- Email: [email protected]
Comparative crossover study of the PATH Woman's Condom and the FC Female Condom®
Contraception 78 (2008) 465 – 473
Original research article
Comparative crossover study of the PATH Woman's Condom and the FC Female Condom® Jill L. Schwartza,⁎, Kurt Barnhartb , Mitch D. Creininc , Alfred Poindexterd , Angie Wheelesse , Maggie Kilbourne-Brookf , Christine K. Maucka , Debra H. Weinere , Marianne M. Callahana a Department of Obstetrics and Gynecology, CONRAD, Eastern Virginia Medical School, Arlington, VA 22209, USA Department of Obstetrics and Gynecology and Center for Clinical Epidemiology and Biostatistics, University of Pennsylvania, Philadelphia, PA 19104, USA c Department of Obstetrics, Gynecology and Reproductive Sciences, University of Pittsburgh and Magee-Women's Research Institute, Pittsburgh, PA 15213, USA d Advances in Health, Inc., Houston, TX 77030, USA e Family Health International (FHI), Research Triangle Park, NC 27713, USA f Program for Appropriate Technology in Health (PATH), Seattle, WA 98107, USA Received 7 May 2008; revised 18 July 2008; accepted 18 July 2008 b
Abstract Background: Only one female condom [FC1 Female Condom (FC1)] is currently marketed, but it is poorly utilized, perhaps due to difficulty with insertion, discomfort and suboptimal functional performance during intercourse. The Program for Appropriate Technology in Health (PATH) Woman's Condom (WC) was developed in an effort to overcome these obstacles. Study Design: This was a randomized crossover study to evaluate the functional performance, safety and acceptability of the FC1 and WC. Seventy-five couples were assigned to one of two condom use sequences (WC/FC1 or FC1/WC) at three centers. Four condoms of the first type were used by couples in four acts of intercourse at home over a 2–4-week period. After a follow-up visit, these procedures were repeated with the second assigned condom type. In a substudy of participants (n=25), a colposcopy was performed prior and subsequent to the first condom use of each of the two condom types. Condom performance was evaluated by calculating measures of function from questionnaires completed by the couple after each condom use. Safety was evaluated by reported urogenital symptoms with a given condom during or immediately following condom use and colposcopic signs of genital irritation in the substudy. Acceptability of each given condom type was measured by questionnaire. Results: Total condom failure (slippage, breakage, etc., divided by the number of female condoms opened) was 31% for the WC and 42% for the FC1. Total clinical failure (slippage, breakage, etc., divided by the number of female condoms used) was 17% for the WC and 24% for the FC1. The proportion of condom failures was 10.9 percentage points less, and the proportion of clinical failure 6.7 percentage points less, when couples used the WC compared to the FC1 [90% CI: −18.5 to −3.3 and −12.6 to −0.8, respectively). Fewer women reported symptoms of urogenital irritation when using the WC vs. the FC1 either overall or when analyzing each use of the condom [woman as unit: −20 percentage points (90% CI: −30.5 to −9.3); condom use as unit: −12.3 percentage points (90% CI: −18.0 to −6.7)]. A similar result was seen for signs of urogenital irritation [woman as unit: −20 percentage points (90% CI: −42.7 to 4.8)]. Among participants with a preference, WC was preferred over the FC1 by twice as many males and by 2.6 times as many females. Conclusions: While both female condoms were safe and acceptable in short-term use, the PATH Woman's Condom leads to less failure, was associated with fewer adverse events, and was more acceptable than the FC1 Female Condom. © 2008 Elsevier Inc. All rights reserved. Keywords: Female condom; PATH Woman's Condom; Reality®, FC Female Condom®; Slippage; Breakage
1. Introduction
⁎ Corresponding author. Tel.: +1 703 524 4744; fax: +1 703 524 4770. E-mail address: [email protected] (J.L. Schwartz). 0010-7824/$ – see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.contraception.2008.07.020
The Reality® Female Condom (currently called FC Female Condom®) (FC1; Female Health Company, Chicago, IL, USA) was approved by the US Food and Drug
466
J.L. Schwartz et al. / Contraception 78 (2008) 465–473
Administration (FDA) in 1993 as the first barrier contraceptive for women that offered limited protection against sexually transmitted infections (STIs). This female condom has a 6-month typical use pregnancy rate of 12.4 per 100 woman-years in a US population [1], has not been associated with trauma to the genital tract [2], prevents trichomoniasis recurrence [3], and may offer as much protection from STIs as male condoms [4]. The FC2 has been introduced as an updated version of this condom using synthetic latex for greater affordability [5]. Despite global need for a female controlled barrier device, there has been poor sustained utilization of this method and the potential impact of the device, as an HIV prevention strategy has not been fully evaluated. In female condom studies conducted over the past decade, varying proportions of users have liked the device and succeeded in using it, resulting in differing acceptability ratings in short-term acceptability studies, depending on the study population and the type of intervention [6–10]. These studies have identified a constellation of problems with female condom use including difficulty with insertion, discomfort and suboptimal functional performance during intercourse that may contribute to poor utilization. The Program for Appropriate Technology in Health (PATH), a nonprofit, international health organization, has used an iterative user-driven development process to produce a novel female condom product in an attempt to overcome these obstacles. The PATH Woman's Condom (WC) has a thin, pliable plastic pouch that conforms to the shape of the vagina, is 9 in. long and has a flexible soft outer ring that is intended to protect the external genitalia (Fig. 1). The four oblong foam pieces on the outside of the pouch are intended to cling lightly to vaginal walls, ensuring stability of the device. The distal end of the pouch and foam pieces are packaged into a capsule which serves as an insertion aid and dissolves quickly when inserted into the vagina. The PATH Woman's Condom was
easy to use and comfortable in a short-term acceptability study [11]. This study was designed to further test the functional performance, safety and acceptability of this novel female condom and to gain additional information for further development. This crossover study compared the PATH Woman's Condom (WC) to the FC1 after four uses of each condom type by 75 couples.
2. Materials and methods Seventy-five couples were enrolled between December 2004 and August 2005 in this Phase I, comparative, openlabel, two-period crossover study comparing the WC to the FC1 at three clinical centers: University of Pennsylvania Medical Center; University of Pittsburgh/Magee Womens Research Institute; and Advances in Health, Houston, TX, USA. The sample size was not selected on the basis of hypothesis testing. The crossover design was chosen because between-patient variation can be eliminated, and smaller groups can be used. Couples who agreed to participate were in good health, at least 18 years old, and at low risk for HIV or STIs. Those couples at risk for pregnancy used a steroidal contraceptive (oral, patch, injectable or implantable) or sterilization (vasectomy or tubal ligation) throughout their participation. Individuals who had sensitivities or allergies to study products or prior experience with female condoms were excluded. The protocol and informed consents for this nonsignificant risk device study were reviewed and approved by the site institutional review boards. Informed consent was obtained from each member of the couple. The female volunteer provided a urine specimen for a pregnancy test and height and weight were measured. Volunteers in the colposcopy substudy at the University of Pennsylvania Medical Center (n=25) also underwent a pelvic
Fig. 1. PATH Woman's Condom.
