Designing Dual Compartment HIV Prevention Products: Women's Sensory Perceptions and Experiences of Suppositories for Rectal and Vaginal Use

Abstract

Dual compartment suppositories are being developed to prevent HIV and other sexually transmitted infections. Such products, for use in the rectum, the vagina, or both, could have a significant public health impact by decreasing global incidence of these diseases. In this study, 16 women each used two rheologically distinct suppositories in their vagina and rectum. User Sensory Perception and Experience (USPE) scales assessed sensory experiences during sexual activity to understand whether, and how, women perceive formulation properties in the vagina and rectum. Qualitative data from individual in-depth interviews captured women's descriptions and comparisons of the experiences. Significant differences and large Cohen's d effect sizes between vaginal and rectal experiences of suppository-A were found for three scales: Application (APP): Product Awareness, SEX: Initial Penetration; and SEX: Effortful. Qualitative data provided user experience details that credibly align with these score differences. Near significant differences and large effect sizes were found for two additional scales: SEX: Perceived Wetness with suppository-A and SEX: Messiness with suppository-B. In addition, other scale scores showed medium-to-large effect sizes that correspond to hypothesized sensations associated with biophysical properties of the suppositories. Statistical significance combined with large effect sizes and qualitative data accurately represent the hypothesized perceptibility of suppository properties and identifies performance characteristics relevant to acceptability and adherence; together these data provide discernment of factors that can guide the development of dual compartment products.

The Clinical Trial Registration number: NCT02744261.

Introduction

HIV-related deaths have declined worldwide and in women overall,¹ and HIV incidence has declined in some populations. HIV prevalence, however, remains high among key populations, including adult females in many regions of the world.² Women remain disproportionately affected by specific sociobehavioral and biological factors that increase this inequality.^3

–6 Less attention, however, has been paid to the fact that women's HIV risk is further magnified because some women engage in both vaginal and anal intercourse^7,8 and the latter is a particularly high-risk behavior for HIV transmission.^9,10 HIV prevention strategies that protect women who engage in both vaginal intercourse (VI) and receptive anal intercourse (RAI) will further protect women's sexual health and address the global HIV pandemic.

Rectal microbicides formulated as semi-solid topical products have been identified as a potentially important element of HIV prevention¹¹ strategies, in part because rectal lubrication is often used during anal intercourse and a semi-solid could potentially provide both lubrication and protection against HIV/sexually transmitted infections (STI). Much of the rectal microbicide research has focused on men who have sex with men (MSM^12

–16); however, important formative work has also sought to include women's anal sex practices and product preferences in rectal product development.¹⁷

Suppositories, well-established drug delivery systems used both vaginally¹⁸ and rectally,^19,20 have been considered a potential HIV prevention formulation throughout the decades-long microbicide development effort.^21

–24 Here, too, work has focused on the needs and preferences of MSM, although some research has included both men and women,²⁵ or focused exclusively on vaginal suppositories.^26,27 Efforts to develop a dual compartment microbicide for women have principally focused on topical gels that could be applied to either, or both, the rectum and vagina.^28,29 Given the established efficacy of suppositories for vaginal and rectal drug delivery, understanding the impact of rheological parameters on perceptibility and acceptability of dual compartment suppositories is a critical step in pipeline progress.

Background to the study

This study (DRUM-S) sought to understand users' sensory perceptions and experiences (USPEs) of two different suppository formulations used rectally by MSM and women who engage in receptive anal sex and, vaginally, by women; a separate element of the study also explored a gel formulation used rectally by both MSM and women (data not presented here). It is an extension of an ongoing line of research by this team.^30

–34 The current analysis focuses on women's experiences using two suppositories, each rectally and vaginally, to understand which product characteristics and USPEs may be most relevant for the formulation of dual compartment prevention products for women.

Prior work by this team of investigators includes Project LINK, which developed and validated USPE measures for topical vaginal gel formulations,³⁴ and Project MIST,^31,32 which expanded the formulation parameter space to include vaginal film, as well as expanding the perceptibility scale validation process to include male sexual partners of women using such formulations during vaginal sex. Latent class modeling conducted on the Project LINK data indicated that four of the sexual activity scales (SEX) were particularly relevant to understanding willingness-to-use (SEX: Initial Penetration, SEX: Perceived Wetness, SEX: Messiness, and SEX: Leakage), establishing a clear relationship between perceptibility and willingness to use a vaginal gel³⁰ for HIV prevention. Analyses from Project MIST allowed the identification of six additional scales associated with perceptibility within an expanded parameter space for vaginal topical formulations³¹ (SEX: Pre-Coital Leakage, SEX: Naturalness, SEX: Lubricity, SEX: Effortful, SEX: Pleasure, and SEX: Noticeable) and suggested that the USPE scales validated in Project LINK generalized well to varying volumes of topical formulations. Furthermore, the addition of USPE assessment among male sexual partners added to the credibility and utility of the scales.

A prior study, Project DRUM, adapted the existing vaginal USPE scales for use with three different 10 mL rectal gels, thus extending psychometric evaluation of the USPE scales from vaginal-only evaluations to both rectal and vaginal formulation evaluations. Psychometric properties of the adapted USPE scales were very good, providing confidence in the USPE scale set. That study was the first to use perceptibility science to evaluate topical rectal prevention formulations, moving the science closer to facilitating development of a dual compartment HIV prevention product. An earlier study compared two volumes (3.5 vs. 10 mL) of a gel used by MSMs to determine luminal distribution of gel in the rectosigmoid and to evaluate USPEs during at-home use. The USPE scales showed conceptually consistent findings regarding volume comparisons in that study.³⁵

The current study, DRUM-S, built directly on these previous findings. DRUM-S is the first study in which USPE scales were used in the development of suppositories for HIV prevention, as well as to examine both vaginal and rectal delivery of suppositories. These efforts are an important step forward in dual compartment product development.

