintaining a good relationship with their doctors preconception, during pregnancy, and postpartum and of the need
to discuss having a baby with their doctor as the first step in planning conception. Women discussed the tests typically performed preconception, that is, pap smear, HIV, and CD4 count to determine the level of immunity in the body. Skill training. Sessions included opportunity for skill training in (1) the use of male and female condoms and vaginal lubricants, (2) positive communication and sexual negotiation strategies, and (3) CB strategies such as reframing perceptions to change behavior. Male and female condoms and lubricants were supplied to each and every individual upon completion of each session. Skill training and group sessions were used to enhance self-efficacy and social support for engaging in new behaviors.
Statistical Analyses In order to examine the relationship between changes in elements targeted by the intervention and changes in condom use, change scores between baseline and follow-up were computed. Follow-up was 6 months postbaseline, which was approximately 4 to 5 months after participants completed the intervention. Then, changes in condom use were regressed on changes in each element of the intervention separately. Differences between genders were tested by testing interactions with gender, and gender effects were dropped if nonsignificant. Subsequently, a full-factorial combination of elements that were found to be significant was simultaneously entered into a multivariable model. Nonsignificant interactions and main effects were removed, beginning with the least significant, until a final model was determined. Changes in condom use in the week preceding assessment was the primary outcome; however, due to the small sample of participants reporting sex at both time points, a measure of condom use ‘‘in general’’ was also analyzed. Because this study assessed couples, models were estimated using generalized estimating equations including a residual correlation for person within couple. All analyses were conducted using SAS PROC GENMOD (SAS 9.3; SAS Corporation, Cary, North Carolina) at a 2-tailed level of significance of P ¼ .05.
Results Demographics Participants (N ¼ 400; 200 couples) averaged 38 years old with 8 years of education and most were unemployed (n ¼ 247, 62%). The majority of participants identified as Protestant (n ¼ 207, 52%) or Catholic (n ¼ 138, 35%). Nearly all were married (n ¼ 388, 97%) and had children (n ¼ 356, 89%); the mean number of children was 3. In all, 371 (93%) participants were HIV sero-positive and 29 (7%) were sero-negative, resulting in 171 seroconcordant and 29 serodiscordant couples. Of the 371 HIV-positive participants, 236 (64%) were taking antiretroviral medications. Table 1 presents further descriptions of participant demographics.
Downloaded from jia.sagepub.com by guest on November 14, 2015
Chitalu et al
5
Table 1. Demographics. Characteristic (N ¼ 400) Age, y Years of education Employment status Employed Unemployed Religion Protestant Catholic Other/not religious Marital status Married Not married At least 1 child No children Number of children HIV status HIV positive HIV negative ARV status (n ¼ 371 HIV positive) On ART Not on ART
M (SD)/n (%) 38.2 (8.2) 8.2 (3.3) 153 (38%) 247 (62%) 207 (52%) 138 (35%) 55 (14%) 388 (97%) 12 (3%) 356 (89%) 44 (11%) 3 (2) 371 (93%) 29 (7%) 236 (64%) 135 (36%)
Abbreviations: ART, antiretroviral therapy; ARV, antiretroviral; M, mean; SD, standard deviation.
Of the 400 participants, 138 reported having sex at least once during both the week before the baseline assessment and the week before the follow-up assessment. The proportion of condom-protected sexual acts in the past week at baseline was 69.0% (standard deviation [SD] ¼ .41), increasing to 83.0% at follow-up (SD ¼ .31). The mean increase in weekly condom use was 14.0% (95% confidence interval [CI] ¼ 0.07-0.21), which did not differ between HIV-negative and -positive participants (t ¼ .73, P ¼ .467) or between men and women (t ¼ .11, P ¼ .910). Similarly, 318 participants contributed both baseline and follow-up general condom use scores. General condom use, which ranged from 1 ¼ never used condoms to 5 ¼ always used condoms, was 4.19 (SD ¼ 1.19) at baseline and increased to 4.44 at follow-up (SD ¼ 1.03). Mean change in general condom use was 0.24 (95% CI ¼ 0.10-0.39), which also did not differ by HIV serostatus (t ¼ .64, P ¼ .522) or gender (t ¼ .64, P ¼ .521). Changes in weekly and general condom use were correlated (r ¼ .46, P < .001).