J.L. Schwartz et al. / Contraception 78 (2008) 465–473
examination with Pap smear, if not performed in the past year; wet mount and Neisseria gonorrhoeae and Chlamydia trachomatis polymerase chain reaction testing. Enrolled couples were randomized to one of two condom use sequences. Participants were given either a WC research study packet or a FC1 research study packet as determined by the assigned condom use sequence for the first condomuse period to use in four separate acts of intercourse over a 2–4-week period. The PATH Woman's Condom (WC) (PATH, Seattle, WA, USA) is an investigational device that is not prelubricated. The FC1, a 6.5-inch long polyurethane plastic pouch with two flexible rings, was used as the control female condom because it is the only FDA-approved female condom. Participants were instructed to wait 30–60 s after inserting the WC to have intercourse to allow for the insertion capsule to dissolve. A total of four condoms, each along with a 3-mL package of Astroglide® (BioFilm, Vista, CA, USA), and instructions for use were distributed. The couples also went home with a detailed condom use questionnaire to complete after each intercourse, and an acceptability questionnaire to be completed separately by each person at the end of each condom use period. Female participants had a follow-up visit after completing the four acts of intercourse, at which time questionnaires and diaries were reviewed and collected, and second assigned condoms and questionnaires and diaries were distributed. Female participants in the substudy underwent a baseline colposcopy prior to the first condom use of each of the two condom types. These women were abstinent from using vaginal products and all sexual activity starting 72 h prior to the baseline colposcopy until after intercourse with the study condom. The couples were to use the condom for the first act of intercourse within 72 h of the baseline colposcopy and the female participant was to return within 6 h after intercourse. After intercourse, colposcopy was performed to examine the immediate effects on the cervicovaginal epithelium of the foam shapes and insertion capsule components of the WC and to compare this to the effect of the inner ring of the marketed FC1, according to the CONRAD/World Health Organization Manual [12]. This was a single-blinded study with only the clinicians performing the colposcopies unaware of the participant's condom assignment. Due to the obvious difference in condom design, the participants could not be blinded to the condom type being used. The random permuted blocks method was used to generate random condom-use sequences for 25 participants per site. The sequences were created by a randomization manager using a verified program based on the random function RANUNI in the SAS® System (SAS Institute, Cary, NC, USA). Allocations were sealed in separate individually sealed envelopes identified by participant number. The condom performance outcome was measured by calculating the functional measures (breakage, slippage, etc.) described in Appendix A from questionnaires completed by the couple after each condom use. The
467
primary safety outcomes were the proportion of participants with urogenital symptoms and the proportion of condom uses during or immediately after which urogenital symptoms were noted. These symptoms included urogenital pain and discomfort (e.g., burning, stinging or pressure), dysuria, genital irritation or rash, genital pruritus, abnormal vaginal discharge, pelvic or lower abdominal pain or intermenstrual bleeding for the female and urogenital pain or discomfort and genital irritation or rash for the male participants. The substudy safety outcome measured the proportion of female participants with signs of genital irritation of the external genitalia, cervix and vagina as observed on colposcopic examination after one act of intercourse with a given condom type. Adverse events (AEs) were coded in the MedDRA dictionary based on blind review of the AE description. Acceptability was measured for each given condom type by questionnaire. All statistical analyses were descriptive or involved calculation of 90% confidence intervals (CIs) about the event proportions. Generalized estimating equations were used to model the differences in proportions of condom failure when couples used the WC versus the FC1, controlling for center and taking into account clustering of condom use within couple. In addition, differences in proportions between the two types of condoms, together with 90% CIs, were calculated for signs and symptoms of urogenital irritation. The 90% CI for the difference in proportions between treatment represents the range of differences in treatment that cannot be rejected (by a twosided test, alpha=.10) given the observed data.
3. Results Of the 82 couples screened for eligibility, 75 were enrolled and 73 completed the study (Fig. 2). The functional performance definitions are detailed in Appendix A. Participant characteristics are described in Table 1. Participants at the Houston site differed from those at the other two sites in that they were likely to be older (female mean=34.3, male mean=35.5), identified themselves as Hispanic or black more frequently, and were less educated (female mean=13.2 years, male mean=12.6). They were also more likely to have had a previous female sterilization or hysterectomy, weigh more (female mean=172.9), and have previously been pregnant. Nine enrolled couples did not open and use all 4 of either or both condom types. Five couples used all four WC but only used one to three FC1 because of discomfort. Three couples used three to four FC1 and zero to two WC. One couple used only one WC after which the female participant developed bilateral labial swelling and erythema and was discontinued early from the study. Condom use questionnaires were completed for 287 WC uses and for 285 FC1 uses. Of those, four WC and seven FC1 broke while opening the packages, four
468
J.L. Schwartz et al. / Contraception 78 (2008) 465–473
FC1, respectively) or complete slippages (6% and 8%, for WC and FC1, respectively). The most common failure type contributing to clinical failure was outer ring displacement which was 9.9% for both condom types (Table 2). Approximately 21 couples (29%) experienced at least one outer ring displacement during WC use compared to 17 couples (23%) during FC1 use. Nine couples reported outer ring displacement with at least one use of both products. Couples used lubrication for the WC in 91% of condom uses and for the FC1 in 61% of condom uses (the FC1 is prelubricated). When asked about the relationship of condom failure to lubricant, 24 (41%) out of 59 respondents who reported a failure (turned inside out, slipped out and/or pulled out) with the WC felt that it was
Table 1 Participant characteristics Female (n=75) Fig. 2. Flow diagram of study.
WC and five FC1 were not inserted, and no vaginal intercourse occurred for five WC uses and one FC1 use. Therefore, a total of 274 WC and 272 FC1 were used during intercourse. Most condoms were inserted by the female participant (80%) and in the vagina as per instructions (97%). The study condoms were worn during vaginal intercourse for an average of 16 min with a range of 1–93 min. A greater percentage of FC1 uses compared to WC uses were associated with difficulty handling the device before insertion (44% compared to 9%), difficulty placing the device in the vagina (36% compared to 8%), discomfort during insertion (21% compared to 8%) and device instability after insertion (16% compared to 9%). About one third of uses of each device were associated with uncertainty as to how deep to place the device (FC1=26.0%, WC=28.1%) and whether the device was correctly placed (FC1=29.7%, WC=26.9%). Total condom failure (Appendix A) was 31% for the WC and 42% for the FC1 and total clinical failure (Appendix A) was 17% for the WC and 24% for the FC1 (Table 2). The proportion of condom failures was 10.9 percentage points less, and the proportion of clinical failure 6.7 percentage points less, when couples used the WC compared to the FC1 (90% CI: −18.5 to −3.3 and −12.6 to −0.8, respectively) (Table 2). WC leads to significantly less failure than FC1 with regard to partial turning inside out and partial slippage and there is a trend toward significance with complete slippage. Approximately 44% of the couples during WC use and 53% of couples during FC1 use experienced at least one event that contributed to total clinical failure. A minority of couples experienced clinical breaks (4% and 0%, for WC and
n Age (years) Mean±S.D. 29.2±8.3 Range 19–60 Ethnicity Hispanic 13 Not Hispanic 62 Race Asian 3 Black or African American 15 White 55 More than one race 2 Other 0 Education Mean±S.D. 14.9±2.5 Ever used Oral contraceptives 70 Male condom 73 Female condom 0 Diaphragm 3 Primary method Tubal ligation 9 Oral contraceptives 46 Hormonal injection 8 Hormonal patch 6 6 Othera Height (in.) Mean±S.D. 64.2±2.5 Range 59–72 Weight Mean±S.D. 153.1±33.8 Range 110–260 Ever pregnant Yes 34 Current relationship with male partner Living together 41 Length of time with male partner (years) Mean±S.D. 5.5±6.8 Range 0.3–40
Male (n=75) %
n
%
30.7±8.5 19–58 17 83
13 62
17 83
4 20 73 3 0
1 15 56 2 1
1 20 75 3 1
14.9±2.7 93 97 0 4 12 61 11 8 8
45 55
a Other methods used while in study include: hysterectomy, post menopausal, and vaginal ring.
J.L. Schwartz et al. / Contraception 78 (2008) 465–473
469
Table 2 Condom functionality PATH WC (n=287)
a
Total number of condoms opened Broke while opening Broke during insertion Broke other (capsule off or missing) Total number of condoms usedb Broke during intercourse Don't know when Type of failurec Clinical breakage Nonclinical breakage Total breakage Incorrect penetration Pullout Complete turning inside out Partial turning inside out Complete slippage Partial slippage Outer ring displacement Total clinical failured Total condom failured
n
%
90% CI
287 4 0 2 274 2 1
100.0 1.4 0.0 0.7 95.5 0.7 0.4
3 6 9 6 37 2 12 4 29 27 47 89
1.1 2.1 3.1 2.2 13.5 0.7 4.4 1.5 10.6 9.9 17.2 31.0
0.3–2.8 0.9–4.1 1.7–5.4 1.0–4.3 10.2–17.4 0.1–2.3 2.6–7.0 0.5–3.3 7.7–14.2 7.1–13.3 13.5–21.3 26.5–35.8
FC1 Female Condom (n=285)
WC-FC 1
n
%
90% CI
Difference
90% CI
285 7 2 0 272 0 0
100.0 2.5 0.7 0.0 95.4 0.0 0.0
0 9 9 7 47 1 26 10 42 27 65 120
0.0 3.2 3.2 2.6 17.3 0.4 9.6 3.7 15.4 9.9 23.9 42.1
0.0–1.1 1.7–5.5 1.7–5.5 1.2–4.8 13.6–21.5 0.0–1.7 6.8–13.0 2.0–6.2 12.0–19.5 7.1–13.4 19.7–28.5 37.2–47.1
– −1.1 0.1 −0.2 −1.9 0.4 −4.6 −2.1 −5.1 −0.4 −6.7 −10.9
−3.0 to 0.8 −2.1 to 2.3 −2.0 to 1.7 −6.7 to 2.9 −0.7 to 1.3 −8.3 to −0.8⁎ −4.2 to 0.0 −9.6 to −0.7⁎ −4.5 to 3.8 −12.6 to −0.8⁎ −18.5 to −3.3⁎
a
Denominator for nonclinical breakage, total breakage and total failure. Denominator for clinical breakage, incorrect penetration, pullout, complete turning inside out, partial turning inside out, complete slippage, partial slippage, outer ring displacement and total clinical failure. c A condom may be counted for more than one type of failure. d A condom only contributed to total failure or clinical failure once, even if it had more than one type of failure. ⁎ Since the upper bounds of these CIs do not include 0, there is a difference (p≤.10, two-tailed) between condom types. b
related to lubrication. Likewise, 40 (46%) of 87 respondents who reported a failure with the FC1 condom felt that it was related to lubrication. The most common reasons given for the failure of both condom types were that there was not enough lubricant and that the lubricant was dried out or sticky. Men ejaculated during 228 of the 274 WC uses (83%) and during 205 of the 272 FC1 uses (75%). Most of the study condoms had to be pulled gently for removal; however, over a third were pulled straight out without twisting to remove which may account for the semen spillage reported in about 7% of the WC and FC uses.