Materials

Participants used two over-the-counter suppositories approved for both rectal and vaginal use. Both suppositories were obtained from their distributors; two doses were provided to study participants in their original per-dose packaging. Suppositories were identified with neutral labels that do not reference any product characteristics; participants and staff referred to them as “the Yellow product” (suppository-A) and “the Green product” (suppository-B). Table 1 provides constituent lists and relevant rheological properties of each product.

Table 1.

Rheological Properties and Constituents of Study Suppositories

Suppository label	Shape	Appearance	Length (cm)	Unit mass (g)	Cross-sectional area (cm²)	Firmness G′ (kPa)^a	Dissolve time (min)^b	Dissolved osmolality (mOsm/kg)^c	pH
A	Bullet	Off-white, waxy	2.5	2.070	0.689 ± 0.039	350.29 ± 57.23	15	0	7.77
B	Bullet with slightly bulbous midsection	Light-medium brown	2.9	3.154	0.828 ± 0.060	594.52 ± 40.21	8	1,045 ± 45	7.47

The firmness (G′) was measured on the texture analyzer instrument that measures the maximum force needed to crush a cross-sectional area of the suppository.

“Dissolve” time was measured through visual inspection performed in 50 mL of DI water heated to 37°C.

Osmolality of the suppository was performed with an osmometer in 50 mL of DI water with a completely dissolved/melted suppository sample.

Suppository-A Constituents: hydrogenated coconut oil, hydrogenated palm oil, natural source vitamin E; did not require refrigeration.

Suppository-B Constituents: plant and fruit butters, vegetable acids, medium chain triglycerides, lecithin, vegetable starches, cellulose, oligosaccharides derivatives and polysaccharide derivatives; refrigeration optional.

Suppositories historically have fallen into one of two major types: lipophilic based or hydrophilic based. Lipophilic fat-based suppositories melt at body temperature to release drug to the body. They readily solubilize typical insoluble small-molecule drugs and require no localized fluids to spread and release drug (i.e., temperature-based melting diffusion). Typically, such suppositories are ideal for the rectum where there is little available fluid in the lower large intestinal tract. The hydrophilic water-based suppositories are unaffected by body temperature and require water to dissolve the suppository and release the drug (i.e., water-based dissolving diffusion). In contrast to the fat-based suppositories, hydrophilic suppositories can more easily support drug delivery of soluble drug compounds and use the body fluids to dissolve the suppository to transport the drug into the body. Such water-based suppositories are appropriate for vaginal application where there is more local fluid.

A suppository will first dissolve in the fluid or melt on the mucous layer depending on whether it is hydrophilic or lipophilic. Owing to osmotic effects of a dissolving hydrophilic suppository, water is drawn to the rectum or vagina, and as the suppository dissolves, drugs in the suppository will diffuse out toward the mucosal epithelial surfaces. For lipophilic melting suppositories, there is no need for fluid to be present to soften and spread the suppository. As the suppository liquifies under heat, the same drug transport observed in dissolving suppositories will occur in melting suppositories. These properties and performance features of lipophilic and hydrophilic suppositories are at the crux of the challenge in developing dual compartment suppositories for HIV prevention.

These mechanisms of delivery can be physically distinguished by their rheological properties. In its unaltered form, the firmness (G′) of each of the suppositories distinguishes each from the other. Lipophilic-based suppositories have a relatively lower firmness compared with hydrophilic-based suppositories. This is because of the temperature-dependent fatty acids that comprise lipophilic suppositories. Such ingredients are softer at room temperature than water-dissolvable ingredients, which can be press molded into hard dosage forms not susceptible to ambient temperatures. This is demonstrated in comparing suppository-A (lipophilic), which is 40% less firm, with suppository-B (hydrophilic). Even before insertion, the rheological differences in firmness can be distinguished in the hands of the user. The lipophilic suppository will feel softer and start melting if continued to be handled. The hydrophilic suppository will feel harder and maintain that firmness.

Once inserted, the rheological differences between the two suppository formulations should continue to be perceived. Suppository-A (lipophilic) will melt once inserted into the body. Spreading and perception of the suppository in the vagina or rectum occurs as the suppository base moves from the insertion point outwards to coat the tissue. Such coverage is similar to a cream that acts as an initial barrier to the tissue. This imbues the surface of the compartment with any rheological properties of the melted suppository (lubricity, stickiness, etc.) during and after penetration. Suppository-B (hydrophilic) will dissolve once inserted into the body. It draws water from the tissue and spreading and perception of the suppository in the vagina or rectum occurs as the suppository shrinks at the point of insertion and the existing surrounding fluids transport to the tissue. Thus, a user's potential sensory perception of the suppository is that of the dosage form disappearing. Any rheological properties of the excipients in the dissolved suppository will be diluted and diffused away.

In the current study, which focuses on measuring and understanding user sensory perceptions and experiences within the parameter space of potential suppository formulations, our rationale was to cover the two primary categories of suppositories, lipophilic and hydrophilic, to evaluate the utility of the USPE scales and qualitatively describe users' sensory experiences within the range of potential suppository formulation properties and performance.

Methods

Project DRUM-S was a single-site pre-phase 1 USPE measure development study, utilizing a formative mixed-method design. The primary objective was to evaluate the sensory experience of suppository use in the context of RAI among men and women, and, for women, in the context of VI, enabling evaluation of formulations for dual compartment use. This article compares and contrasts women's dual compartment suppository formulation experiences using both quantitative and qualitative data. Data related to associations with participants' willingness-to-use similar formulations for HIV prevention are presented elsewhere.

All study procedures were reviewed and approved by the local Institutional Review Board.

Participants

Eligible participants were healthy sexually active women (18–45 years), who reported RAI and VI each at least twice in the previous 6 months and were willing to use each of three study products in conjunction with RAI and VI on at least one occasion (resulting in a minimum of three RAI and three VI events). Female volunteers were required to report consistent use of an effective birth control method and were excluded if they were breastfeeding, pregnant, or planning to get pregnant during the study timeline. If volunteers were found to be positive for a curable STI at screening, they could rescreen following treatment and test of cure. All participants were compensated for their time and effort to complete study activities.