Impact of Elements of the Intervention on Condom Use In order to determine the elements of the intervention that were associated with changes in sexual barrier use, changes in condom use were regressed on changes in condom acceptability, communication, IPV, information, self-efficacy, and social support. The first series of analyses considered each element separately and were conducted for both condom use in the past week and condom use in general. Results of these analyses, including b coefficients, standard errors, and 95% CIs, are presented in Table 2. In summary, changes in sexual barrier acceptability and willingness to use sexual barriers as well as
relationship consensus, self-efficacy, and social support impacted sexual barrier use. Changes in information, IPV, and negotiation had no impact. Significant predictors and interactions were entered into multivariable models predicting weekly condom use and condom use in general. The results of these analyses are presented in Table 3, including both unstandardized and standardized b coefficients along with standard errors and 95% CIs. As each predictive variable was measured using a different scale, the percentages presented are not directly comparable. The standardized b coefficients are more comparable as they represent the effect of 1 SD change in the predictor on 1 SD of the outcome. In summary, a significant positive relationship between changes in sexual barrier acceptability and weekly condom use was found. A 1-point increase in the acceptability of male condoms was associated with a 5% increase in condom use (standardized b ¼ .228). Social support positively impacted weekly condom use, such that a 1-unit increase in social support was associated with a 2% increase in condom use (b ¼ .211). Finally, willingness to use male condoms also impacted condom use, such that a 1 point increase in willingness was associated with an 7% increase in use (b ¼ .146). No interactions were significant, and self-efficacy did not impact condom use in multivariable analysis. Sexual barrier acceptability also impacted condom use in general; increases in the acceptability of male condoms were associated with significant increases in general condom use (b ¼ .10; standardized b ¼ .159). Additionally, relationship consensus positively impacted general condom use (b ¼ .02; b ¼ .182). No interaction between acceptability and relationship consensus was found.
Discussion This study examined the relative impact of elements of a couple-based group sexual risk reduction intervention on sexual barrier use. Changes in acceptability had the greatest overall influence on sexual barrier use, followed by social support, relationship consensus, and willingness to use sexual barriers. Self-efficacy was associated with the use of sexual barriers, but increased self-efficacy did not contribute to increased barrier use in multivariable analysis. Finally, negotiation, IPV, and information had no apparent influence on sexual barrier use. Study results provide further support for the predictive relationship between acceptability and self-reported willingness to use barriers and sexual barrier use, in line with earlier findings by this team using individual rather than dyadic analyses.28 One of the key elements of CB therapy is the premise that attitudes influence behavior, for example, acceptability of condoms will predict their use. In this study, acceptability may have been increased by CB strategies designed to promote positive attitudes and reduce negative opinions regarding condom use. Condoms are highly effective in the prevention of HIV transmission, but their impact as a prevention strategy has been limited by low acceptability and inconsistent use. Metaanalyses of positive prevention studies in developing countries
Downloaded from jia.sagepub.com by guest on November 14, 2015
6
Journal of the International Association of Providers of AIDS Care
Table 2. Bivariate Analyses of Changes in Condom Use Regressed on Changes in Condom Acceptability, Communication, IPV, Information, Self-Efficacy, and Social Support.a
Condom use in past week Male condom acceptability Willingness to use male condoms Relationship consensus Negotiation Intimate partner violence HIV/AIDS-related knowledge Self-efficacy Social support Condom use in general Male condom acceptability Willingness to use male condoms Relationship consensus Negotiation Intimate partner violence HIV/AIDS-related knowledge Self-efficacy Social support
b (SE)
P
95% CI (b)
.060 (.019) .113 (.036) .005 (.003) .001 (.007) .004 (.004) .030 (.021) .010 (.004) .026 (.009)
.002 .002 .100 .894 .404 .153 .018 .004
0.023-0.096 0.043-0.183 0.011-0.001 0.012-0.014 0.005-0.012 0.011-0.072 0.002-0.019 0.008-0.043
.101 (.038) .046 (.096) .021 (.007) .007 (.015) .019 (.012) .065 (.049) .001 (.010) .029 (.020)
.008 .636 .005 .646 .134 .188 .881 .146
0.026-0.176 0.143-0.234 0.006-0.035 0.036-0.022 0.043-0.006 0.032-0.161 0.018-0.020 0.010-0.068
Abbreviations: CI, confidence interval; IPV, intimate partner violence; SE, standard error. a Statistically significant variables are indicated in boldface.
