|Home | About | Journals | Submit | Contact Us | Français|
Vaginal microbicides could provide an important option for women in the prevention of HIV and other STIs. Researchers have examined women's preferences for specific product characteristics; however, much remains to be learned about women's willingness to use them with a variety of partners. This study examined high-risk women's experiences using simulated microbicides during a two-week trial. Ninety-six women completed the trial by using the product and reporting their experiences with different partner types. Analyses were conducted to examine differences between women who always used the product and those who used it less than all the time and whether risk and relationship attitudes were associated with simulated microbicide use during participants' most recent sexual encounters. All participants used the product and on an average of 79-94% of sexual encounters. The majority of the participants reported liking the simulated product. Women indicated that they would want to use microbicides in order to feel in control of their health, to avoid having to ask their partner, and because microbicides would be easier to use than condoms. Women who used the product 100% of the time had less relationship power; however, greater sexual assertiveness for STI prevention predicted simulated microbicide use during the most recent sexual encounter, and interventions may need to focus on promoting sexual assertiveness for disease prevention.
Despite an increase in HIV prevention efforts (Belcher, et al., 1998; Kamb, et al., 1998; Stanton, et al., 2000), including condom promotion interventions (Collins, Kohler, DiClemente, & Wang, 1999; Kalichman, Cherry, & Browne-Sperling, 1999; Stall, Paul, Barrett, Crosby, & Bein, 1999) and abstinence-based educational programs (Aten, Siegel, Enaharo, & Auinger, 2002), no notable decrease in yearly incidence of HIV infection due to sexual transmission has occurred (Centers for Disease Control and Prevention [CDC], 2004). Further, women and adolescent girls comprise the fastest growing segment of new HIV infections (CDC, 2004). Their risk might be magnified in situations in which they do not have the power to make independent decisions about disease or pregnancy prevention. Microbicides and other female-controlled HIV prevention methods are needed to address this continuing epidemic.
The term “microbicides” refers to a range of topical products in gel, cream, film or suppository form, being developed to prevent the transmission of HIV and other sexually transmitted infections (STIs). Microbicides, although not currently available,1 could prove to be an important tool in the prevention of HIV and STI transmission, particularly in the absence of an HIV vaccine. A number of studies have been conducted to examine the acceptability of various forms of microbicides (Darroch & Frost, 1999; Hardy, de Padua, Jimenez, & Zaneveld, 1998a-d; Hammett, et al., 2000; Severy, 2000; Soto, Espinoza, & Baum, 2000; Srirak, et al., 2000; Weeks, Mosack, Abbott, Novick Sylla, Valdes, & Prince, 2004). Specific product characteristics seem to be important. Brazilian women preferred a product in the form of a cream that could be inserted with an applicator, that had no taste, and that would both allow for spontaneity and last for a long period of time (Hardy, et al., 1998a-d). Other researchers examined user characteristics and found that U.S. women at highest risk of HIV--namely, women of color who are younger, single, and had a lower income-- were most interested in the use of microbicides (Darroch & Frost, 1999). Personal characteristics and context of use, such as whether the participants were married or cohabiting, were associated with different personal preferences, including product type, method of application, duration of effect, and willingness to pay more (Darroch & Frost, 1999). Other studies have found that certain characteristics (e.g., product's lubricating ability, concern about touching genitalia, smell and taste of product) were viewed either positively or negatively according to cultural preferences and contexts of use (Hammet et al., 2000).
In our previous work, we found that after controlling for the effects of demographic variables, microbicide acceptability scores were negatively related to having experienced either physical or sexual violence and to perceptions of HIV/AIDS risk, but positively related to STI locus of control (Weeks, et al., 2004). Additionally, having power within the sexual relationship was negatively correlated with microbicide acceptability scores. The current study examined high-risk women's experiences using simulated microbicides during a two-week trial. We explored the differences between women who used the simulated microbicide consistently and those who did not. We assessed the predictors of simulated microbicide use during participants' most recent sexual encounter, and we surveyed their interest in using actual microbicides with different partner types.