Recruitment and screening

Volunteers were recruited through advertisements (e.g., print, social media, and internet), community-based organizations, and word-of-mouth. Volunteers contacted research staff who described study procedures, including risks and benefits; following verbal consent, participants responded to basic eligibility questions. Preliminarily eligible participants were scheduled for a clinical screening visit (Visit 1) that included written consent to be screened, and STI (Neisseria gonorrhoeae, Chlamydia trachomatis, and syphilis), HIV (with pre- and post-test counseling), and pregnancy testing. Limited visual rectal and pelvic examinations were also conducted. Participants completed a baseline survey during this clinical visit (e.g., marital status, financial situation, educational achievement, HIV/STI risk perception, and needs in a sexual/reproductive health product). Once participants were deemed fully eligible, they were scheduled for their first product evaluation (study enrollment) visit.

Formulation evaluation

At visit 2 (enrollment visit), all participants completed written consent to participate in the product evaluation study and completed a Sexual and Reproductive Health (SRH) History Survey including sexual behavior and SRH product use (e.g., douche; lubricant). Female participants were screened for pregnancy at the start of each visit. Participants were then provided with two doses of their assigned study formulation. The order in which formulations were evaluated by each participant was randomized. Study staff reviewed written instructions for correct insertion of each formulation. Condom use and lubricant use was allowed. Encouragement (i.e., not requirement) of condom use for all sex acts was a stipulation of the funder: lubricant use was not prohibited to approximate, and be able to inquire about, real-world use by participants. Staff provided a condom use demonstration at the first formulation visit and as necessary thereafter. At visits 3 and 4, participants returned to complete an interview for the first and second formulations evaluated, respectively, and receive their next formulation. At visit 5, participants completed an interview for the third formulation evaluated, as well as the final study survey.

Quantitative USPE data

Between visits 2 and 5, each participant completed an automated daily phone survey (<3 min) capturing data related to product use/nonuse and sexual behaviors and monitoring for adverse events or participant concerns. Responses endorsing product use associated with either RAI and/or VI prompted completion of a web-based user experience survey. Web-based surveys included USPE items, conventional acceptability items, and behavioral and psychosocial measures capturing participant experiences during RAI/VI (e.g., partner type; sexual behavior; product use self-efficacy; willingness-to-use product). USPE items are statements, written in a factual tone, to which participants respond with varying levels of agreement on a 5-point Likert scale. Each item is designed to capture a specific construct, concept, or experience hypothesized to be related to biophysical and other rheological parameters of suppository and other semi-solid formulations. The USPE scales fall into three broad categories: Application (APP: i.e., product insertion), Ambulation (AMB: i.e., walking around, daily activity), and Sexual Activity (SEX). Women completed separate surveys for VI and RAI, whether or not they occurred in the same sexual episode.

Qualitative USPE data

Once all formulations had been used in each compartment and quantitatively evaluated, the participant completed a single qualitative in-depth interview about suppository perceptibility, acceptability, and willingness to use. Interviews were guided by a qualitative research agenda; questions focused on users' sensory perceptions, as elicited by suppository properties. Participants were asked to describe and discuss: (1) steps they took to prepare for RAI or VI (e.g., douching/enema, lubricant use); (2) the insertion process, including ease or difficulty; (3) if, when, and under what conditions each product leaked out of the rectum/vagina; (4) awareness of the suppository during RAI/VI, including perceptions of presence/absence, wetness/dryness, perceived penetration effort, and so on; (5) product behavior as it impacted the experience of sex, including lubricating or desiccating properties, perceptions of interference, sexual pleasure and/or comfort, and so on; (6) whether the product might be used covertly; and (7) volume considerations in the context of RAI/VI.

Interviews were audio recorded and transcribed. Applied thematic analysis³⁶ was conducted by a team of analysts who developed a codebook based on questions in the interview agenda as well as emergent topics raised by participants. Each transcript was individually coded by two or more coders who then convened to compare and agree upon final coding and entered all data into the NVivo qualitative data project. Throughout this process, the coding team met regularly to discuss the codes, clarify their definitions, and keep track of their use. An audit trail of the codebook and these meetings was maintained. Codes were used to identify key USPEs, to track the product being discussed (e.g., suppository-A and suppository-B) and the type of intercourse referenced in the discussion (RAI/VI).

After all transcripts were coded, analysts identified relevant codes and used the coding query system to sort comments by product and type of intercourse. Summaries were written describing the aggregated participant experiences, keeping track of participants who reported similar experiences, as well as those noting unique experiences, and capturing relevant descriptive quotes. Five analysts participated in this process; over 250 individual summaries were written.

Mixed methods analysis

The mixed methods analysis used an integrated convergent design³⁷ in which the quantitative and qualitative data were collected and analyzed separately but concurrently. All quantitative analyses were conducted using IBM SPSS Statistics 20.0. Descriptive statistics were calculated for demographics and behavioral variables. Exploratory dimensional analyses used principal component analysis to re-examine the internal structure of the eight original USPE scales³⁴ as well as six novel scales identified in a separate study.³¹ Individual scale scores were calculated by summing the individual items on a scale and dividing by the number of items in each scale [Averaged Scale Item Score (ASIS)]. This provided a common scale metric to aid interpretation. Effect size differences were calculated for the 20 paired t-test comparisons using an adjusted version of Cohen's d statistic.³⁸

The current analysis focuses on pairwise comparisons of ASIS of sensory experiences during VI and RAI that had both large effect sizes and were statistically significant to explore participant experiences of each suppository during rectal and vaginal use. This focus both minimized potential errors because of overinterpretation (given sample size) and directed the qualitative inquiry to areas where there were strong signals of difference. The resulting analysis allows the qualitative data to provide descriptive participant experiences that add to the interpretation of quantitative results.