Table 3. Multivariable Analysis of Changes in Condom Use Regressed on Significant Bivariate Predictors.a
Weekly condom use Male condom acceptability Willingness to use male condoms Social support Condom use in general Male condom acceptability Relationship consensus
b (SE)
P
95% CI (b)
b (SE)
95% CI (b)
.049 (.016) .071 (.032) .022 (.009)
.006 .019 .012
0.018-0.081 0.007-0.134 0.005-0.039
.228 (.084) .146 (.067) .211 (.084)
0.063-0.393 0.014-0.277 0.047-0.375
.102 (.040) .021 (.007)
.010 .004
0.025-0.180 0.007-0.035
.159 (.061) .182 (.063)
0.038-0.279 0.059-0.305
Abbreviations: CI, confidence interval; IPV, intimate partner violence; SE, standard error. a Changes in condom use were regressed on changes in condom acceptability, communication, IPV, information, self-efficacy, and social support individually. Significant predictors were entered simultaneously into a multivariable model; the results are shown here.
have reported that interventions targeting HIV sero-positive individuals are successful in increasing condom use, in both seroconcordant and discordant relationships.29 Recent studies, for example, microbicides and preexposure prophylaxis (PreP), have highlighted the importance of assessing the acceptability of prevention strategies as a predictor of uptake. In fact, low product acceptability may be the Achilles’ heel of HIV prevention. Enhancement of product acceptability is likely a critical intervention element and has important implications for the uptake of existing prevention strategies as well as new strategies such as PreP. Social support and relationship consensus also played a key role in the uptake of sexual barrier products. In psychotherapy, the group format is often used to increase the overall impact of the intervention by providing a variety of viewpoints on interpersonal issues, but the group format may also increase the perception of the availability of supportive like-minded peers. In addition, the supportive nature of the group format may be more comfortable for members of collective communities and
cultures, although results of previous research regarding the impact of the group format on condom use are mixed.2 Although no interaction was discovered, future studies should continue to explore the association between social support and relationship consensus. Both the elements of social support and the focus on achieving consensus appear to be critical intervention elements. Changes in information, IPV, self-efficacy, and negotiation within the relationship did not influence sexual barrier use. This may be due to a variety of factors. Widespread information on HIV is available in Zambia; most participants displayed high levels of knowledge about HIV, and there may have been a ceiling effect for the measure. Additionally, in Zambia, men are the primary sexual decision makers,30 and negotiation within sexual relationships may be limited. However, it is clear that relationship dynamics influence HIV transmission risk,2,31 but exactly how these dynamics contribute or can be influenced to enhance prevention is unclear. Future studies should continue this exploration.