Condom failure rates for all types of failure were higher during the first condom use period compared to the second condom use period, regardless of condom type. Types of urogenital symptoms are listed in Table 3. Fewer women reported symptoms of urogenital irritation when using the WC versus the FC1 either overall (e.g., taking the woman as the unit of analysis) or when analyzing each use of the condom [woman as unit: −20 percentage points (90% CI: −30.5 and −9.3); condom use as unit: −12.3 percentage points (90% CI: −18.0 and −6.7)]. Five women had at least one AE with both condom types. One participant had bilateral labial swelling and
Table 3 Urogenital symptoms Adverse event
Genital pain Genital discomfort Genital rash Pelvic pain Labial swelling Intermenstrual bleeding Genital pruritus Vaginal candidiasis Dysuria Total number of events/women with at least one eventa
PATH Woman's Condom n=73
FC1 Female Condom n=74
Number of events
Number of women (%)
Number of events
Number of women (%)
8 7 2 0 1 0 0 1 0 19
6 (8%) 4 (5%) 1 (1%) 0 1 (1%) 0 0 1 (1%) 0 9 (12%)
30 25 0 6 0 1 1 1 3 67
13 (18%) 10 (14%) 0 4 (5%) 0 1 (1%) 1 (1%) 1 (1%) 3 (4%) 24 (32%)
a WC-FC1 difference (90% CI) for condom as a unit: −12.3 (−18.0 to −6.7) and for women as a unit: −20.0 (−30.5 to −9.3). Since the upper bounds of these CIs do not include 0 there is a difference (p≤.10, two-tailed) between condom types.
470
J.L. Schwartz et al. / Contraception 78 (2008) 465–473
Table 4 Colposcopic findings Finding
PATH Woman's Condom Baseline PATH Woman's Condom FC1 Female Condom Baseline (before intercourse) (before intercourse) New findingsa
FC1 Female Condom New findingsa
n=23 Erythema 14 Petechiae 2 Ecchymoses 0 Edema 4 Ulcer 1 Peeling 0 Grossly white finding 0 Other 8 Total number of findings 29 Total number of women with at 14 least one findingd
n=23
n=25
n=25
1 2 0 0 0 1 1 2b 7 6 (26%)
10 3 0 0 0 1 1 8 23 14
13 1 0 1 0 0 0 2c 17 11 (44%)
a New findings include all new, noniatrogenic colposcopic findings that were not observed on the baseline colposcopy or findings observed on the baseline colposcopy that worsened during intercourse with the given condom, regardless of relatedness as determined by the site study clinician. b Nabothian cyst and prominent vessel. c Prominent vessels and rash-like erythema. d WC-FC1 difference for new findings (90% CI) for women as a unit=−20.0 (42.7 to 4.8). Since the upper bounds of this CI include 0 there is no difference (p≤.10, two-tailed) between condom types.
erythema after use of her first WC. Among male participants, there were a total of 14 AEs associated with product use reported by 6 (8%) out of 73 men using at least one WC and a total of 13 AEs associated with product use reported in 8 (11%) out of 74 using at least one FC1. Mild penile discomfort, pressure, irritation and itching were the most common male complaints. Similar results were observed for male symptoms when condom use was the unit of analysis. Overall, men experienced slightly fewer symptoms of urogenital irritation when using the WC versus the FC1 (man as unit: −2.7 percentage points, 90% CI: −10.7 to 5.0; condom use as unit: −0.4, 90% CI: −4.2 to 3.4). No severe or serious AEs were reported by male or female participants. Types of colposcopic findings are listed in Table 4. Fewer women had colposcopic findings of urogenital irritation when using the WC versus the FC1 (woman as unit: −20 percentage points, 90% CI: −42.7 to 4.8). All of the seven findings observed with the WC were located on the cervix. Thirteen of the 17 findings observed with the FC1 were erythema. Eleven findings were located on the external genitalia, five were located on the cervix, and one was located on the vaginal wall. None of the findings involved deep disruption of the epithelium and none were considered serious. After using both condom types, 41 (57%) out of 72 of the female partners liked the WC more, 16 (22%) out of 72 liked the FC1 more and 15 (21%) out of 72 had no preference. After using both condom types, 31 (43%) out of 72 of the male partners liked the WC more, 16 (22%) out of 72 liked the FC1 more and 25 (35%) out of 72 had no preference. Female participants gave more favorable mean scores to the WC for the majority of acceptability parameters (Table 5). On average, female participants liked the WC minimally better and the FC1 slightly worse than male condoms used
prior to the study (Table 5). Female participants gave more favorable mean scores (1=very easy to 7=very difficult) on only one aspect of the FC1; convenience of lubrication (WC −2.6, FC1 −1.9) (Table 5). Male partners gave more favorable or similar means scores to the WC in all categories except for convenience of lubrication.
4. Discussion This was the first breakage and slippage study of the PATH Woman's Condom. The WC performed better than the FC1 with lower total clinical failure and total failure using female condom performance definitions adapted from male condom definitions of slippage and breakage [13]. There was significantly less failure with WC than with FC1 in regard to partially turning inside out and partial slippage and a trend toward significance with complete slippage. The statistical significance of these results should be interpreted cautiously, as the small sample size and relatively few condom uses per couple meant that “nuisance” variables for period effects (a difference in results between the condom type used during the first crossover period vs. the condom type used in the second) and carryover (an effect of the first condom type lingering past the washout period and affecting results with the second condom type) could not be tested with sensitivity. WC production was still at the hand-made prototype stage when inventory was developed for this study, so a few clinical breaks probably resulting from the hand welding was not a surprising result. Using a recently revised definition of failure modes specific for female condoms [14], which does not include partial turning inside out, total clinical failure rate for WC is reduced from 17% to 13% and that for FC1 is reduced
J.L. Schwartz et al. / Contraception 78 (2008) 465–473
471
Table 5 Female partner acceptability PATH Woman's Condom (n=73) Mean±S.D.
FC1 Female Condom (n=73) Mean±S.D.
Ease of handling condom Ease of inserting condom Ease of following instructions Convenience of lubrication Ease of removing condom
2.1±1.4 2.3±1.3 1.5±.82 2.6±2.0 1.7±1.1
3.7±2.0 3.8±2.0 2.0±1.4 1.9±1.3 1.9±1.3
Comfort during insertion of condom Feel of lubricant during insertion Feel of material Overall comfort during use Comfort/fit of outer ring during use Feel of lubricant during use of condom Comfort during removal of condom
2.6±1.3 2.4±1.5 3.0±1.7 3.2±1.6 3.5±1.7 2.5±1.5 2.0±1.2
3.5±1.8 2.4±1.2 3.5±1.7 4.2±1.9 3.8±1.9 2.8±1.3 2.5±1.4
1=Very good 7=Very poor
General appearance after insertion General fit outside vagina General fit within vagina
3.5±1.9 3.0±1.7 2.8±1.7
3.7±1.9 3.5±1.9 3.4±1.9
1=Very little 7=Too much
Amount of lubricant Noise during use
4.1±1.2 2.8±1.4
4.4±1.3 3.2±1.7
1=Too long/wide 7=Too short/narrow
Length of condom outside vagina Length of condom within vagina Width of condom
3.3±1.1 3.6±1.1 3.5±1.2
3.3±1.1 3.6±1.1 3.4±1.1
1=Cool 7=Warm
Sensation of coolness
4.2±1.2
4.1±1.2
1=Very pleasant 7=Very unpleasant
Smell of condom
3.5±1.4
3.4±1.5
1=Much better 7=Much worse
Sensitivity/stimulation during use compared with no use
4.6±1.9
4.8±1.7
1=Very satisfied 7=Very unsatisfied
Satisfaction with stability during use Satisfaction with feeling of sex
2.9±1.8 3.8±2.0
3.9±2.0 4.2±2.1
1=This condom much better 7=This condom much worse
In general, how did study condoms compare to male condoms used prior to study
3.9±1.9
4.7±2.0
1=More sexual satisfaction 7=Less sexual satisfaction
How did sexual satisfaction with study condoms compare to male condoms used prior to study
3.8±1.9
4.6±2.0
1=Very easy 7=Very difficult
1=Very comfortable 7=Very uncomfortable
from 24% to 15%. Reports of partial turning inside out are not included as failure modes in the new definitions because the condom, while less stable, continues to serve as an intact barrier. This instability, however, can lead to user dissatisfaction which might result in discontinuation of the method. The WC was found to be more stable during use as evidenced by reports of slipping out completely less often than the FC1. The outer ring displacement rate of 10% observed in this study for both condom types is higher than that observed for the FC1 and FC2 (3.1% and 3.0%, respectively) in a previous study in which slippage was also rare [15]. Inadequate lubricant was felt to play a role in poor performance in condom failures of each type. Couples used lubrication for more of the WC uses than for FC1 most likely because the WC comes unlubricated and the user instructions recommend adding lubricant to the inside of the condom pouch before sex. A more convenient lubrication approach and/or more clear instructions about applying lubricant might improve acceptability and performance.