Women were recruited among those participating in a parent study entitled, Microbicides Readiness Survey, in which high-risk women from Hartford, Connecticut (i.e., heroin or cocaine users, those with injection drug-using sex partners, or those participating in commercial sex work) were surveyed regarding their interest in vaginal microbicides (Weeks, et al., 2004). The recruitment strategies and eligibility requirements for the parent study have been enumerated elsewhere (Weeks, et al., 2004). Based on participant responses to the Microbicides Readiness Survey, interviewers were able to determine whether participants were eligible for participation in the Simulated Microbicides Trial. Additional requirements for the Simulated Microbicides Trial were that the women not be pregnant or trying to become pregnant and that they reported no current STI or STI-like symptoms. Thus, upon completion of the Microbicides Readiness Survey, eligible participants were invited to participate in the Simulated Microbicides Trial. Potential participants were informed that they would be asked to use an over-the-counter vaginal moisturizer during vaginal sex for a two-week period and that they would need to document their sexual activity, condom and product use for that time period. They were then screened for their interest in and ability to follow the trial protocol. All but one woman who was invited to participate in the trial was interested in doing so. One hundred nine of the 134 women screened for participation in the trial were eligible. Women who were deemed ineligible or who were uninterested in participating were thanked and paid $25 for their participation in the Microbicides Readiness Survey.
All research activities were reviewed and approved by an Institutional Review Board for acceptable protections of human subjects; all participants provided informed consent before engaging in any research activities. Following written informed consent, trial participants were instructed to use a simulated microbicides product with male or female condoms (to protect participants from STIs and unwanted pregnancy) during vaginal intercourse during the subsequent two-week period. Participants were shown a tube of the simulated microbicide, Replens®, and told that this product was a water-based vaginal moisturizer that contained no microbicidal properties. Participants were given oral and written directions on the proper way to apply the product for the Simulated Microbicides Trial, strongly cautioned both verbally and in the written instructions that the product did not prevent pregnancy or HIV/STIs, and asked to apply one dose of the product before and to use a male or female condom during every act of vaginal intercourse over the next two-week period. Participants were provided with an adequate supply of condoms and doses of Replens. A follow-up appointment, in which participants completed the Simulated Microbicides Survey, was scheduled for two weeks after intake. Participants were paid an additional $40 for their participation in the trial.
Replens is an over-the-counter vaginal moisturizer primarily consisting of water and glycerin. It does not inhibit contraception, has no microbicidal attributes, and is compatible with condom use (Sato, Sugiuchi, & Nakamura, 1998; Trishman, Luciano, & Maier, 1992 in LDS Consumer Products, n.d.). We chose to use Replens as the simulated microbicide for this trial because its physical properties closely resemble several vaginal microbicides currently in advanced clinical trials, and it has been used as the placebo in some of these safety trials (personal communication, A. Profy and T. Moench, April, 2001). Replens is provided in single-dose, pre-filled plastic applicators, which can be opened by breaking off a small plastic tab. The lubricant is delivered by inserting the applicator directly into the vagina and compressing the plunger that contains the lubricant.
Data from the Microbicides Readiness Survey served as baseline data for the current Simulated Microbicides Trial study. The baseline survey measured socio-demographic characteristics, drug use and sexual practices, perceived HIV/AIDS risk, exposure to partner violence/abuse, STI and other health history, reproductive history and intentions, partner and sexual relationship factors, psychological factors, and willingness to participate in the Simulated Microbicides Trial and, if available, a clinical trial of actual microbicide products. Data gathered from the Simulated Microbicides Trial Survey included sexual behavior reports (including vaginal sexual encounters, Replens use, and condom use) and interest in actual vaginal microbicide use. Measures relevant to the present study are outlined below.
Microbicide acceptability was measured at baseline through a modified version of the Bridgeport Survey of Acceptability of Vaginal Microbicides that was developed to assess acceptability among similar high-risk populations in Bridgeport, Connecticut (Hammet, et al., 2000; Hardy, et al., 1998a). The index included 21 items on the relative acceptability of certain characteristics, using a Likert scale from 1-4 (1=very unacceptable; 4=very acceptable). Participants responded to questions that were neutral (“What if microbicides were a gel?”), negative (“What if a small amount leaked out after sex?”), and positive (“What if the microbicide had a pleasant taste?”). The microbicide acceptability index derived from this set of items demonstrated high internal consistency, as measured by Cronbach's alpha (α=.90). Interest in microbicide clinical trial participation was measured by a dichotomous one-item question.
