Contraception 92 (2015)

261–267

Original research article

A randomized noninferiority crossover controlled trial of the functional

performance and safety of new female condoms: an evaluation of the
☆,☆☆,★ Velvet,
Cupid2, and FC2
⁎
Mags Beksinskaa,b, , Ross Greenera, Immo Kleinschmidtb, Lavanya Pillaya,
Virginia Maphumuloa, Jennifer Smita,c
a
MatCH Research (Maternal, Adolescent and Child Health Research), Department of Obstetrics and Gynaecology, Faculty of Health Sciences, University of
the Witwatersrand, 34 Essex Terrace, Westville, Durban 3629, South Africa
b
Faculty of Epidemiology and Population Health, London School of Hygiene and Tropical Medicine, Keppel Street, London WC1E, UK cSchool
of Pharmacy and Pharmacology, Faculty of Health Sciences, University of KwaZulu-Natal, University Road, Chiltern Hills, Westville 3629, South Africa
Received 2 February 2015; revised 1 May 2015; accepted 14 May 2015

Abstract

Objectives: New designs of female condoms have been developed to lower cost and/or improve acceptability. To secure regulatory
approvals, clinical studies are required to verify performance. We aimed to assess the functional performance and safety of two new female
condom types — Velvet and Cupid2 female condom — against the existing FC2 female condom.
Study design: This was a three-period crossover, randomized noninferiority clinical trial with 300 women randomized to condom-type
order in one South African site. Primary end points were total clinical failure and total female condom failure. Noninferiority of component
modes, clinical breakage, nonclinical breakage, slippage, misdirection and invagination were also determined. Safety data were also
assessed for each female condom. Participants were asked to use five of each female condom type and to collect information on use in a
condom diary at home and were interviewed after use of each type. Frequencies and percentages were calculated by condom type for each
failure mode, and differences in performance of the three female condoms using FC2 as reference, with 95% confidence intervals, were
estimated using generalized estimating equation models.
Results: A total of 282 (94%) participants completed follow-up, using at least one condom of each type. Total clinical failure (clinical
breakage, invagination, misdirection, slippage) was b5% for all female condoms: FC2 (4.50%), Cupid2 (4.79%) and Velvet (3.93%).
Noninferiority was demonstrated for all condom failure modes for the two new female condoms with respect to FC2, within the margin of
3% difference in mean failure, at the 5% significance level.
Conclusion: Noninferiority for the two new female condoms was demonstrated with respect to the marketed FC2. These data are used to
support manufacturer dossiers for World Health Organization (WHO)/United Nations Population Fund (UNFPA) prequalification.
Implications: Data from this study have been submitted to WHO/UNFPA and will contribute to the prequalification submission
requirements for the Cupid2 and Velvet female condoms. © 2015 Elsevier Inc. All rights reserved.