Overall, the WC leads to fewer urogenital symptoms and a trend toward fewer colposcopic signs of genital irritation among women. Although there were baseline differences at the Houston site, the trend toward less genital irritation with the WC condom was observed at all three centers. There did not appear to be any adverse immediate effects of the foam shapes and insertion capsule on the cervicovaginal epithelium evidenced by fewer abnormal colposcopic findings within 6 h after intercourse with the WC compared to the FC1 (26% vs. 44%). There were more colposcopic findings of erythema and external genital findings observed after FC1 compared to WC use. Among participants with a preference, WC was preferred over the FC1 by twice as many males and by 2.6 times as many females. The WC received more favorable or as favorable mean scores in most individual parameters except for convenience of lubrication. Women rated the WC as easier to handle and insert and were more satisfied with the stability of fit which are two significant problems that have
472
J.L. Schwartz et al. / Contraception 78 (2008) 465–473
been identified with the FC1. Feedback suggests that future user instructions for the WC should include more explicit instructions on how deep to place the device and cues for correct placement. Although the WC performed better than the FC1, women's experience with the WC in this study was not as universally positive as was reported in a previous single arm short-term acceptability study of the WC in three countries [11]. In the current study, experience improved condom functionality in that condom failure rates for all types of failure were higher during the first condom use period compared to the second condom use period, regardless of condom type. A study using prostate-specific antigen as a marker for female condom failure found that mechanical problems and semen exposure rates declined with increased use, also suggesting a positive effect of experience [16]. In short-term use, the WC has fewer reports of common user acceptability problems and was more often preferred over the FC1. Given these improvements, further development of the PATH Woman's Condom is warranted as a promising method of female initiated barrier protection.
•
•
•
Acknowledgments The authors thank Sarah Maitz for preparing the manuscript. Support for this study was provided by CONRAD through USAID. The opinions expressed herein do not necessarily reflect the opinions of USAID.
•
Appendix A. Functional performance definitions • Clinical breakage is defined as female condom breakage during intercourse or withdrawal of the condom from the vagina and is calculated by dividing the number of female condoms reported to have broken during intercourse or withdrawal by the number of female condoms used during intercourse. Clinical breakage includes instances where the inner or outer ring breaks off. • Nonclinical breakage is defined as female condom breakage before intercourse or after withdrawal of the condom from the vagina. It is calculated by dividing the number of female condoms reported to have broken before intercourse or after withdrawal by the number of female condom packages opened. • Total breakage is defined as female condom breakage at any time before, during or after intercourse (both clinical and nonclinical), and is calculated by dividing the number of female condoms that broke by the number of female condom packages opened. • Incorrect penetration is defined as vaginal penetration whereby the penis is inserted between the female condom and the vaginal wall and is calculated by dividing the number of reported incidences of incorrect
•
•
•
penetration by the number of female condoms used during intercourse. Pullout is defined as a female condom that is completely removed from the vagina during intercourse or withdrawal without turning inside out (no eversion) by adhering to the male partner's penis and acting as a male condom and is calculated by dividing the number of female condoms that pulled out completely during intercourse or withdrawal by the number of female condoms used during intercourse. Complete turning inside out is defined as a female condom that is completely removed from the vagina during intercourse or withdrawal by adhering to the male partner's penis and turning completely inside out (eversion) during intercourse or withdrawal and is calculated by dividing the number of female condoms that turned completely inside out during intercourse or withdrawal by the number of female condoms used during intercourse. Partial turning inside out is defined as a female condom that is partially removed from the vagina by adhering to the male partner's penis and partially turning inside out (eversion) during intercourse or withdrawal and is calculated by dividing the number of female condoms that turned partially inside out during intercourse or withdrawal by the number of female condoms used during intercourse. Complete slippage is defined as a female condom that slips completely out of the vagina, without turning inside out and without acting as a male condom, during intercourse or withdrawal and is calculated by dividing the number of female condoms that slipped completely out during intercourse or withdrawal by the number of female condoms used during intercourse. Partial slippage is defined as a female condom that slips partially out of the vagina, without turning inside out and without acting as a male condom, during intercourse or withdrawal and is calculated by dividing the number of female condoms that slipped partially out during intercourse or withdrawal by the number of female condoms used during intercourse. Outer ring displacement is defined as an outer ring/ stiffener that is pushed into the vagina during intercourse and is calculated by dividing the number of reported incidences of outer ring displacement during intercourse by the number of female condoms used during intercourse. Total clinical failure is defined as the number of female condoms that clinically break, completely or partially turn inside out, completely slip out during intercourse, or are associated with reported incidences of incorrect penetration, or outer ring displacement during intercourse (i.e., any event that might lead to semen spillage) and is calculated by dividing the sum of these events by the number of female condoms used during intercourse.
J.L. Schwartz et al. / Contraception 78 (2008) 465–473
• Total condom failure is defined as total breakage, incorrect penetration, total (complete and partial) turning inside out, total (complete and partial) slippage, outer ring displacement, and pullout and is calculated by dividing the sum of these events by the number of female condoms opened. References [1] Farr G, Gabelinick H, Sturgen K, Dorflinger L. Contraceptive efficacy and acceptability of the female condom. Am J Public Health 1994;84: 1960–4. [2] Soper DE, Brockwell NJ, Dalton HP. Evaluation of the effects of a female condom on the female lower genital tract. Contraception 1991; 44:21–9. [3] Soper DE, Shoupe D, Shangold GA, Shangold MM, Gutmann J, Mercier L. Prevention of vaginal trichomoniasis by compliant use of the female condom. Sex Transm Dis 1993;20:137–9. [4] Minnis AM, Padian NS. Effectiveness of female controlled barrier methods in preventing sexually transmitted infections and HIV: current evidence and future research directions. Sex Transm Inf 2005;81: 193–200. [5] Smit J, Beksinska M, Vijayakumar G, Mabude Z. Short-term acceptability of the Reality® polyurethane female condom and a synthetic latex prototype: a randomized crossover trial among South African women. Contraception 2006;73:394–8. [6] Jivasak-Apimas S, Saba J, Chandeying V, et al. Acceptability of the female condom among sex workers in Thailand: results from a prospective study. Sex Transm Dis 2001;28:648–54.
473
[7] Hoffman S, Mantell J, Exner TM, Stein Z. The future of the female condom. Perspect Sex Reprod Health 2004;36:120–6. [8] Deniaud F. Dynamics of female condom acceptability among prostitutes and young women in Abidjan, Ivory Coast. Contracept Fertil Sex 1997;25:921–32. [9] el-Bassel N, Krishnan SP, Schilling RF, Witte S, Gilbert L. Acceptability of the female condom among STD clinic patients. AIDS Educ Prev 1998;10:465–80. [10] Bounds W, Guillebaud J, Newman GB. Female condom (Femidom): a clinical study of its use-effectiveness and patient acceptability. Br J Fam Plann 1992;18:36–41. [11] Coffey PS, Kilbourne-Brooke M, Au G, Seamans Y, Cohen J. Shortterm acceptability of the PATH Woman's Condom among couples at three sites. Contraception 2006;73:588–93. [12] World Health Organization, CONRAD. Manual for the standardization of colposcopy for the evaluation of vaginally administered products update 2004. Geneva, Switzerland/Arlington, VA: WHO/CONRAD; 2004. [13] Steiner M, Trussell J, Glover L, Joanis C, Spruyt A, Dorflinger L. Standardized protocols for condom breakage and slippage trials: a proposal. Am J Public Health 1994;84:1897–900. [14] Beksinska M, Joanis C, Manning J, et al. A commentary: standardized definitions of failure modes for female condoms. Contraception 2006; 75:251–5. [15] Beksinska M, Smit J, Mabude Z, Vijayakumar G, Joanis C. Performance of the Reality polyurethane female condom and a synthetic latex prototype: a randomized crossover trial among South African women. Contraception 2006;73:386–93. [16] Macaluso M, Lawson L, Hortin G, et al. Efficacy of the female condom as a barrier to semen during intercourse. Am J Epidemiol 2003;157: 289–97.