Qualitative data summaries (i.e., memos) brought together data from codes most relevant to the constructs each scale captured. For example, memos associated with APP: Product Awareness included data from codes capturing the insertion process and the sensations the participant was aware of during insertion (i.e., being physically aware that the suppository was present). Memos associated with SEX: Initial Penetration captured smoothness and lubricity as the penis passed the introitus/anal sphincter, including ease of penetration. Memos associated with SEX: Effortful included data from codes capturing sensations of lubrication upon penetration, sensations of effort to complete penetration, and difficulty with strokes (i.e., penis moving/thrusting inward/outward during sex) during sex. We did this for both suppositories in both compartments, to compare, contrast, and therefore better understand what formulation properties elicit sensations that may contribute to differences in ASIS.

Results

Forty-three female volunteers were prescreened by phone for initial eligibility; 21 were deemed eligible and 18 enrolled in the clinical screening process (visit 1). Of these, two were withdrawn: one screened positive for an STI at visit 1 but did not return following treatment [i.e., she did not enroll in product evaluation (visits 2–5)] and one was unable to follow study guidelines.

Sixteen women completed the study; demographics and SRH data are given in Table 2. As noted, condom and personal lubricant use was allowed. Ten women used the study suppositories in separate RAI and VI sexual episodes; the others reported having both RAI and VI in the same sexual episode for each suppository evaluated. Half of the women (8/16) used condoms with each suppository for both RAI and VI. All 16 women reported lubricant use during VI, whereas only half the women (8/16) reported lubricant use for RAI.

Table 2.

Participant Demographics, Sexual and Reproductive Health Needs and Sexual and Reproductive History

Participant demographics (N = 16)
Age: mean (SD; range)	28.88 (6.65; 20–42)
Hispanic/Latina, n (%)	4 (25)
Race, n (%)
Black	1 (6)
White	10 (63)
American Indian/Alaska Native	1 (6)
More than one race^a	1 (6)
Refused	3 (19)
Marital status, n (%)
Never married	9 (56)
Married	2 (13)
Separated	2 (13)
Divorced	1 (6)
Other^b	2 (13)
Annual income, n (%)
<$15,000	6 (38)
$15,000–$36,000	9 (56)
>$36,000	1 (6)
Sexual and Reproductive Health Needs, next 6 months
Importance of reducing HIV risk (next 6 months): n (%)
Not at all important	3 (19)
A little important	2 (13)
Somewhat important	1 (6)
Very important	2 (13)
Extremely important	8 (50)
Importance of reducing other STI risk (next 6 months): n (%)
Not at all important	3 (19)
A little important	0
Somewhat important	3 (19)
Very important	2 (13)
Extremely important	8 (50)
	Rectal/rectum	Vaginal/vagina
Importance of lubrication for [type of sex] in next 6 months? n (%)
Not at all important	0	1 (6)
A little important	1 (6)	3 (19)
Somewhat important	5 (31)	5 (31)
Very important	6 (38)	4 (25)
Extremely important	4 (25)	3 (19)
Sexual and Reproductive History, including prior product use

	Rectal/rectum	Vaginal/vagina
Total number of sex episodes (by type) in past 6 months
Mean (range)	10.38 (2–32)	89.75 (30–218)
Douche/enema to prepare for [type of sex] in past 6 months? n (%)
None of the time	12 (75)	13 (81)
Some of the time	2 (13)	2 (13)
Half of the time	2 (13)	1 (6)
Most of the time	0	0
Every time	0	0
Use of male condom for [type of sex] in past 6 months? n (%)
None of the time	9 (56)	10 (63)
Some of the time	3 (19)	4 (25)
Half of the time	1(6)	0
Most of the time	1 (6)	1 (6)
Every time	2 (13)	1 (6)
Use of lubricant for [type of sex] in past 6 months, n (%)
None of the time	1 (6)	4 (25)
Some of the time	3 (19)	6 (38)
Half of the time	1 (6)	4 (25)
Most of the time	2 (13)	1 (6)
Every time	9 (53)	1 (6)
Ever used an applicator to insert lubricant into [compartment]? n (%)
Yes	1 (6)	2 (13)
No	15 (94)	14 (88)
Medication use in past 6 months? n (%)
Yes	0	5 (31)
No	16 (100)	11 (69)
Ever used a tampon? n (%)
Yes	n/a	15 (94)
[If yes:] Typically use tampons?		10 (63)
No		1 (6)
Ever used a menstrual cup? n (%)
Yes	n/a	3 (19)
[If yes:] Typically use menstrual cups?		3 (19)
No		13 (81)
Used a dessicant to make vagina dry or tight during vaginal sex the past 6 months? n (%)
No	n/a	16 (100)
Used a spermicide during vaginal sex in past 6 months? n (%)
No	n/a	16 (100)

Black/African American, White/Caucasian and American Indian/Alaska Native (n = 1).

“Living with partner” (n = 1), “partnership” (n = 1).

STI, sexually transmitted infection; SD, standard deviation.

Scale scores and effect sizes

Previously established perceptibility measures were adapted in a prior run-in study for use with the dual compartment suppositories. The measures generally displayed good to excellent internal consistency reliability with Cronbach's coefficient alpha ranging from 0.73 to 0.94 for 31 of the 40 measures (10 measures across 2 products and 2 compartments). Acceptable reliability was found for 6 measures (0.58–0.67), and poor reliability for 3 measures (0.47 for both suppositories in rectal SEX: Noticeable measures, and 0.39 for suppository-B vaginal SEX: Leakage measure). Table 3 presents relevant scale scores for vaginal and rectal use comparisons by product for those scales that yielded at least one medium/large effect size in paired t-test comparisons.

Table 3.