Keywords: Female condom; Function; Barrier methods; Safety

☆
There are no conflicts of interest known to the authors.
☆☆
262 M. Beksinska et al. / Contraception 92 (2015) 261–267
This study wasfunded by theUniversal Access to Female
CondomsJoint Programme, The William and Flora Hewlett Foundation,
and Cupid Ltd.
★ programs, add value to the method mix and respond to the
This trial is registered on the South African Clinical Trials
database DOH-27-0113-4272.
⁎ Corresponding author at: MatCH Research (Maternal, Adolescent
and Child Health Research), Department of Obstetrics and Gynaecology,
Faculty of Health Sciences, University of the Witwatersrand, 34 Essex
Terrace, Westville, Durban 3629, South Africa.
E-mail address: mags.beksinska@lshtm.ac.uk (M. Beksinska).
http://dx.doi.org/10.1016/j.contraception.2015.05.008 0010-
7824/© 2015 Elsevier Inc. All rights reserved.
1. Introduction
Like male condoms, female condoms should provide a
mechanical barrier against unplanned pregnancy and most
sexually transmitted infections (STIs) including HIV [1].
Although female condom distribution continues to increase
globally, increasing from 25 million units in 2007 to over
60 million units in 2012 [2], distribution remains
significantly lower compared to male condoms, accounting
for only
Box 1
Description of study products
Feature FC2
reproductive health technologies having the potential to
expand choice in reproductive health and family planning
needs of a range of clients [4].
Several new female condom products, in the final stages
of development or recently available, aim to reduce unit
cost and/or improve acceptability [5]. The first female
condom (FC1) made by the Female Health Company
(FHC) was approved by the US Food and Drug
Administration (USFDA) in 1993. The FHC ceased
production of the polyurethane FC1 in 2009 and replaced it
with a similarly designed synthetic latex product called
FC2. Classified as class III medical devices by the USFDA,
the regulatory process for female condoms is more
complex than for male condoms. This was compounded by
the lack of an international standard until 2011 to verify the
quality of new female condom devices. To secure
regulatory approvals, including World Health Organization
(WHO)/United Nations Population Fund (UNFPA)
prequalification, manufacturers must conduct clinical
studies to verify the performance of new female condom
designs. Two female condoms (FC2 and Cupid) are now
prequalified based on the results of clinical noninferiority
studies [6,7]. Condom functional performance studies
typically collect detailed data on small numbers of condom
uses (5–10 uses) over a short period of time (4–6 weeks).
The objectives of this study
Velvet were to assess functional
Cupid2
performance, safety and acceptability of the Velvet and
Cupid2® female condoms as compared to the FC2 control
device. We report here on performance and safety data;
acceptability data will be reported elsewhere.