We used the 23-item Sexual Relationship Power Scale in the baseline survey (Pulerwitz, Gortmaker, & DeJong, 2000). This scale consists of two subscales: Relationship Control and Decision-Making Dominance. For the Relationship Control subscale, participants responded on a Likert scale from 0 (Strongly Agree) to 3 (Strongly Disagree). A sample item from this subscale is “Most of the time, we do what my partner wants to do.” For the Decision-Making Dominance subscale, response choices were 0 (your partner), 1 (both you and your partner), and 2 (you). A sample question from this subscale is, “Who usually has more say about how often you see one another?” The subscales were standardized to compute the overall scale score. Higher scores indicate greater relational power. For the current study, α=.88.
The Sexual Assertiveness Scale consists of three subscales: Sexual Assertiveness for Initiation of Sex, Sexual Assertiveness for Refusal, and Sexual Assertiveness for Pregnancy/Sexually Transmitted Disease (STD) Prevention (Morokoff et al., 1997). We used a modified version of the Sexual Assertiveness for Pregnancy/STD Prevention subscale in the baseline survey. An example of the items in this subscale is “I refuse to have sex if my partner refuses to use a condom or latex barrier.” The five-item subscale used Likert scoring from 0-4 (0=Never; 4=Always). Higher scores reflected more sexual assertiveness. For the current study, α = 70.
The items used were taken from the Risk Behavior Assessment (RBA) developed for and used in the National Institute on Drug Abuse (NIDA) Cooperative Agreement for AIDS Community Based Outreach/Intervention Research Program (Joe, Menon, Copher, & Simpson, 1990; Joe & Simpson, 1991; NIDA Cooperative Agreement for AIDS Community Based Outreach/Intervention Research Program, 1993). Only those six questions related to HIV risk prevention attitudes were used in the baseline survey. An example is, “My present behavior will protect me against HIV or AIDS.” Items were scored on a 4-point Likert scale (0=Strongly Agree; 3=Strongly Disagree) such that higher numbers indicated a greater perceived risk. For the current study, α = .63.
The Simulated Microbicide Trial survey assessed participants' sexual activity during the entire two-week trial, including the total number of sexual partners, the total number of vaginal sexual encounters, the total number of times the simulated microbicides were used, and the total number of condoms used for these encounters. Additionally, the number of sexual encounters and number of times in which simulated microbicides and condoms were used for encounters with each of the three types of partners (primary, casual, and paying) were assessed. Primary partners were defined as those individuals who participants identified as being their “main, steady, or primary partner.” Accordingly, participants could identify a maximum of one primary partner. Paying partners were defined as those partners with whom participants had sex “only in exchange for drugs or money.” Casual partners were defined as those partners with whom participants had sex who were not considered primary or paying partners. Lastly, sexual activity, product, and condom use information was gathered about participants' most recent sexual encounter.
Trial participants were asked to answer questions about their experiences using Replens during sexual encounters with each of the three partner types, and for their most recent sexual encounter. They were provided a list of possible reactions and were asked the question, “What experiences did you have when you used Replens during sex with your primary (casual, paying) partner(s)?” Participants could endorse any of a number of applicable experiences. They were also asked about whether their partners knew about their Replens use, their reasons for not telling partners (when applicable), and their partners' attitudes toward the simulated microbicides. Finally, they were asked about their intentions to use actual microbicides.
Frequencies and descriptive statistics were computed using SPSS (Version 12.0) for the number of sexual encounters with different types of partners, condom use, Replens use, participant experiences with Replens use with different partner types, partner knowledge of product use, willingness to participate in microbicide clinical trials, interest in use of microbicides with different types of partners, and specific characteristics of participants' most recent sexual encounter. Independent samples t-tests were performed to examine differences between those who used the simulated microbicides 100% of the time and those who used it less than 100% of the time in relation to attitudinal variables (including microbicide acceptability, relationship power, sexual assertiveness, and perceived HIV/AIDS risk). Further, a multiple logistic regression analysis was performed using these same attitudinal variables to examine their association with Replens use during the most recent sexual encounter. Model fit was assessed by using the Wald χ2 test of difference between the full and null models; a .05 criterion of statistical significance was employed.