Paired T-Test Comparisons of Relevant USPE Scale Scores for Suppository-A (Supp-A) and Suppository-B (Supp-B) Used in the Rectum and Vagina During RAI and VI, Respectively: ASIS, Standard Deviations, Effect Sizes, and p-Values

Application/Insertion (APP) Scale Scores
Scale	N ^a	Supp-ARectalASIS ^b (SD)^c	Supp-AVaginalASIS ^b (SD)^c	d ^d	p	N ^a	Supp-BRectalASIS ^b (SD)^c	Supp-BVaginalASIS ^b (SD)^c	d ^d	p
APP: Ease	16	4.13 (0.73)	4.25 (0.68)	0.28	.441	15	3.52 (1.09)	3.81 (1.11)	0.56	.145
APP: Product Awareness	16	3.03 (0.88)	2.46 (0.97)	0.99	.014	15	2.65 (0.82)	2.29 (0.75)	0.58	.137
Sexual Activity (SEX) Scale Scores
SEX: Initial Penetration	16	3.31 (1.14)	3.69 (0.84)	0.85	.045	14	2.12 (1.45)	2.40 (1.41)	0.31	.423
SEX: Messiness	16	1.98 (0.75)	2.36 (0.77)	0.57	.126	15	1.79 (0.67)	2.33 (0.86)	0.71	.076
SEX: Perceived Wetness	13	2.77 (0.97)	3.10 (1.06)	0.81	.065	15	2.30 (1.15)	2.35 (1.13)	0.07	.861
SEX: Leakage	16	2.31 (1.01)	2.65 (1.07)	0.30	.413	15	1.85 (0.85)	2.31 (0.97)	0.51	.184
SEX: Effortful	16	1.75 (0.84)	1.31 (0.47)	0.85	.042	15	2.67 (1.37)	2.75 (1.60)	0.15	.704
SEX: Pleasure	15	2.78 (1.22)	2.62 (1.14)	0.15	.683	15	2.04 (1.45)	1.42 (0.74)	0.59	.150
SEX: Pre-Coital Leakage	16	1.35 (0.58)	1.67 (0.79)	0.47	.206	15	1.18 (0.33)	1.44 (0.99)	0.67	.228
SEX: Noticeable	16	2.27 (1.08)	2.23 (1.10)	0.06	.876	15	2.95 (1.25)	3.40 (0.89)	0.61	.145

If a participant did not answer a specific scale item, their score was not calculated for that specific scale for that product. Reported comparison analyses for each scale are based on the actual respondent sample that answered all items on a specific scale for each product.

ASIS calculated by summing the individual item responses on a scale and dividing by the number of items in each scale, providing a common scale metric to aid interpretation; 5-point Likert-type scale, where 1 = “do not agree at all,” 2 = “agree a little,” 3 = “agree somewhat,” 4 = “agree a lot,” and 5 = “agree completely.”

Mean values with standard deviations are given in parentheses.

Adjusted Cohen's d effect size using the correlation between measures; effect sizes of 0.2, 0.5, and 0.8 are considered respectively small, medium, and large.

ASIS, Averaged Scale Item Score; RAI, receptive anal intercourse; USPE, User Sensory Perception and Experience; VI, vaginal intercourse.

Using previous guidelines where effect sizes of 0.2, 0.5, and 0.8 are considered, respectively, small, medium, and large, 13 of 20 comparisons had values in the medium to large range. There were three statistically significant and large effect size differences for perceived sensations in the rectum as compared with the vagina. All three occurred with suppository-A: APP: Product Awareness (p = .014, d = 0.99); SEX: Initial Penetration (p = .045, d = 0.85); and SEX: Effortful (p = .042, d = 0.85). Guided by these results, we reviewed relevant qualitative data, including coding summaries and written memos.

Insertion experience

From a qualitative perspective, the majority (12 of 16 regarding rectal insertion; 13 of 16 regarding vaginal insertion) of our participants discussed feeling suppository-A, using the same term, “waxy bullet,” for both rectal and vaginal insertion. Whereas “torpedo” was the other primary descriptor for rectal experience, a “tampon,” not surprisingly, was the descriptor when commenting on vaginal insertion sensations. Some felt friction during insertion, whereas one felt the suppository shift inside her rectum. One participant noted, “it just felt like I had something small going inside my body [rectum], but it didn't hurt, and it didn't necessarily feel good. It just…didn't bother me.”

Data regarding the physical awareness of suppository-A (i.e., APP: Product Awareness) suggest that the product was felt, at least by some women, in both compartments, but that once inserted, overall, women were less aware of suppository-A in the vagina (ASIS = 2.46) than in the rectum (ASIS = 3.03). One participant described it this way: “I just pushed it all the way till I didn't feel the suppository anymore” and “I felt its presence through the whole insertion process…But once I got it to its destination in my vaginal canal, I didn't feel it anymore…” A different participant specifically noted that suppository-A felt less solid in the vagina than in the rectum.

Sexual activity experience

About one-third of participants reported experiences with suppository-A similar to their typical initial penetration experiences during RAI: one indicated initial penetration felt the way it typically does, another that it felt the way it does when using typical additional lubrication; and three women indicated that rectal initial penetration had the level of discomfort they typically experience. Another one-third commented that initial penetration during RAI using suppository-A was more difficult than it typically is. Other relevant comments included that, during rectal penetration, suppository-A felt like it was being pushed around but did not hinder penetration per se. Two participants described the consistency as like a “very thick gel” or “maple syrup” and as lubricating, but also tacky. Although two participants commented that initial penetration during RAI was easy because suppository-A provided lubrication, several others (5) chose to apply their own lubricant for initial penetration when using suppository-A.

With respect to VI with suppository-A, two participants described their VI initial penetration experiences as typical and/or as it would be with their own natural lubrication. The lubricating quality of suppository-A was often raised in the context of VI: two commented that initial penetration felt lubricated; three others specifically said the lubrication suppository-A provided made vaginal penetration easier. Another said it was less slippery than typical penetration during VI and that suppository-A did not mix well with her own vaginal lubrication. Some participants found suppository-A so lubricating that it decreased sensation, made initial penetration faster and slick, or made it feel more wet. One participant said the product felt thick and sticky during initial penetration, whereas another liked that her vagina felt tighter and fuller during vaginal penetration with suppository-A.