2. Materials and methods

2.1. Study design
Material Synthetic nitrile rubber latex Natural rubber latex
Thiswasathree- Natural rubber latex
Length (MM) 170 170 125
Internal retention mechanism Flexible inner ring
period,randomizednoninferioritycrossover
Flexible inner ring Medical grade sponge
lubricant Silicone oil clinicaltrialtocomparedevicefunction,safetyandacceptabilit
Silicone oil Silicone oil
Outer retention mechanism Circular ring y of two
Circular ring new female condom typesouter
Octagonal with the currently
frame
at the open end of the condom available FC2 as the control device. Of the two available
Shelf life (y) 5 5 3
Other features – prequalified female condoms,
Natural and flavored varieties
FC2 was chosen as the
Natural and pink color,
control as there are more clinical data available
vanilla scented variety on FC2
Manufacturer Female Health Company, USA HLL Lifecare, India Cupid Ltd, India
Approvals USFDA, WHO/UNFPA comparative
India Drug Controlperformance
Authority of India
the Drug
reference
Control condom
Authority, (FC2).
Women were recruited from an urban reproductive health clinic in Durban. The study CE Markwas conducted
of the EU between
August 2013 and April 2014. Potential participants had to be at least 18 years of age and no older than 45, with no known
allergies to the study
0.19% of global condom procurement [3]. In 2011, the
female condom was identified by the Reproductive Health and its functional performance is well documented [6–
Supplies Coalition1 as one of several underused 10]. Practically, the trial could not be blinded as all three
condom designs are quite distinct, each requiring product
1 A global partnership of public, private and nongovernmental
specific training for correct use. The target population was
organizations. Its aim is to ensure that all people in low- and 300 sexually active women in Durban, South Africa. The
middleincome countries can choose, obtain and use the supplies and participants were similar in profile and recruited from the
appropriate services they need to safeguard their reproductive health.
 M. Beksinska et al. / Contraception 92 (2015) 261–267
same site which was used to establish the products (latex,
synthetic latex, polyurethane), using a reliable, nonbarrier
method of contraception, and free of STIs, as determined
by pelvic examination and use of a syndromic diagnostic
tool. Pregnant women (according to urine pregnancy test)
were excluded. Participants were required to be sexually
active and monogamous and could not be practicing sex
workers. Since take-home condom logs were used to
gather data, participants were also required to be literate.
Participants
readandsignedaninformedconsentformpriortoscreeningand
enrolment. In this study, each woman was asked to use
five of each of the three female condom types and to
complete a condom log at home after each condom use.
After completing use of each condom type, they were
asked to return to the clinic to be interviewed about their
experiences and the condom logs were reviewed.
2.2. Study products
The three study female condoms are described in Box 1.
Each female condom product was shipped by the
manufacturer to FHI360 for quality assurance testing to
ensure that study products were of the quality specified by
the manufacturer and met International Standards
Organization (ISO) Standard 25841-2011 [8].
2.3. Ethical considerations
This research study followed the CONSORT guidelines
for the reporting of trials and is registered on the South
African Clinical Trials Database DOH-27-0113-4272.
This protocol and related documents were reviewed and
approved by the Human Research Ethics Committee of
the University of the Witwatersrand, Johannesburg, South
Africa, prior to study initiation and enrollment of
participants (M120522). The study was additionally
approved by the provincial, district and local Departments
of Health in KwaZulu-Natal Province. The consent
process, all questionnaires and condom use logs used were
written and conducted in a language understandable to
participants. Male partners were informed of the study by
the participants through use of a fact sheet which
explained the purpose of the study, provided information
on the study products and on their role as the partner in
the research.
2.4. Primary study objectives and end points
The primary objective of this research was to compare
the functional performance of three female condom types.
The primary noninferiority end points were self-reported
total clinical failure and total female condom failure. Their
component events, clinical breakage, nonclinical
breakage, total breakage, slippage, misdirection and
invagination, were assessed as well. Safety and
acceptability data were also collected. Female condom
263
failure events (modes) are recognized by the WHO and
other regulatory agencies; definitions of each failure mode
analyzed are as follows [9]. In the risk assessment of the
three female condoms used in this study, no other failure
modes were identified.
• Clinical breakage: breakage during sexual intercourse
or during withdrawal of the female condom from the
vagina. Clinical breakage has potential adverse
clinical consequences.
• Nonclinical breakage: breakage noticed before