Downloaded from jia.sagepub.com by guest on November 14, 2015
Chitalu et al
7
This study had several limitations. Serodiscordancy32 and the length of relationship have previously been related to increases in condom use. This study did not have sufficient numbers of serodiscordant couples to examine the potential influence of serostatus and did not assess the length of participant relationships. Additionally, although this study did not find changes in IPV to influence sexual barrier uptake, future studies should continue to assess gender power dynamics as potentially influential determinants of study outcomes. This team previously identified that a couple-based intervention conducted in the CHC setting in Lusaka, Zambia, was effective in increasing sexual barrier use in comparison with the standard of care.16 Although it cannot be concluded that the intervention increased sexual barrier use through enhancing these elements, it does appear that these components were associated with increased use. Translation of interventions, while including adaptation to local settings, must also emphasize the retention of evidence-based components to ensure optimal outcomes. Declaration of Conflicting Interests The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by a grant from NIH/NICHD RO1HD058481.
References 1. UNAIDS. Report on the global AIDS epidemic 2013; 2013. http://www.unaids.org/en/media/unaids/contentassets/documents/ epidemiology/2013/gr2013/UNAIDS_Global_Report_2013_en. pdf. Accessed November 5, 2013. 2. Noar SM. Behavioral interventions to reduce HIV-related sexual risk behavior: review and synthesis of meta-analytic evidence. AIDS Behav. 2008;12(3):335-353. 3. Johnson BT, Scott-Sheldon LA, Huedo-Medina TB, Carey MP. Interventions to reduce sexual risk for human immunodeficiency virus in adolescents: a meta-analysis of trials, 1985-2008. Arch Pediatr Adolesc Med. 2011;165(1):77-84. 4. Earl A, Albarracin D. Nature, decay, and spiraling of the effects of fear-inducing arguments and HIV counseling and testing: a metaanalysis of the short- and long-term outcomes of HIV-prevention interventions. Health Psychol. 2007;26(4):496-506. 5. Lacroix JM, Pellowski JA, Lennon CA, Johnson BT. Behavioural interventions to reduce sexual risk for HIV in heterosexual couples: a meta-analysis. Sex Transm Infect. 2013;89(8):620-627. 6. Albarracin D, McNatt PS, Klein CT, Ho RM, Mitchell AL, Kumkale GT. Persuasive communications to change actions: an analysis of behavioral and cognitive impact in HIV prevention. Health Psychol. 2003;22(2):166-177. 7. Prowse KM, Logue CE, Fantasia HC, Sutherland MA. Intimate partner violence and the CDC’s Best-evidence HIV risk reduction interventions [published online September 30, 2013]. Public Health Nurs. 2013.
8. Crepaz N, Horn AK, Rama SM, et al. The efficacy of behavioral interventions in reducing HIV risk sex behaviors and incident sexually transmitted disease in black and Hispanic sexually transmitted disease clinic patients in the United States: a meta-analytic review. Sex Transm Dis. 2007;34(6):319-332. 9. Scott-Sheldon LA, Johnson BT. Eroticizing creates safer sex: a research synthesis. J Prim Prev. 2006;27(6):619-640. 10. Edgar T, Noar SM, Murphy B. Communication Skills Training in HIV Prevention Interventions. New York: Lawrence Erlbaum; 2008. 11. Noar SM. An interventionist’s guide to AIDS behavioral theories. AIDS Care. 2007;19(3):392-402. 12. Crepaz N, Lyles CM, Wolitski RJ, et al. Do prevention interventions reduce HIV risk behaviours among people living with HIV? A meta-analytic review of controlled trials. AIDS. 2006;20(2): 143-157. 13. Johnson BT, Carey MP, Chaudoir SR, Reid AE. Sexual risk reduction for persons living with HIV: research synthesis of randomized controlled trials, 1993 to 2004. J Acquir Immune Defic Syndr. 