Finally, USPE scale scores for the SEX: Effortful scale also indicated significantly different sensory experiences between vaginal and rectal intercourse experiences when using suppository-A. With respect to VI experiences, discussions were primarily related to the additional lubrication suppository-A provided. Comments ranged from suppository-A making strokes smoother, more lubricated, and not sticky, to being so slippery that, in one case, the vaginal opening felt wider and/or her partner's penis slipped out unexpectedly and, in another, that her partner lost his erection because the slipperiness impeded feeling. To the contrary, one participant felt suppository-A was thick and another said strokes felt dry. RAI experiences with suppository-A were most often discussed in reference to “typical” RAI. Participants described RAI with suppository-A that was easier, had less friction, and was more slippery. Strokes felt smooth, were easier after initial penetration, and lubrication lasted throughout sex, and did not dry out. Two participants commented that their partner's penis was able to go in deeper into their rectum and that strokes were faster than typical. Although one said she felt comfortable and not as dry as typical, another found RAI was dry and more difficult than usual. One participant indicated it was difficult for her partner to thrust, that he could not go in as deeply and that she elected to apply her own additional lubricant. Overall, comparing rectal and vaginal suppository-A experiences shows that the qualitative comments in both compartments mostly pertained to levels of lubrication; in the vagina, the level of lubrication was for some too much, whereas in the rectum, it was, mostly, appropriate and appreciated.

In addition to these three statistically significant pairwise comparisons between vaginal and rectal sensory experiences with suppository-A, there are two nearly statistically significant comparisons with large effect sizes. The SEX: Messiness effect size for suppository-B has a p-value of .076 but a medium effect size (d = 0.71). The SEX: Perceived Wetness effect size for suppository-A has a p-value of .065 but a large effect size (d = 0.81). In addition, there are seven other rectal–vaginal compartment comparisons with moderate effect sizes (but not statistically significant). For suppository-B moderate effect sizes are seen in the following scales: APP: Ease (d = 0.56); APP: Product Awareness (d = 0.58); SEX: Leakage (d = 0.51); APP: Pleasure (d = 0.59); SEX: Pre-coital Leakage (d = 0.67), and SEX: Noticeable (d = 0.61). For suppository-A a moderate effect size is seen for the SEX: Messiness scale (d = 0.57).

Discussion

In this study, two distinct formulations of suppositories were investigated that encompass distinct rheological properties from each other. Suppository-A is a lipophilic formulation that uses temperature to dissolve. Suppository-B is a hydrophilic formulation that relies upon the residual moisture of the rectum/vagina to dissolve. The use of fatty acids and oil in suppositories like suppository-A are nonirritating and well tolerated by rectal and vaginal tissue.^22,39 However, because such suppositories are designed to melt at body temperature, they are susceptible to higher ambient temperatures (tropical) during storage that may reduce structural integrity and insertion. Conversely, hydrophilic suppositories like suppository-B are not temperature dependent. However, drawing moisture from the surrounding tissue can result in a hyperosmolar solution (>275–290 mOsm/kg), which leaves the epithelial and underlying stromal tissue drier and susceptible to microscopic lacerations that may cause irritation and enhance infection. Thus, these formulations represent the major rheological classes of suppositories ranging from “soft” lipophilic molded suppositories to “hard” pressed-formed hydrophilic suppositories, and are necessary to evaluate USPE scale performance, as well as understand the range of sensory experiences users perceive in each compartment.

USPE scale score comparisons indicate that women can clearly distinguish between two rheologically distinct suppositories when the same suppository is used in the vagina and/or the rectum. That women can distinguish sensations elicited by two rheologically distinct suppositories confirms previous data noted with other semi-solid formulations.^31,34 That women can discern variations in sensory experiences within the same formulation across rectal and vaginal compartments is the first result of its kind to our knowledge and underscores the opportunity to leverage perceptibility studies in product development.

The qualitative data provide compelling experiential narrative and the quantitative data support this narrative. The integrated, convergent mixed methods design allowed us to compare and contrast product experiences by each suppository and each compartment, and to provide users' experiential details, and thus credibility, to the quantitative scale scores. Taken together, these data indicate that women were able to clearly discern differences in how suppository-A felt in their rectums versus their vaginas in terms of product awareness during insertion, what sensations they felt as their partner's penis entered their vagina or rectum, and how much effort was perceived to be needed during sex when the suppository was present. Women felt suppository-A more in their rectum than in their vagina during insertion. During sex using suppository-A, women found initial vaginal penetration was easier than initial rectal penetration. Results also indicate that more effort was required for penile penetration and strokes during RAI than for VI using suppository-A.

The scale scores and effect sizes align well with the known rheology of the two suppositories evaluated here, as well as their hypothesized performance in each compartment. In addition to examining those individual scale score differences that reached statistical significance, the other comparisons that had medium-to-large effect sizes that did not reach statistical significance are important indicators of potentially meaningful differences that might be significant and should be investigated with larger samples. Suppository-A is a softer suppository that is porous and melts faster, becoming more gel-like quickly; this helps to explain why differences, as noted by medium-to-large effect sizes, occur early in the sexual encounter, hence the greater effect sizes for the SEX: Initial Penetration and SEX: Perceived Wetness scales. In contrast, dissolution takes longer for suppository-B. Suppository-B, therefore shows larger differences as indicated by effect sizes on scales that measure sensory experiences that occur later in the sexual encounter: for example, SEX: Pleasure, SEX: Messiness, and SEX: Noticeable. It is differences in sensation experienced, as well as the meanings users derive of these experiences, that should be considered in formulation design and development, given the link previously established between USPEs and users' willingness to use products that elicit such sensations.

Study considerations

Our findings of the sensory experiences of dual compartment suppository use presented here were drawn from 16 participants in the northeastern United States. The meaning-making of sensory experiences may vary in samples that differ demographically or culturally. Consistency within participant in their use of condoms and/or lubricants provide little opportunity for study of sensory experiences of suppository use-alone versus with a condom and/or personal lubricant in any individual, as well as comparisons in experiences across compartments. That said, qualitative data do not vary in any meaningful way between women who used or did not use these products. Overall, what we can infer from these data is that, given a woman's “typical” sexual episode (whether with or without condom, and/or with or without lubricant), the suppositories seem to elicit unique sensory experiences separate from (or above and beyond) those typically experienced by each woman.