Participants were predominantly of African American (52.6%), Hispanic (29.5%), and European American (15.8%) descent. They ranged in age from 18 to 53 years (Mean (M)=36.0; Standard Deviation (SD)=8.8). Most were single (51.6%), divorced (24.2%) or married/partnered (21.0%). Most participants had given birth to at least one child (85.3%), and 42.1% were raising children at the time of the study. The majority of the sample had either some high school education (38.9%) or had earned a high school diploma or passed the graduate equivalency exam (40.0%). One-third of the sample was homeless, and two-thirds were unemployed or laid off from work. Over one-third (35.8%) of the sample smoked crack cocaine, 17.9% snorted cocaine or heroin, and 15.8% injected drugs. Nearly sixty percent (58.9%) of the women reported using an illicit drug in the last 30 days. Nearly a quarter of the sample (24.2%) reported having been paid by partners for sex in the 30 days prior to the baseline interview. Ninety-six of those enrolled (88.1%) completed the study.
During the two-week trial period, all participants reported using the product at least once (Range=1-30 doses; M=4.97, SD=3.84). Eighty participants reported having a primary partner, 11 participants reported having between 1-5 casual partners, and 14 participants reported having between 2-45 paying partners. Table 1 provides descriptive information about participants' simulated microbicides and condom use with different partners. All participants with primary partners and all those with casual partners reported using simulated microbicides at least once with these partner types. All but one individual with paying partners reported being able to use them at least once with individuals of that partner type. An independent samples t test revealed that those who reported using the product 100% of the time (M=2.77, SD=.37) differed from those who had less than perfect compliance (M=2.80, SD=.36), t (94)=-2.08, p <.05 with regard to relationship power. Specifically, compliant participants had less relationship power.
The most frequently endorsed responses to questions about their experiences using the simulated microbicide across partner types are shown in Table 2. Eighty-six percent of respondents indicated that their primary partners knew that they were using the product, 46% indicated that their casual partners knew, and only 23% reported that their paying partners knew. Those whose partners did not know were asked for the reasons they had not told them. Twenty-seven percent of those with primary partners, 60% of those with casual partners, and 100% of those with paying partners agreed with the item, “I didn't think it was any of his business.”
Two-thirds of participants reported having used a male condom during their most recent sexual encounter; however, 40% (n=38) failed to use Replens. Seventy-nine percent indicated that their partner knew about their Replens use. Of these, 52% indicated that the partners did not care one way or another about its use, 35% indicated that the partners liked the product, and 13% indicated that the partners disliked it.
Table 3 shows the results of the multiple logistic regression analysis to examine factors related to simulated microbicide use during the most recent sexual encounter. Microbicide acceptability, relationship power, sexual assertiveness, and AIDS risk scale scores were included as independent variables to examine the degree to which they predicted use during participants' most recent sexual encounter. The results indicate that the full model is significantly different from the null model, χ2 (4, N=91)=10.35, p=.04. The model correctly classifies 78.6% of those who used Replens and 31.4% of those who did not, for an overall success rate of 60.4%. The Nagelkerke R2, which is an estimate of the R2 for logistic regression models, was .15. Sexual assertiveness for pregnancy/STD prevention was significantly related to use (p=.03). For every increase of one point on the sexual assertiveness scale, the odds of having used the product increased by 1.6 times.
All but one participant (n=95) indicated that she would like to participate in a microbicides trial, if one were to become available. Participants were also asked whether they intended to use actual microbicides, when they become available, with different partner types. Nearly all participants with the particular partner types indicated that they intended to use such a product with primary partners (98.8%), casual partners (100%), and paying partners (100%). Further, when asked how often they would use the product with different partners, they reported that they would always use the product with primary partners (73.8%), with casual partners (97.4%), and with paying partners (96.2%). When asked the reasons for which they would want to use microbicides with different partner types, participants provided responses indicating the importance of having control over their own disease prevention decisions, that they would prefer microbicides to condoms or that microbicides would be a convenient option. The most commonly endorsed reasons by partner type are enumerated in Table 4.