Original research article
Comparative crossover study of the PATH Woman's Condom and the FC Female Condom® Jill L. Schwartza,⁎, Kurt Barnhartb , Mitch D. Creininc , Alfred Poindexterd , Angie Wheelesse , Maggie Kilbourne-Brookf , Christine K. Maucka , Debra H. Weinere , Marianne M. Callahana a Department of Obstetrics and Gynecology, CONRAD, Eastern Virginia Medical School, Arlington, VA 22209, USA Department of Obstetrics and Gynecology and Center for Clinical Epidemiology and Biostatistics, University of Pennsylvania, Philadelphia, PA 19104, USA c Department of Obstetrics, Gynecology and Reproductive Sciences, University of Pittsburgh and Magee-Women's Research Institute, Pittsburgh, PA 15213, USA d Advances in Health, Inc., Houston, TX 77030, USA e Family Health International (FHI), Research Triangle Park, NC 27713, USA f Program for Appropriate Technology in Health (PATH), Seattle, WA 98107, USA Received 7 May 2008; revised 18 July 2008; accepted 18 July 2008 b
Abstract Background: Only one female condom [FC1 Female Condom (FC1)] is currently marketed, but it is poorly utilized, perhaps due to difficulty with insertion, discomfort and suboptimal functional performance during intercourse. The Program for Appropriate Technology in Health (PATH) Woman's Condom (WC) was developed in an effort to overcome these obstacles. Study Design: This was a randomized crossover study to evaluate the functional performance, safety and acceptability of the FC1 and WC. Seventy-five couples were assigned to one of two condom use sequences (WC/FC1 or FC1/WC) at three centers. Four condoms of the first type were used by couples in four acts of intercourse at home over a 2–4-week period. After a follow-up visit, these procedures were repeated with the second assigned condom type. In a substudy of participants (n=25), a colposcopy was performed prior and subsequent to the first condom use of each of the two condom types. Condom performance was evaluated by calculating measures of function from questionnaires completed by the couple after each condom use. Safety was evaluated by reported urogenital symptoms with a given condom during or immediately following condom use and colposcopic signs of genital irritation in the substudy. Acceptability of each given condom type was measured by questionnaire. Results: Total condom failure (slippage, breakage, etc., divided by the number of female condoms opened) was 31% for the WC and 42% for the FC1. Total clinical failure (slippage, breakage, etc., divided by the number of female condoms used) was 17% for the WC and 24% for the FC1. The proportion of condom failures was 10.9 percentage points less, and the proportion of clinical failure 6.7 percentage points less, when couples used the WC compared to the FC1 [90% CI: −18.5 to −3.3 and −12.6 to −0.8, respectively). Fewer women reported symptoms of urogenital irritation when using the WC vs. the FC1 either overall or when analyzing each use of the condom [woman as unit: −20 percentage points (90% CI: −30.5 to −9.3); condom use as unit: −12.3 percentage points (90% CI: −18.0 to −6.7)]. A similar result was seen for signs of urogenital irritation [woman as unit: −20 percentage points (90% CI: −42.7 to 4.8)]. Among participants with a preference, WC was preferred over the FC1 by twice as many males and by 2.6 times as many females. Conclusions: While both female condoms were safe and acceptable in short-term use, the PATH Woman's Condom leads to less failure, was associated with fewer adverse events, and was more acceptable than the FC1 Female Condom. © 2008 Elsevier Inc. All rights reserved. Keywords: Female condom; PATH Woman's Condom; Reality®, FC Female Condom®; Slippage; Breakage
1. Introduction
⁎ Corresponding author. Tel.: +1 703 524 4744; fax: +1 703 524 4770. E-mail address: [email protected] (J.L. Schwartz). 0010-7824/$ – see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.contraception.2008.07.020
The Reality® Female Condom (currently called FC Female Condom®) (FC1; Female Health Company, Chicago, IL, USA) was approved by the US Food and Drug
466
J.L. Schwartz et al. / Contraception 78 (2008) 465–473
Administration (FDA) in 1993 as the first barrier contraceptive for women that offered limited protection against sexually transmitted infections (STIs). This female condom has a 6-month typical use pregnancy rate of 12.4 per 100 woman-years in a US population [1], has not been associated with trauma to the genital tract [2], prevents trichomoniasis recurrence [3], and may offer as much protection from STIs as male condoms [4]. The FC2 has been introduced as an updated version of this condom using synthetic latex for greater affordability [5]. Despite global need for a female controlled barrier device, there has been poor sustained utilization of this method and the potential impact of the device, as an HIV prevention strategy has not been fully evaluated. In female condom studies conducted over the past decade, varying proportions of users have liked the device and succeeded in using it, resulting in differing acceptability ratings in short-term acceptability studies, depending on the study population and the type of intervention [6–10]. These studies have identified a constellation of problems with female condom use including difficulty with insertion, discomfort and suboptimal functional performance during intercourse that may contribute to poor utilization. The Program for Appropriate Technology in Health (PATH), a nonprofit, international health organization, has used an iterative user-driven development process to produce a novel female condom product in an attempt to overcome these obstacles. The PATH Woman's Condom (WC) has a thin, pliable plastic pouch that conforms to the shape of the vagina, is 9 in. long and has a flexible soft outer ring that is intended to protect the external genitalia (Fig. 1). The four oblong foam pieces on the outside of the pouch are intended to cling lightly to vaginal walls, ensuring stability of the device. The distal end of the pouch and foam pieces are packaged into a capsule which serves as an insertion aid and dissolves quickly when inserted into the vagina. The PATH Woman's Condom was
easy to use and comfortable in a short-term acceptability study [11]. This study was designed to further test the functional performance, safety and acceptability of this novel female condom and to gain additional information for further development. This crossover study compared the PATH Woman's Condom (WC) to the FC1 after four uses of each condom type by 75 couples.
2. Materials and methods Seventy-five couples were enrolled between December 2004 and August 2005 in this Phase I, comparative, openlabel, two-period crossover study comparing the WC to the FC1 at three clinical centers: University of Pennsylvania Medical Center; University of Pittsburgh/Magee Womens Research Institute; and Advances in Health, Houston, TX, USA. The sample size was not selected on the basis of hypothesis testing. The crossover design was chosen because between-patient variation can be eliminated, and smaller groups can be used. Couples who agreed to participate were in good health, at least 18 years old, and at low risk for HIV or STIs. Those couples at risk for pregnancy used a steroidal contraceptive (oral, patch, injectable or implantable) or sterilization (vasectomy or tubal ligation) throughout their participation. Individuals who had sensitivities or allergies to study products or prior experience with female condoms were excluded. The protocol and informed consents for this nonsignificant risk device study were reviewed and approved by the site institutional review boards. Informed consent was obtained from each member of the couple. The female volunteer provided a urine specimen for a pregnancy test and height and weight were measured. Volunteers in the colposcopy substudy at the University of Pennsylvania Medical Center (n=25) also underwent a pelvic
Fig. 1. PATH Woman's Condom.