intercourse or occurring after withdrawal of the
condom from the vagina. Nonclinical breakage is
without potential adverse clinical consequences.
• Total breakage: the sum of all female condom
breakages at any time before, during or after sexual
intercourse, and includes both clinical breakages and
nonclinical breakages
• Slippage: when a female condom slips completely
out of the vagina during sexual intercourse
• Misdirection: vaginal penetration whereby the penis
is inserted between the female condom and the
vaginal wall
• Invagination: when the external retention feature of
the female condom is partially or fully pushed into
the vagina during sexual intercourse
• Total clinical failure: the sum of female condoms that
clinically break or slip, or are associated with
misdirection, invagination or any additional failure
mode(s) identified in the risk assessment, which
result in reduction of the female condom-protective
function • Total female condom failure: a female
condom for which a nonclinical breakage, clinical
breakage or slippage occurs, or is associated with
misdirection, invagination or any additional failure
modes(s) identified in the risk assessment
2.5. Secondary study objectives and end points
Secondary objectives were to assess the safety and
acceptability of each of the female condom types. Safety of
each device was measured and evaluated according to
number, severity, relatedness and duration of adverse
events. Standard acceptability measures were collected, and
these data will be reported elsewhere.
2.6. Randomization and concealment
Each of the 300 women was allocated the three types of
condoms in random order. A Williams design [10] was
used to generate the six possible sequences to which
women could be allocated. This ensured that carryover
effects from the use of one condom type to the following
condom type were balanced. To ensure balance of the six
sequences, block randomization was used [11] with
randomly permutated block sizes of 6 and 12, to conceal
264 M. Beksinska et al. / Contraception 92 (2015) 261–267
the next randomization sequence. Overall, each of the 2.8. Statistical methods
sequences occurred 50 times to give the 300 allocations.
The main analysis for primary and secondary end points
In order to devise a simple and effective method of
(condom failure events) was a comparison of reported
concealment, we used scratch cards, containing the
failure rates of new female condoms with FC2 among the
allocated treatment sequence per participant, where each
subset of participants who provided relevant follow-up data
code for a given female condom type was concealed
on at least one condom of each type.
The
hypothesis
for the
primary
end points,
total
clinical
failure and
total female
condom
failure and
their
component
failure
events, was
that each of
the new
condoms
Velvet and
Cupid2 was
“noninferio
Fig. 1. CONSORT flowchart. r” to FC2
beneath a separate foil square. On each study visit, research (the reference female condom).
staff would remove (scratch) the foil corresponding to the To demonstrate noninferiority, the upper limit of the
visit number printed above the square, thereby revealing two-sided 90% confidence interval (CI), (equivalent to
the next visit allocation in the sequence. The development one-sided 95% CI) for the difference in the occurrence of
and use of the cards is reported elsewhere [12]. failure events (new female condom — FC2) was required
to be below 3.0%, that is, demonstration that failure rate
2.7. Sample size differences were below 3% [8]. If noninferiority was
demonstrated and the difference was less than zero, that is,
We calculated the power to demonstrate noninferiority
the new condom had lower failure rates, the hypothesis of
obtained for different sample sizes, starting with a
superiority was tested at 5% level of significance.
minimum of 200 women completing the study
Primary end points were analyzed using generalized
recommended by ISO 25841 on female condom
estimating equations assuming a binomial distribution of
functionality studies of acute failure events based on self-
the response, an exchangeable correlation matrix and an
reports [8]. We assumed a total failure rate of 4% for the
identity link function. Type of condom was included in the
FC2 as reported from past research [6,7, 13–15], and a
model at the couple-use level defining couples as clusters,
correlation of repeated measures (condom events) within
to take into account correlation of responses for repeated
participants of 0.15, as reported for male condoms [16].
observations for the same couple.
Using 3% as a (clinically determined) margin of
noninferiority and a significance level of 5% for the
hypothesis of noninferiority, 250 women completing the
3. Results
study would provide 98% power. We expected a
noncompletion rate of up to 15% and increased the sample Data were collected between August 2013 and April
size to 300. 2014. A total of 305 women were recruited and screened
from the client population of the reproductive health
clinic. Three
hundredwomenwereenrolledinthestudy.Intotal,282women
 M. Beksinska et al. / Contraception 92 (2015) 261–267 265
Table 1
Baseline sociodemographic characteristics of the main analysis population (N=282)
Characteristic