Participants who indicated that they would not always use microbicides were asked about the reasons for which they would never or not always use such a product. They could endorse as many reasons as were applicable and they could specify additional reasons. The most common reasons for not wanting to use such a product included concerns about it being too messy or drippy (n=8), forgetting to use it (n=7), the inconvenience of it (n=5), and trusting one's partner (n=5).
Product acceptability will play a key role in the success of introducing any method of protection to the market. One product that has been introduced more recently is the female condom. Factors that detracted from acceptability included insertion difficulties (Beksinska, Rees, McIntyre, & Wilkinson, 2001; Cecil, Perry, Seal, & Pinkerton, 1998; Neilands & Choi, 2002; UNAIDS/WHO, 2000), cost (Pettifor, Beksinska, Rees, Mqoqi, & Dickson-Tetteh, 2001), appearance (Beksinska, et al., 2001), partner objection (Beksinska, et al., 2001; Ford & Mathie, 1993; Pool, et al., 2000; Kaler, 2001; Sapire, 1995), and provider bias (Welsh, Feldblum, Kuyoh, Mwarogo, & Kungu, 2001). Factors that enhanced acceptability included enhanced partner communication (Rivers, et al., 1998), perception of HIV risk (Nwaneka & Inyang, 2004; Sakondhavat, et al., 2001), and receiving instruction and opportunity to practice use (Artz, Demand, Pulley, Posner, & Macaluso, 2002; UNAIDS/WHO, 2000). While numerous studies have addressed female condom acceptability, there is little published data on male condom acceptability. However, in a review of male methods of contraception, Ringheim (1996) found both acceptability and use of male condoms to be low, in both developing and developed nations.
Previous research on microbicide acceptability has indicated that women have particular physical preferences for the products (Darroch & Frost, 1999; Hardy, et al., 1998a-d; Hammett, et al., 2000; Severy, 2000; Soto, et al., 2000; Srirak, et al., 2000; Weeks, et al., 2004). However, since microbicides are not yet available, researchers have been unable to assess women's likelihood of using such a product with different partner types and under different circumstances. This study was an attempt to assess not only the degree to which women would be able to comply with instructions to use simulated vaginal microbicides during intercourse for a two-week trial period, but also to examine their experiences using them. The results indicated that all participants used the product at least once during the trial period with primary, casual, and paying partners. Further, they tended to like the product and were willing to participate in clinical trials, if they became available. When asked about the possibility of using microbicides with different partner types, the results were encouraging as at least three-quarters indicated an interest in always using the product, particularly those with casual and paying partners. Of course, while we can imitate the physical properties of simulated microbicides, we cannot simulate the meaning microbicide use would have in particular relationships. Still, given the strong interest in using this product with different partner types, we can conclude that these women were interested in having another tool available in the prevention of STIs and HIV. Further, their reasons for wanting to use such a product seemed to be a function primarily of having control over their own sexual decision-making.
Not surprisingly, relationship dynamics played a role in the experience of using the product. Participants were likely to tell primary partners, who did not seem to mind the product. Low expected cell counts and non-independence of cells for those with casual and paying partners precluded the use of analyses to assess independence of partner type and partner knowledge of Replens use. Nonetheless, a clear trend developed in that individuals who were classified as belonging to less intimate partner type categories (i.e., casual, paying) appeared to be less likely to have known about the use of simulated microbicides, and the study participants were more likely to believe that their use of the simulated product was “none of his business”.
Further, multiple logistic regression results indicated that sexual assertiveness for pregnancy/STD prevention was associated with use of the simulated microbicide during participants' most recent sexual encounter. This result was similar to the results from an examination of factors associated with condom use during the previous sexual encounter (results not shown). The same mechanism may be influencing women's decision to introduce either product into the sexual encounter; thus, we need to be cautious in the promotion of microbicides as a “cure-all” for situations in which women are unlikely to ask their partners to use condoms. On the other hand, these results suggest that sexual assertiveness at baseline is a key determinant of (simulated and actual) protective behaviors. Sexual assertiveness reflects a behavioral orientation, and as such, may be amenable to modification. Interestingly, microbicide acceptability and perceived AIDS risk were not associated with use, suggesting that personal assessments of product characteristics or risk are less important than the ability to be assertive in sexual decision-making. Previous research indicates that women are much more likely to insist upon condom use with non-primary partners (Macaluso, Demand, Artz, & Hook III, 2000), and ultimately, interventions may need to focus on promoting sexual assertiveness for disease prevention. It is unclear whether partner buy-in or particular relationship dynamics influence sexual assertiveness.