The order of product use may have played a role in participants' experiences and ratings. First use could benefit from novelty or sensory vigilance, whereas subsequent uses could benefit from learning. Conversely, recall requirements alone can mute or highlight order experiences, respectively. Participants with more frequent use of vaginal or rectal products and more frequent vaginal or rectal intercourse may find inserting products into respective compartments easier, and more familiar overall, whereas those, for instance, reporting less frequent intercourse in either compartment may find such experiences less familiar and/or more salient. Larger sample sizes in future studies may help address these concerns. Our qualitative data were collected in one interview held after all product use was concluded and during which participants reflected on the use of multiple products. Differing recall, as well as order effects, may have shaped those data as well. Nevertheless, the USPE scales identify relevant and qualitatively credible differences in the dual compartment use of these two suppositories. Furthermore, both the quantitative and qualitative data provide a clear user experience narrative suggesting that women do experience differences between vaginal and rectal use of suppositories with different rheological properties.

Conclusion

Consistent and correct dual compartment suppository use will depend on the development of a product that women find acceptable in the vagina and the rectum. Yet, as our qualitative data show, ideal product characteristics may differ for each compartment. For example, users may want a dual compartment product to provide similar levels of pleasure during intercourse in both the vagina and rectum but want more lubrication at initial penetration in the rectum than in the vagina. Understanding, and effectively measuring, the experiences women have with dual compartment products will enable developers to identify a product that hits the “sweet spot” for both the vaginal and rectal sensory experience. To do this, product developers need to elucidate ideal characteristics for vaginal and rectal use, identify where those characteristics overlap for both compartments, and use that information to guide ideal product design. Identifying which scales and perceived experiences are most relevant between compartments is a significant step in identifying the meaningful range in which to develop dual compartment formulations.

Disclaimer

The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Footnotes

Acknowledgments

The authors express their deepest gratitude to the participants who engaged in this study: their willingness to do so provides critical knowledge toward the ultimate goal of ending the HIV pandemic. The authors also express their sincere thanks to Bruce Barton and Renee Fraatz for their unfailing attention to the details of research pharmacy, and to Candelaria Barroso for her skill and clarity in the qualitative data, as well as her enduring support of this team and work. J.J.R. is now affiliated with LifePro Financial Services, Inc., San Diego, CA. A.S.H. is now affiliated with Novavax, Gaithersburg, MD. K.A.S. is now affiliated with the Population Studies and Training Center (PSTC), Brown University, Providence, RI.

Authors' Contributions

K.M.G.: study conceptualization, study design, data analysis and interpretation, article writing. R.K.R.: study design, data collection, data analysis and interpretation, article writing. M.G.: study procedures implementation, data collection, data management and analysis, article writing and review. J.J.R.: study procedures implementation, data collection, data management, article review. S.E.V.: study design, data collection, data analysis and interpretation, article review. J.L.F.: quantitative data analysis, article review. A.S.H.: formulation design, data interpretation, article writing and review. D.F.K.: formulation design, data interpretation, article review. S.C.-U. and S.T.: clinical screening, article review. K.A.S.: study procedures implementation, data collection, data management and analysis, article review. K.W.B.: article review. R.W.B.: study conceptualization, article review.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

Research reported in this publication was supported by The National Institute of Allergies and Infectious Diseases (NIAID) of the National Institutes of Health (NIH) under Grant No. U19AI101961.

References

National Center for HIV/AIDS. HIV and Women. 2019. Available at https://www.cdc.gov/hiv/pdf/group/gender/women/cdc-hiv-women.pdf, accessed January 23, 2021 .

Joint United Nations Programme on HIV/AIDS (UNAIDS). Ending AIDS: progress towards the 90-90-90 targets. 2017. Available at https://www.unaids.org/sites/default/files/media_asset/UNAIDS_2017_ENDINGAIDS_Slides_en.pdf, accessed January 23, 2021 .

Myer

, Kuhn

, Stein

, Wright

Jr ., Denny

: Intravaginal practices, bacterial vaginosis, and women's susceptibility to HIV infection: Epidemiological evidence and biological mechanisms. Lancet Infect Dis, 2005; 5:786–794.

Chersich

, Rees

: Vulnerability of women in southern Africa to infection with HIV: Biological determinants and priority health sector interventions. AIDS, 2008; 22 Suppl 4:S27–S40.

Pulerwitz

, Amaro

, De Jong

, Gortmaker

, Rudd

: Relationship power, condom use and HIV risk among women in the USA. AIDS Care, 2002; 14:789–800.

Santelli

, Edelstein

, Mathur

, et al.: Behavioral, biological, and demographic risk and protective factors for new HIV infections among youth in Rakai, Uganda. J Acquir Immune Defic Syndr, 2013; 63:393–400.

McBride

, Fortenberry

: Heterosexual anal sexuality and anal sex behaviors: A review. J Sex Res, 2010; 47:123–136.

Flannery

, Ellingson

, Votaw

, Schaefer

: Anal intercourse and sexual risk factors among college women, 1993–2000. Am J Health Behav, 2003; 27:228–234.

Baggaley

, White

, Boily

: HIV transmission risk through anal intercourse: Systematic review, meta-analysis and implications for HIV prevention. Int J Epidemiol, 2010; 39:1048–1063.

10.

Centers for Disease Control and Prevention. HIV Risk Behaviors. 2019. Available at https://www.cdc.gov/hiv/group/gender/women/risk-behaviors.html, accessed January 23, 2021 .

11.

McGowan

: Rectal microbicides: Can we make them and will people use them?. AIDS Behav, 2011; 15 Suppl 1:S66–S71.

12.