Age (y) Mean (SD) 27.4 (5.78)

Min–Max 18–45

Ethnic group, n (%) Black 282 (100)

Completed years of school, n (%) 8 6 (2.1)
9 3 (1.1)

10 14 (5)

11 55 (19.5)

12 204 (72.3)

Primary occupation, n (%) None/Unemployed 160 (56.7)

Self-employed + other 5 (1.7)

Unskilled labor 54 (19.1)

Health/Medical 1 (0.4)

Office 4 (1.4)

Public service/government 1 (0.4)

Sales 2 (0.7)

Student 53 (18.8)

Teacher/Lecturer 1 (0.4)

Volunteer 1 (0.4)

Marital status, n (%) Married or living together 62 (22)

Not married and not living together 219 (77.6)

Married and not living together 1 (0.4)

Length of relationship with spouse/partner, n (%) Less than 1 y 6 (2.1)

1–5 y 161 (57.1)

6–10 y 80 (28.4)

More than 10 y 35 (12.4)

Number of living children, n (%) 0 27 (9.6)

1 138 (48.9)

2 71 (25.2)

3 or more 46 (16.3)

completedthestudy(94%),with each womanusing at
leastone condom of each of the three female condom
types. These women comprise the main analysis
population. The study
flowchartisshowninFig.1.Therewereonlythreewomenwho
did not return for any follow-up, eight came for one
follow-up visit and seven for two follow-up visits.
The mean age of participants was 27.4, and the
majority (96.8%) had 10 years or more of schooling
(Table 1). Almost a fifth (18.8%) were students, and just
over half (56.7%) were
unemployed (Table 1). Most participants were not married
or living with their partners (77.6%). The majority of
participants (98.9%) had ever used male condoms, while
266 M. Beksinska et al. / Contraception 92 (2015) 261–267
just over a fifth (22%) had ever used female condoms (data
not shown).
Intotal,4214condompacketswereopenedand4147female
condoms were used in the study. All three female
condoms had similar rates of nonclinical breakage of 2%.
Rates of invagination, slippage and misdirection were all
under 2%. Noninferiority was demonstrated for total
female condom failure for the two female condoms
Cupid2 and Velvet with respect to the FC2 within the
margin of 3% difference in mean failure, at the 5%
significance level (Table 2). The p values in Table 2
indicate the differences between the test condoms and
FC2foreachfailuremodeInonefailuremode—misdirection—
the Cupid2 was significantly different from FC2 with 25
events compared to 16, respectively (p=.026). The
additional analysis performed with women with complete
or incomplete condom series provided similar results (not
shown).
There was no evidence of superiority for any of the
new female condoms with respect to FC2 in any of the
failure modes (95% CIs in Table 2 include the null value;
p values for two-sided superiority tests all greater than .
05).
Overall, the occurrence of adverse events occurred in
approximately 2% of condom uses. No serious adverse
events were reported. A total of 110 events were recorded
in 84 condoms. Over two-thirds were reported as very
slight (69.1%) and mild by the remaining third (30.9%).
No action was taken in any of the 110 events, nor was any
treatment used. Under half of the events were reported to
have occurred in men (41.8%; n=46) and over half in
women (58.2%; n=64). Of the 110 events, 55 events
occurred with Cupid2 in 38 condoms; 31 events occurred
in 25 FC2 condoms and 24 events occurred in 21 Velvet
condoms. Hence, adverse events were highest in Cupid2.
 M. Beksinska et al. / Contraception 92 (2015) 261–267 267
4. Discussion ⁎ p Value indicates differences between the test condom and FC2.

The two new female condoms (Cupid2 and Velvet)

evaluated in this study are noninferior to the reference
condom (FC2) within a margin of 3% failure for all the
functional parameters. The reporting of adverse events was
1% higher than the rates reported in a previous study [7],
which included the same control FC2 and the currently
available Cupid. In this study, adverse events for Cupid2
were the highest of the three condoms evaluated. The FC2
is made of synthetic latex and therefore does not expose
users to the proteins in natural rubber latex that can cause
latex allergies. The Velvet and Cupid2 are made of latex,
and so it would be expected that these condoms may cause
more irritation than those made of synthetic latex. The
outer ring/ frame of the Cupid2 is also more rigid compared
to velvet and FC2 and may have resulted in more
discomfort in some users.
Table 2
Mean failure and failure difference of Cupid2, Velvet in relation to FC2, with two-sided 90% CIs for noninferiority hypothesis at α=5% and two-sided 95%
CIs for superiority hypothesis, for the main analysis population
The total clinical failure, total female condom failure
and component failure rates reported are expected and
consistent with findings from earlier, similarly conducted
studies [6,7, 13–15]. Unlike male condoms, new female
condoms are required to provide evidence of functional
noninferiority with respect to the marketed FC2. Data from
this study have been submitted to WHO/UNFPA and will
contribute to the prequalification submission requirements
for the Cupid2 and Velvet female condoms.
Evidence for the effect of choice of method on increased
uptake of contraceptive methods is limited; however, a
systematic review of the evidence in 2006 supports the
argument that increased contraceptive choice for women is
associated with increased uptake and better health