Similar to our earlier findings (Weeks, et al., 2004), those who used the simulated product for every sexual encounter reported less relationship power at intake. The Relationship Power scale reflects a more global assessment of relationship power, whereas the Sexual Assertiveness scale is specific to STI prevention behaviors, which might provide insight into the apparent inconsistency in our findings. Application of the simulated microbicide might have been very similar in nature to other behaviors women typically engage in to reduce their risk of STI transmission. At the same time, simulated microbicide use might have also resonated with women who want but do not have enough power in their relationships to make self-protective decisions. Further examination is necessary to more fully understand the role that relationship dynamics play in simulated (and eventually, actual) microbicide use.
We are limited in the conclusions and comparisons we can draw, given the constraints of the data for this pilot study. This sample was drawn from a larger parent study. As such, there may have been selection biases based on the characteristics of those individuals who decided to participate in that study. Thus, the results might not be generalizeable to women living in different contexts, especially for non-high risk women. Additionally, although all but one woman who was approached was interested in participating in the trial, thirteen eligible women did not return for the follow-up interview. We might assume that these women were similar to those who did complete the trial; however, due to poor statistical power, i.e., our ability to detect statistical differences between the two subsamples, we were unable to draw any conclusions about their propinquity. As such, the results of this study may not be generalizeable to women who might not be willing to participate in such a study, particularly women who might experience repercussions if their partners' became aware of product usage. Because individual participants could have more than one partner type, we could not make independent comparisons between participants. Further, because we asked about categories of partners, we were unable to analyze the data at the individual partner level. Finally, and most importantly, we relied on the participants' self-reports regarding their sexual behavior. While self-report was the most feasible way of collecting simulated microbicide usage data, it may not have yielded entirely reliable data. It is important to note that this was a pilot study in a relatively new area of research and we were able to make certain conclusions that will guide further hypothesis testing.
It would be illuminating to examine partner perceptions directly. Although we asked about partners' perceptions of product use, we asked this entirely from the perspective of the study's participants. Future researchers may want to include partners' perspectives. Such research would have to be carefully constructed to provide safeguards against a strong self-selection bias of partner involvement in a study of this nature. However, researchers could potentially recruit male participants who are instructed to use the product with their female partners and report back on their experiences. Although an improvement in perceived sexual assertiveness may be an important component of women initiating use of this product with their male partners, garnering partner support for this method as well as for the use of condoms continues to be indicated as an important point of intervention.
This research was supported by National Institute on Mental Health grant R01 MH063631, NIMH center grant P30-MH52776, and NRSA postdoctoral training grant T32-MH19985. Neither the researchers nor the research project has any relationship with or bias toward LDS Consumer Products, which produces Replens® vaginal moisturizer. Further, the researchers do not have a financial stake in the results of the study. An earlier version of this article was presented at the Annual Meeting of the American Public Health Association, November 2004. The authors wish to thank Barbara Valdes, Mary Prince, Lance Weinhardt, Steve Pinkerton, and Tim McAuliffe for their contributions to the project and comments on earlier versions of this manuscript.
1As of this writing, there were at least four microbicides entering or completing Stage III clinical trials (Alliance for Microbicide Development, n.d.).
Katie E. Mosack, Center for AIDS Intervention Research, 2071 N. Summit Avenue, Milwaukee, WI 53202, 414-456-7740.
Margaret R. Weeks, Institute for Community Research, 2 Hartford Square West, Suite 100, Hartford, CT 06106, 860-278-2044.
Laurie Novick Sylla, Connecticut AIDS Education and Training Center, Yale University School of Nursing, New Haven, CT 06536, 203-737-2361.
Maryann Abbott, Institute for Community Research, 2 Hartford Square West, Suite 100, Hartford, CT 06106, 860-278-2044.