Giguere

, Dolezal

, Bauermeister

, et al.: Influence of partner type on acceptability and likelihood of use of a rectal microbicide among young men who have sex with men in the United States and Puerto Rico. J Sex Res, 2016; 53:633–641.

13.

Carballo-Dieguez

, Giguere

, Dolezal

, et al.: Adherence to rectal gel use among mainly ethnic minority young men who have sex with men during a 3-month placebo gel trial: Implications for microbicide research. AIDS Behav, 2014; 18:1726–1733.

14.

Chakrapani

, Newman

, Shunmugam

, et al.: “Like Holding an Umbrella Before It Rains”: Acceptability of future rectal microbicides among men who have sex with men in India-a modified technology acceptance model. Qual Health Res, 2017; 27:1236–1248.

15.

Bauermeister

, Giguere

, Dolezal

, et al.: To use a rectal microbicide, first insert the applicator: Gel and applicator satisfaction among young men who have sex with men. AIDS Educ Prev, 2016; 28:1–10.

16.

Gross

, Buchbinder

, Celum

, Heagerty

, Seage

, 3rd: Rectal microbicides for

U.S

. gay men. Are clinical trials needed? Are they feasible? HIVNET Vaccine Preparedness Study Protocol Team. Sex Transm Dis, 1998; 25:296–302.

17.

Exner

, Correale

, Carballo-Dieguez

, et al.: Women's anal sex practices: Implications for formulation and promotion of a rectal microbicide. AIDS Educ Prev, 2008; 20:148–159.

18.

Hussain

, Ahsan

: The vagina as a route for systemic drug delivery. J Control Release, 2005; 103:301–313.

19.

van Hoogdalem

, de Boer

, Breimer

: Pharmacokinetics of rectal drug administration, Part II. Clinical applications of peripherally acting drugs, and conclusions. Clin Pharmacokinet, 1991; 21:110–128.

20.

de Boer

, Moolenaar

, de Leede

, Breimer

: Rectal drug administration: Clinical pharmacokinetic considerations. Clin Pharmacokinet, 1982; 7:285–311.

21.

Nunes

, Sarmento

, das Neves

: Formulation and delivery of anti-HIV rectal microbicides: Advances and challenges. J Control Release, 2014; 194:278–294.

22.

Ham

, Buckheit

Jr : Designing and developing suppository formulations for anti-HIV drug delivery. Ther Deliv, 2017; 8:805–817.

23.

Garg

, Tambwekar

, Vermani

, et al.: Development pharmaceutics of microbicide formulations. Part II: Formulation, evaluation, and challenges. AIDS Patient Care STDS, 2003; 17:377–399.

24.

Garg

, Kandarapu

, Vermani

, et al.: Development pharmaceutics of microbicide formulations. Part I: Preformulation considerations and challenges. AIDS Patient Care STDS, 2003; 17:17–32.

25.

Pines

, Gorbach

, Weiss

, et al.: Acceptability of potential rectal microbicide delivery systems for HIV prevention: A randomized crossover trial. AIDS Behav, 2013; 17:1002–1015.

26.

Zaveri

, Powell

, Guthrie

, Bakke

, Ziegler

, Hayes

: Qualitative exploration of intrinsic and extrinsic factors that influence acceptability of semisoft vaginal suppositories. BMC Womens Health, 2018; 18:170.

27.

, Zaveri

, Ziegler

, Hayes

: Shape of vaginal suppositories affects willingness-to-try and preference. Antiviral Res, 2013; 97:280–284.

28.

Ham

, Nugent

, Peters

, et al.: The rational design and development of a dual chamber vaginal/rectal microbicide gel formulation for HIV prevention. Antiviral Res, 2015; 120:153–164.

29.

Dezzutti

, Rohan

, Wang

, et al.: Reformulated tenofovir gel for use as a dual compartment microbicide. J Antimicrob Chemother, 2012; 67:2139–2142.

30.

Guthrie

, Dunsiger

, Vargas

, et al.: Perceptibility and the “Choice Experience”: User Sensory perceptions and experiences inform vaginal prevention product design. AIDS Res Hum Retroviruses, 2016; 32:1022–1030.

31.

Guthrie

, Fava

, Vargas

, et al.: The role of volume in the perceptibility of topical vaginal formulations: User sensory perceptions and experiences of heterosexual couples during vaginal sex. AIDS Res Hum Retroviruses, 2020; 36:1059–1070.

32.

Guthrie

, Rohan

, Rosen

, et al.: Vaginal film for prevention of HIV: Using visual and tactile evaluations among potential users to inform product design. Pharm Dev Technol, 2018; 23:311–314.

33.

Guthrie

, Rosen

, Vargas

, et al.: User input in iterative design for prevention product development: Leveraging interdisciplinary methods to optimize effectiveness. Drug Deliv Transl Res, 2017; 7:761–770.

34.

Morrow

, Fava

, Rosen

, et al.: Designing preclinical perceptibility measures to evaluate topical vaginal gel formulations: Relating user sensory perceptions and experiences to formulation properties. AIDS Res Hum Retroviruses, 2014; 30:78–91.

35.

Weld

, Hiruy

, Guthrie

, et al.: A comparative pre-phase I study of the impact of gel vehicle volume on distal colon distribution, user experience, and acceptability. AIDS Res Hum Retroviruses, 2017; 33:440–447.

36.

Guest G, MacQueen KM, Namey EE: Applied Thematic Analysis. Sage, Thousand Oaks, CA, 2011.

37.

Creswell JW, Clark VLP: Designing and Conducting Mixed Methods Research. Sage, Thousand Oaks, CA, 2017.

38.

Cohen

: Statistical Power Analysis for the Behavioral Sciences. 2nd ed. Lawrence Erlbaum, Hillsdale, NJ, 1988.

39.

Moreton

: Suppository bases, hard fats. In: Handbook of Pharmaceutical Excipients (Rowe RC, Sheskey PJ, Cook WG, Fenton ME, eds.) The Pharmaceutical Press, Inc., London, 2012.