J.L. Schwartz et al. / Contraception 78 (2008) 465–473
examination with Pap smear, if not performed in the past year; wet mount and Neisseria gonorrhoeae and Chlamydia trachomatis polymerase chain reaction testing. Enrolled couples were randomized to one of two condom use sequences. Participants were given either a WC research study packet or a FC1 research study packet as determined by the assigned condom use sequence for the first condomuse period to use in four separate acts of intercourse over a 2–4-week period. The PATH Woman's Condom (WC) (PATH, Seattle, WA, USA) is an investigational device that is not prelubricated. The FC1, a 6.5-inch long polyurethane plastic pouch with two flexible rings, was used as the control female condom because it is the only FDA-approved female condom. Participants were instructed to wait 30–60 s after inserting the WC to have intercourse to allow for the insertion capsule to dissolve. A total of four condoms, each along with a 3-mL package of Astroglide® (BioFilm, Vista, CA, USA), and instructions for use were distributed. The couples also went home with a detailed condom use questionnaire to complete after each intercourse, and an acceptability questionnaire to be completed separately by each person at the end of each condom use period. Female participants had a follow-up visit after completing the four acts of intercourse, at which time questionnaires and diaries were reviewed and collected, and second assigned condoms and questionnaires and diaries were distributed. Female participants in the substudy underwent a baseline colposcopy prior to the first condom use of each of the two condom types. These women were abstinent from using vaginal products and all sexual activity starting 72 h prior to the baseline colposcopy until after intercourse with the study condom. The couples were to use the condom for the first act of intercourse within 72 h of the baseline colposcopy and the female participant was to return within 6 h after intercourse. After intercourse, colposcopy was performed to examine the immediate effects on the cervicovaginal epithelium of the foam shapes and insertion capsule components of the WC and to compare this to the effect of the inner ring of the marketed FC1, according to the CONRAD/World Health Organization Manual [12]. This was a single-blinded study with only the clinicians performing the colposcopies unaware of the participant's condom assignment. Due to the obvious difference in condom design, the participants could not be blinded to the condom type being used. The random permuted blocks method was used to generate random condom-use sequences for 25 participants per site. The sequences were created by a randomization manager using a verified program based on the random function RANUNI in the SAS® System (SAS Institute, Cary, NC, USA). Allocations were sealed in separate individually sealed envelopes identified by participant number. The condom performance outcome was measured by calculating the functional measures (breakage, slippage, etc.) described in Appendix A from questionnaires completed by the couple after each condom use. The
467
primary safety outcomes were the proportion of participants with urogenital symptoms and the proportion of condom uses during or immediately after which urogenital symptoms were noted. These symptoms included urogenital pain and discomfort (e.g., burning, stinging or pressure), dysuria, genital irritation or rash, genital pruritus, abnormal vaginal discharge, pelvic or lower abdominal pain or intermenstrual bleeding for the female and urogenital pain or discomfort and genital irritation or rash for the male participants. The substudy safety outcome measured the proportion of female participants with signs of genital irritation of the external genitalia, cervix and vagina as observed on colposcopic examination after one act of intercourse with a given condom type. Adverse events (AEs) were coded in the MedDRA dictionary based on blind review of the AE description. Acceptability was measured for each given condom type by questionnaire. All statistical analyses were descriptive or involved calculation of 90% confidence intervals (CIs) about the event proportions. Generalized estimating equations were used to model the differences in proportions of condom failure when couples used the WC versus the FC1, controlling for center and taking into account clustering of condom use within couple. In addition, differences in proportions between the two types of condoms, together with 90% CIs, were calculated for signs and symptoms of urogenital irritation. The 90% CI for the difference in proportions between treatment represents the range of differences in treatment that cannot be rejected (by a twosided test, alpha=.10) given the observed data.
3. Results Of the 82 couples screened for eligibility, 75 were enrolled and 73 completed the study (Fig. 2). The functional performance definitions are detailed in Appendix A. Participant characteristics are described in Table 1. Participants at the Houston site differed from those at the other two sites in that they were likely to be older (female mean=34.3, male mean=35.5), identified themselves as Hispanic or black more frequently, and were less educated (female mean=13.2 years, male mean=12.6). They were also more likely to have had a previous female sterilization or hysterectomy, weigh more (female mean=172.9), and have previously been pregnant. Nine enrolled couples did not open and use all 4 of either or both condom types. Five couples used all four WC but only used one to three FC1 because of discomfort. Three couples used three to four FC1 and zero to two WC. One couple used only one WC after which the female participant developed bilateral labial swelling and erythema and was discontinued early from the study. Condom use questionnaires were completed for 287 WC uses and for 285 FC1 uses. Of those, four WC and seven FC1 broke while opening the packages, four
468
J.L. Schwartz et al. / Contraception 78 (2008) 465–473
FC1, respectively) or complete slippages (6% and 8%, for WC and FC1, respectively). The most common failure type contributing to clinical failure was outer ring displacement which was 9.9% for both condom types (Table 2). Approximately 21 couples (29%) experienced at least one outer ring displacement during WC use compared to 17 couples (23%) during FC1 use. Nine couples reported outer ring displacement with at least one use of both products. Couples used lubrication for the WC in 91% of condom uses and for the FC1 in 61% of condom uses (the FC1 is prelubricated). When asked about the relationship of condom failure to lubricant, 24 (41%) out of 59 respondents who reported a failure (turned inside out, slipped out and/or pulled out) with the WC felt that it was
Table 1 Participant characteristics Female (n=75) Fig. 2. Flow diagram of study.
WC and five FC1 were not inserted, and no vaginal intercourse occurred for five WC uses and one FC1 use. Therefore, a total of 274 WC and 272 FC1 were used during intercourse. Most condoms were inserted by the female participant (80%) and in the vagina as per instructions (97%). The study condoms were worn during vaginal intercourse for an average of 16 min with a range of 1–93 min. A greater percentage of FC1 uses compared to WC uses were associated with difficulty handling the device before insertion (44% compared to 9%), difficulty placing the device in the vagina (36% compared to 8%), discomfort during insertion (21% compared to 8%) and device instability after insertion (16% compared to 9%). About one third of uses of each device were associated with uncertainty as to how deep to place the device (FC1=26.0%, WC=28.1%) and whether the device was correctly placed (FC1=29.7%, WC=26.9%). Total condom failure (Appendix A) was 31% for the WC and 42% for the FC1 and total clinical failure (Appendix A) was 17% for the WC and 24% for the FC1 (Table 2). The proportion of condom failures was 10.9 percentage points less, and the proportion of clinical failure 6.7 percentage points less, when couples used the WC compared to the FC1 (90% CI: −18.5 to −3.3 and −12.6 to −0.8, respectively) (Table 2). WC leads to significantly less failure than FC1 with regard to partial turning inside out and partial slippage and there is a trend toward significance with complete slippage. Approximately 44% of the couples during WC use and 53% of couples during FC1 use experienced at least one event that contributed to total clinical failure. A minority of couples experienced clinical breaks (4% and 0%, for WC and
n Age (years) Mean±S.D. 29.2±8.3 Range 19–60 Ethnicity Hispanic 13 Not Hispanic 62 Race Asian 3 Black or African American 15 White 55 More than one race 2 Other 0 Education Mean±S.D. 14.9±2.5 Ever used Oral contraceptives 70 Male condom 73 Female condom 0 Diaphragm 3 Primary method Tubal ligation 9 Oral contraceptives 46 Hormonal injection 8 Hormonal patch 6 6 Othera Height (in.) Mean±S.D. 64.2±2.5 Range 59–72 Weight Mean±S.D. 153.1±33.8 Range 110–260 Ever pregnant Yes 34 Current relationship with male partner Living together 41 Length of time with male partner (years) Mean±S.D. 5.5±6.8 Range 0.3–40
Male (n=75) %
n
%
30.7±8.5 19–58 17 83
13 62
17 83
4 20 73 3 0
1 15 56 2 1
1 20 75 3 1
14.9±2.7 93 97 0 4 12 61 11 8 8
45 55
a Other methods used while in study include: hysterectomy, post menopausal, and vaginal ring.
J.L. Schwartz et al. / Contraception 78 (2008) 465–473
469
Table 2 Condom functionality PATH WC (n=287)
a
Total number of condoms opened Broke while opening Broke during insertion Broke other (capsule off or missing) Total number of condoms usedb Broke during intercourse Don't know when Type of failurec Clinical breakage Nonclinical breakage Total breakage Incorrect penetration Pullout Complete turning inside out Partial turning inside out Complete slippage Partial slippage Outer ring displacement Total clinical failured Total condom failured
n
%
90% CI
287 4 0 2 274 2 1
100.0 1.4 0.0 0.7 95.5 0.7 0.4
3 6 9 6 37 2 12 4 29 27 47 89
1.1 2.1 3.1 2.2 13.5 0.7 4.4 1.5 10.6 9.9 17.2 31.0
0.3–2.8 0.9–4.1 1.7–5.4 1.0–4.3 10.2–17.4 0.1–2.3 2.6–7.0 0.5–3.3 7.7–14.2 7.1–13.3 13.5–21.3 26.5–35.8
FC1 Female Condom (n=285)
WC-FC 1
n
%
90% CI
Difference
90% CI
285 7 2 0 272 0 0
100.0 2.5 0.7 0.0 95.4 0.0 0.0
0 9 9 7 47 1 26 10 42 27 65 120
0.0 3.2 3.2 2.6 17.3 0.4 9.6 3.7 15.4 9.9 23.9 42.1
0.0–1.1 1.7–5.5 1.7–5.5 1.2–4.8 13.6–21.5 0.0–1.7 6.8–13.0 2.0–6.2 12.0–19.5 7.1–13.4 19.7–28.5 37.2–47.1
– −1.1 0.1 −0.2 −1.9 0.4 −4.6 −2.1 −5.1 −0.4 −6.7 −10.9
−3.0 to 0.8 −2.1 to 2.3 −2.0 to 1.7 −6.7 to 2.9 −0.7 to 1.3 −8.3 to −0.8⁎ −4.2 to 0.0 −9.6 to −0.7⁎ −4.5 to 3.8 −12.6 to −0.8⁎ −18.5 to −3.3⁎
a
Denominator for nonclinical breakage, total breakage and total failure. Denominator for clinical breakage, incorrect penetration, pullout, complete turning inside out, partial turning inside out, complete slippage, partial slippage, outer ring displacement and total clinical failure. c A condom may be counted for more than one type of failure. d A condom only contributed to total failure or clinical failure once, even if it had more than one type of failure. ⁎ Since the upper bounds of these CIs do not include 0, there is a difference (p≤.10, two-tailed) between condom types. b
related to lubrication. Likewise, 40 (46%) of 87 respondents who reported a failure with the FC1 condom felt that it was related to lubrication. The most common reasons given for the failure of both condom types were that there was not enough lubricant and that the lubricant was dried out or sticky. Men ejaculated during 228 of the 274 WC uses (83%) and during 205 of the 272 FC1 uses (75%). Most of the study condoms had to be pulled gently for removal; however, over a third were pulled straight out without twisting to remove which may account for the semen spillage reported in about 7% of the WC and FC uses.