2006;41(5):642-650. 14. Jones DJ, Chitalu N, Ndubani P, et al. Sexual risk reduction among Zambian couples. SAHARA J. 2009;6(2):69-75. 15. Jones DL, Weiss SM, Waldrop-Valverde D, Chitalu N, Mumbi M, Vamos S. Community-based risk reduction in Zambia. Open Health Serv Pol J. 2010;1:38-44. 16. Jones D, Weiss SM, Arheart K, Cook R, Chitalu N. Implementation of HIV prevention interventions in resource limited settings: the partner project [published online August 21, 2013]. J Community Health. 2013. 17. Lusaka ranks highest in HIV prevalence rate. Lusaka Times. June 26, 2013. http://www.lusakatimes.com/2013/06/26/lusaka-rankshighest-in-hiv-prevalence-rate/. Accessed January 17, 2014. 18. Neumann MS, Johnson WD, Semaan S, et al. Review and metaanalysis of HIV prevention intervention research for heterosexual adult populations in the United States. J Acquir Immune Defic Syndr. 2002;30(Suppl 1):S106-S117. 19. Herbst JH, Beeker C, Mathew A, et al. The effectiveness of individual-, group-, and community-level HIV behavioral riskreduction interventions for adult men who have sex with men: a systematic review. Am J Prev Med. 2007;32(Suppl 4):S38-S67. 20. Jones DL, Weiss SM, Bhat GJ, Bwalya V. Influencing sexual practices among HIV-positive Zambian women. AIDS Care. 2006;18(6):629-634. 21. Jones DL, Weiss SM, Bhat GJ, Feldman D, Bwalya V, Budash D. A sexual barrier intervention for HIVþ/ Zambian women: acceptability and use of vaginal chemical barriers. J Multicult Nurs Health. 2004;10(1):24-31. 22. Spanier GB. Measuring dyadic adjustment: new scales for assessing the quality of marriage and similar dyads. J Marriage Fam. 1976;38(1):15-28. 23. Straus MA. Measuring intrafamily conflict and violence: the conflict tactics scales. J Marriage Fam. 1979;41(1):75-88. 24. Fisher JD, Fisher WA, Misovich SJ, Kimble DL, Malloy TE. Changing AIDS risk behavior: effects of an intervention emphasizing AIDS risk reduction information, motivation, and behavioral skills in a college student population. Health Psychol. 1996;15(2):114-123.
Downloaded from jia.sagepub.com by guest on November 14, 2015
8
Journal of the International Association of Providers of AIDS Care
25. Smith KW, McGraw SA, Costa LA, McKinlay JB. A self-efficacy scale for HIV risk behaviors: development and evaluation. AIDS Educ Prev. 1996;8(2):97-105. 26. Carver CS. You want to measure coping but your protocol’s too long: consider the brief COPE. Int J Behav Med. 1997;4(1): 92-100. 27. Jones DL, Weiss SM, Malow R, et al. A brief sexual barrier intervention for women living with AIDS: acceptability, use, and ethnicity. J Urban Health. 2001;78(4):593-604. 28. Jones DL, Weiss SM, Chitalu N, et al. Acceptability and use of sexual barrier products and lubricants among HIV-seropositive Zambian men. AIDS Patient Care STDS. 2008;22(12):1015-1020. 29. Kennedy CE, Medley AM, Sweat MD, O’Reilly KR. Behavioural interventions for HIV positive prevention in developing countries:
a systematic review and meta-analysis. Bull World Health Organ. 2010;88(8):615-623. 30. Jones DL, Ross D, Weiss SM, Bhat G, Chitalu N. Influence of partner participation on sexual risk behavior reduction among HIV-positive Zambian women. J Urban Health. 2005;82(3 suppl 4):iv92-iv100. 31. Vamos S, Cook R, Chitalu N, Mumbi M, Weiss SM, Jones D. Quality of relationship and sexual risk behaviors among HIV couples in Lusaka, Zambia. AIDS Care. 2013;25(9):1102-1108. 32. Bouhnik AD, Preau M, Schiltz MA, Lert F, Obadia Y, Spire B. Unprotected sex in regular partnerships among homosexual men living with HIV: a comparison between sero-nonconcordant and seroconcordant couples (ANRS-EN12-VESPA Study). AIDS. 2007;21(Suppl 1):S43-S48.
Downloaded from jia.sagepub.com by guest on November 14, 2015