Condom failure rates for all types of failure were higher during the first condom use period compared to the second condom use period, regardless of condom type. Types of urogenital symptoms are listed in Table 3. Fewer women reported symptoms of urogenital irritation when using the WC versus the FC1 either overall (e.g., taking the woman as the unit of analysis) or when analyzing each use of the condom [woman as unit: −20 percentage points (90% CI: −30.5 and −9.3); condom use as unit: −12.3 percentage points (90% CI: −18.0 and −6.7)]. Five women had at least one AE with both condom types. One participant had bilateral labial swelling and
Table 3 Urogenital symptoms Adverse event
Genital pain Genital discomfort Genital rash Pelvic pain Labial swelling Intermenstrual bleeding Genital pruritus Vaginal candidiasis Dysuria Total number of events/women with at least one eventa
PATH Woman's Condom n=73
FC1 Female Condom n=74
Number of events
Number of women (%)
Number of events
Number of women (%)
8 7 2 0 1 0 0 1 0 19
6 (8%) 4 (5%) 1 (1%) 0 1 (1%) 0 0 1 (1%) 0 9 (12%)
30 25 0 6 0 1 1 1 3 67
13 (18%) 10 (14%) 0 4 (5%) 0 1 (1%) 1 (1%) 1 (1%) 3 (4%) 24 (32%)
a WC-FC1 difference (90% CI) for condom as a unit: −12.3 (−18.0 to −6.7) and for women as a unit: −20.0 (−30.5 to −9.3). Since the upper bounds of these CIs do not include 0 there is a difference (p≤.10, two-tailed) between condom types.
470
J.L. Schwartz et al. / Contraception 78 (2008) 465–473
Table 4 Colposcopic findings Finding
PATH Woman's Condom Baseline PATH Woman's Condom FC1 Female Condom Baseline (before intercourse) (before intercourse) New findingsa
FC1 Female Condom New findingsa
n=23 Erythema 14 Petechiae 2 Ecchymoses 0 Edema 4 Ulcer 1 Peeling 0 Grossly white finding 0 Other 8 Total number of findings 29 Total number of women with at 14 least one findingd
n=23
n=25
n=25
1 2 0 0 0 1 1 2b 7 6 (26%)
10 3 0 0 0 1 1 8 23 14
13 1 0 1 0 0 0 2c 17 11 (44%)
a New findings include all new, noniatrogenic colposcopic findings that were not observed on the baseline colposcopy or findings observed on the baseline colposcopy that worsened during intercourse with the given condom, regardless of relatedness as determined by the site study clinician. b Nabothian cyst and prominent vessel. c Prominent vessels and rash-like erythema. d WC-FC1 difference for new findings (90% CI) for women as a unit=−20.0 (42.7 to 4.8). Since the upper bounds of this CI include 0 there is no difference (p≤.10, two-tailed) between condom types.
erythema after use of her first WC. Among male participants, there were a total of 14 AEs associated with product use reported by 6 (8%) out of 73 men using at least one WC and a total of 13 AEs associated with product use reported in 8 (11%) out of 74 using at least one FC1. Mild penile discomfort, pressure, irritation and itching were the most common male complaints. Similar results were observed for male symptoms when condom use was the unit of analysis. Overall, men experienced slightly fewer symptoms of urogenital irritation when using the WC versus the FC1 (man as unit: −2.7 percentage points, 90% CI: −10.7 to 5.0; condom use as unit: −0.4, 90% CI: −4.2 to 3.4). No severe or serious AEs were reported by male or female participants. Types of colposcopic findings are listed in Table 4. Fewer women had colposcopic findings of urogenital irritation when using the WC versus the FC1 (woman as unit: −20 percentage points, 90% CI: −42.7 to 4.8). All of the seven findings observed with the WC were located on the cervix. Thirteen of the 17 findings observed with the FC1 were erythema. Eleven findings were located on the external genitalia, five were located on the cervix, and one was located on the vaginal wall. None of the findings involved deep disruption of the epithelium and none were considered serious. After using both condom types, 41 (57%) out of 72 of the female partners liked the WC more, 16 (22%) out of 72 liked the FC1 more and 15 (21%) out of 72 had no preference. After using both condom types, 31 (43%) out of 72 of the male partners liked the WC more, 16 (22%) out of 72 liked the FC1 more and 25 (35%) out of 72 had no preference. Female participants gave more favorable mean scores to the WC for the majority of acceptability parameters (Table 5). On average, female participants liked the WC minimally better and the FC1 slightly worse than male condoms used
prior to the study (Table 5). Female participants gave more favorable mean scores (1=very easy to 7=very difficult) on only one aspect of the FC1; convenience of lubrication (WC −2.6, FC1 −1.9) (Table 5). Male partners gave more favorable or similar means scores to the WC in all categories except for convenience of lubrication.
4. Discussion This was the first breakage and slippage study of the PATH Woman's Condom. The WC performed better than the FC1 with lower total clinical failure and total failure using female condom performance definitions adapted from male condom definitions of slippage and breakage [13]. There was significantly less failure with WC than with FC1 in regard to partially turning inside out and partial slippage and a trend toward significance with complete slippage. The statistical significance of these results should be interpreted cautiously, as the small sample size and relatively few condom uses per couple meant that “nuisance” variables for period effects (a difference in results between the condom type used during the first crossover period vs. the condom type used in the second) and carryover (an effect of the first condom type lingering past the washout period and affecting results with the second condom type) could not be tested with sensitivity. WC production was still at the hand-made prototype stage when inventory was developed for this study, so a few clinical breaks probably resulting from the hand welding was not a surprising result. Using a recently revised definition of failure modes specific for female condoms [14], which does not include partial turning inside out, total clinical failure rate for WC is reduced from 17% to 13% and that for FC1 is reduced
J.L. Schwartz et al. / Contraception 78 (2008) 465–473
471
Table 5 Female partner acceptability PATH Woman's Condom (n=73) Mean±S.D.
FC1 Female Condom (n=73) Mean±S.D.
Ease of handling condom Ease of inserting condom Ease of following instructions Convenience of lubrication Ease of removing condom
2.1±1.4 2.3±1.3 1.5±.82 2.6±2.0 1.7±1.1
3.7±2.0 3.8±2.0 2.0±1.4 1.9±1.3 1.9±1.3
Comfort during insertion of condom Feel of lubricant during insertion Feel of material Overall comfort during use Comfort/fit of outer ring during use Feel of lubricant during use of condom Comfort during removal of condom
2.6±1.3 2.4±1.5 3.0±1.7 3.2±1.6 3.5±1.7 2.5±1.5 2.0±1.2
3.5±1.8 2.4±1.2 3.5±1.7 4.2±1.9 3.8±1.9 2.8±1.3 2.5±1.4
1=Very good 7=Very poor
General appearance after insertion General fit outside vagina General fit within vagina
3.5±1.9 3.0±1.7 2.8±1.7
3.7±1.9 3.5±1.9 3.4±1.9
1=Very little 7=Too much
Amount of lubricant Noise during use
4.1±1.2 2.8±1.4
4.4±1.3 3.2±1.7
1=Too long/wide 7=Too short/narrow
Length of condom outside vagina Length of condom within vagina Width of condom
3.3±1.1 3.6±1.1 3.5±1.2
3.3±1.1 3.6±1.1 3.4±1.1
1=Cool 7=Warm
Sensation of coolness
4.2±1.2
4.1±1.2
1=Very pleasant 7=Very unpleasant
Smell of condom
3.5±1.4
3.4±1.5
1=Much better 7=Much worse
Sensitivity/stimulation during use compared with no use
4.6±1.9
4.8±1.7
1=Very satisfied 7=Very unsatisfied
Satisfaction with stability during use Satisfaction with feeling of sex
2.9±1.8 3.8±2.0
3.9±2.0 4.2±2.1
1=This condom much better 7=This condom much worse
In general, how did study condoms compare to male condoms used prior to study
3.9±1.9
4.7±2.0
1=More sexual satisfaction 7=Less sexual satisfaction
How did sexual satisfaction with study condoms compare to male condoms used prior to study
3.8±1.9
4.6±2.0
1=Very easy 7=Very difficult
1=Very comfortable 7=Very uncomfortable
from 24% to 15%. Reports of partial turning inside out are not included as failure modes in the new definitions because the condom, while less stable, continues to serve as an intact barrier. This instability, however, can lead to user dissatisfaction which might result in discontinuation of the method. The WC was found to be more stable during use as evidenced by reports of slipping out completely less often than the FC1. The outer ring displacement rate of 10% observed in this study for both condom types is higher than that observed for the FC1 and FC2 (3.1% and 3.0%, respectively) in a previous study in which slippage was also rare [15]. Inadequate lubricant was felt to play a role in poor performance in condom failures of each type. Couples used lubrication for more of the WC uses than for FC1 most likely because the WC comes unlubricated and the user instructions recommend adding lubricant to the inside of the condom pouch before sex. A more convenient lubrication approach and/or more clear instructions about applying lubricant might improve acceptability and performance.
Overall, the WC leads to fewer urogenital symptoms and a trend toward fewer colposcopic signs of genital irritation among women. Although there were baseline differences at the Houston site, the trend toward less genital irritation with the WC condom was observed at all three centers. There did not appear to be any adverse immediate effects of the foam shapes and insertion capsule on the cervicovaginal epithelium evidenced by fewer abnormal colposcopic findings within 6 h after intercourse with the WC compared to the FC1 (26% vs. 44%). There were more colposcopic findings of erythema and external genital findings observed after FC1 compared to WC use. Among participants with a preference, WC was preferred over the FC1 by twice as many males and by 2.6 times as many females. The WC received more favorable or as favorable mean scores in most individual parameters except for convenience of lubrication. Women rated the WC as easier to handle and insert and were more satisfied with the stability of fit which are two significant problems that have
472
J.L. Schwartz et al. / Contraception 78 (2008) 465–473
been identified with the FC1. Feedback suggests that future user instructions for the WC should include more explicit instructions on how deep to place the device and cues for correct placement. Although the WC performed better than the FC1, women's experience with the WC in this study was not as universally positive as was reported in a previous single arm short-term acceptability study of the WC in three countries [11]. In the current study, experience improved condom functionality in that condom failure rates for all types of failure were higher during the first condom use period compared to the second condom use period, regardless of condom type. A study using prostate-specific antigen as a marker for female condom failure found that mechanical problems and semen exposure rates declined with increased use, also suggesting a positive effect of experience [16]. In short-term use, the WC has fewer reports of common user acceptability problems and was more often preferred over the FC1. Given these improvements, further development of the PATH Woman's Condom is warranted as a promising method of female initiated barrier protection.
•
•
•
Acknowledgments The authors thank Sarah Maitz for preparing the manuscript. Support for this study was provided by CONRAD through USAID. The opinions expressed herein do not necessarily reflect the opinions of USAID.
•
Appendix A. Functional performance definitions • Clinical breakage is defined as female condom breakage during intercourse or withdrawal of the condom from the vagina and is calculated by dividing the number of female condoms reported to have broken during intercourse or withdrawal by the number of female condoms used during intercourse. Clinical breakage includes instances where the inner or outer ring breaks off. • Nonclinical breakage is defined as female condom breakage before intercourse or after withdrawal of the condom from the vagina. It is calculated by dividing the number of female condoms reported to have broken before intercourse or after withdrawal by the number of female condom packages opened. • Total breakage is defined as female condom breakage at any time before, during or after intercourse (both clinical and nonclinical), and is calculated by dividing the number of female condoms that broke by the number of female condom packages opened. • Incorrect penetration is defined as vaginal penetration whereby the penis is inserted between the female condom and the vaginal wall and is calculated by dividing the number of reported incidences of incorrect
•
•
•
penetration by the number of female condoms used during intercourse. Pullout is defined as a female condom that is completely removed from the vagina during intercourse or withdrawal without turning inside out (no eversion) by adhering to the male partner's penis and acting as a male condom and is calculated by dividing the number of female condoms that pulled out completely during intercourse or withdrawal by the number of female condoms used during intercourse. Complete turning inside out is defined as a female condom that is completely removed from the vagina during intercourse or withdrawal by adhering to the male partner's penis and turning completely inside out (eversion) during intercourse or withdrawal and is calculated by dividing the number of female condoms that turned completely inside out during intercourse or withdrawal by the number of female condoms used during intercourse. Partial turning inside out is defined as a female condom that is partially removed from the vagina by adhering to the male partner's penis and partially turning inside out (eversion) during intercourse or withdrawal and is calculated by dividing the number of female condoms that turned partially inside out during intercourse or withdrawal by the number of female condoms used during intercourse. Complete slippage is defined as a female condom that slips completely out of the vagina, without turning inside out and without acting as a male condom, during intercourse or withdrawal and is calculated by dividing the number of female condoms that slipped completely out during intercourse or withdrawal by the number of female condoms used during intercourse. Partial slippage is defined as a female condom that slips partially out of the vagina, without turning inside out and without acting as a male condom, during intercourse or withdrawal and is calculated by dividing the number of female condoms that slipped partially out during intercourse or withdrawal by the number of female condoms used during intercourse. Outer ring displacement is defined as an outer ring/ stiffener that is pushed into the vagina during intercourse and is calculated by dividing the number of reported incidences of outer ring displacement during intercourse by the number of female condoms used during intercourse. Total clinical failure is defined as the number of female condoms that clinically break, completely or partially turn inside out, completely slip out during intercourse, or are associated with reported incidences of incorrect penetration, or outer ring displacement during intercourse (i.e., any event that might lead to semen spillage) and is calculated by dividing the sum of these events by the number of female condoms used during intercourse.
J.L. Schwartz et al. / Contraception 78 (2008) 465–473
• Total condom failure is defined as total breakage, incorrect penetration, total (complete and partial) turning inside out, total (complete and partial) slippage, outer ring displacement, and pullout and is calculated by dividing the sum of these events by the number of female condoms opened. References [1] Farr G, Gabelinick H, Sturgen K, Dorflinger L. Contraceptive efficacy and acceptability of the female condom. Am J Public Health 1994;84: 1960–4. [2] Soper DE, Brockwell NJ, Dalton HP. Evaluation of the effects of a female condom on the female lower genital tract. Contraception 1991; 44:21–9. [3] Soper DE, Shoupe D, Shangold GA, Shangold MM, Gutmann J, Mercier L. Prevention of vaginal trichomoniasis by compliant use of the female condom. Sex Transm Dis 1993;20:137–9. [4] Minnis AM, Padian NS. Effectiveness of female controlled barrier methods in preventing sexually transmitted infections and HIV: current evidence and future research directions. Sex Transm Inf 2005;81: 193–200. [5] Smit J, Beksinska M, Vijayakumar G, Mabude Z. Short-term acceptability of the Reality® polyurethane female condom and a synthetic latex prototype: a randomized crossover trial among South African women. Contraception 2006;73:394–8. [6] Jivasak-Apimas S, Saba J, Chandeying V, et al. Acceptability of the female condom among sex workers in Thailand: results from a prospective study. Sex Transm Dis 2001;28:648–54.
473
[7] Hoffman S, Mantell J, Exner TM, Stein Z. The future of the female condom. Perspect Sex Reprod Health 2004;36:120–6. [8] Deniaud F. Dynamics of female condom acceptability among prostitutes and young women in Abidjan, Ivory Coast. Contracept Fertil Sex 1997;25:921–32. [9] el-Bassel N, Krishnan SP, Schilling RF, Witte S, Gilbert L. Acceptability of the female condom among STD clinic patients. AIDS Educ Prev 1998;10:465–80. [10] Bounds W, Guillebaud J, Newman GB. Female condom (Femidom): a clinical study of its use-effectiveness and patient acceptability. Br J Fam Plann 1992;18:36–41. [11] Coffey PS, Kilbourne-Brooke M, Au G, Seamans Y, Cohen J. Shortterm acceptability of the PATH Woman's Condom among couples at three sites. Contraception 2006;73:588–93. [12] World Health Organization, CONRAD. Manual for the standardization of colposcopy for the evaluation of vaginally administered products update 2004. Geneva, Switzerland/Arlington, VA: WHO/CONRAD; 2004. [13] Steiner M, Trussell J, Glover L, Joanis C, Spruyt A, Dorflinger L. Standardized protocols for condom breakage and slippage trials: a proposal. Am J Public Health 1994;84:1897–900. [14] Beksinska M, Joanis C, Manning J, et al. A commentary: standardized definitions of failure modes for female condoms. Contraception 2006; 75:251–5. [15] Beksinska M, Smit J, Mabude Z, Vijayakumar G, Joanis C. Performance of the Reality polyurethane female condom and a synthetic latex prototype: a randomized crossover trial among South African women. Contraception 2006;73:386–93. [16] Macaluso M, Lawson L, Hortin G, et al. Efficacy of the female condom as a barrier to semen during intercourse. Am J Epidemiol 2003;157: 289–97.