Condom function Condom Condoms, Mean Failure difference (%) p-Value⁎

type n failure
Difference with FC2 90% CI 95% CI
(n=282), (%) n
Nonclinical breakage Cupid2 1397 2.2 (31) −0.15 (−1.04 to 0.75) (−1.22 to 0.92) .798
Velvet 1406 1.98 (28) −0.37 (−1.24 to 0.51) (−1.41 to 0.68) .489

FC2 1410 2.35 (33) (Ref) – – –

Clinical breakage Cupid2 1373 0.66 (9) −0.23 (−0.77 to 0.32) (−0.87 to 0.42) 0.493
Velvet 1391 0.59 (8) −0.30 (−0.83 to 0.23) (−0.93 to 0.33) 0.350

FC2 1383 0.89 (12) (Ref) – – –

Invagination Cupid2 1373 1.03 (14) −0.67 (−1.38 to 0.03) (−1.52 to 0.17) 0.118
Velvet 1391 1.37 (19) −0.34 (−1.08 to 0.41) (−1.22 to 0.55) 0.457

FC2 1383 1.71 (23) (Ref) – –

Misdirection Cupid2 1373 1.82 (25) 0.84 (0.26 to 1.71 (0.12 to 1.85) 0.026
Velvet 1391 1.15 (16) 0.18 (0.43 to 0.79) (−0.55 to 0.90) 0.628

FC2 1383 0.98 (13) (Ref) – – –

Slippage Cupid2 1373 1.47 (20) 0.60 (−0.06 to 1.3) (−0,19 to 1.40) 0.136
Velvet 1391 0.87 (12) 0.00 (−0.58 to 0.56) (−0.68 to 0.67) 0.984

FC2 1383 0.87 (12) (Ref) – – –

Total clinical failurea Cupid2 1373 4.79 (63) 0.29 (−0.94 to 1.51) (−1.72 to 1.75) 0.699
Velvet 1391 3.93 (54) −0.60 (−1.74 to 0.60) (−1.20 to 0.82) 0.422

FC2 1383 4.50 (60) (Ref) – – –

Total female condom failure Cupid2 1398 6.69 (94) 0.87 (−1.30 to 1.56) (−1.58 to 1.83) 0.882
Velvet 1406 5.82 (82) 0.19 (−2.23 to 0.54) (−2.50 to 0.80) 0.314

FC2 1410 6.66 (93) (Ref) – – –

a
Six condoms experienced two failures which are considered as one condom failure for total clinical failure.
268 M. Beksinska et al. / Contraception 92 (2015) 261–267
outcomes (such as lower pregnancy rates and fewer STIs)
[17]. Further, women continue use of their chosen
contraceptives to a greater degree than those denied a
choice of method [18]. Access to a variety of female
condom types could improve choice for women in need of
a contraceptive method or for dual protection against
pregnancy and infection.
The availability of new female condom products will
provide donors and users with more options and has
potential to reduce unit price of devices [4]. The
Reproductive Health Supplies Coalition has published
information comparing different female condom unit costs
to the donor funded public sector, indicating that when
Cupid female condom was prequalified in 2012, it was sold
at a lower price per unit compared to FC2 [4].
The current public sector price of the two available
prequalified female condoms (FC2 and Cupid) are still
much higher than most developing country programs can
support, with female condoms costing approximately 20
times more to produce compared to male condoms [19]. If
new designs become available, increased choice and
competition may lead to a price reduction and increased
distribution of female condoms globally.
5. Limitations
A limitation of this study was that it was not possible to
blind the participants and research staff to the products as
they are all quite different. Allocation concealment was
used to ensure that this limitation was minimized. This
study was based exclusively on self-reported measures of
condom use. To validate condom function more
accurately, the use of PSA would provide stronger
evidence if failures reported led to exposure to semen and
this would supplement, and possibly partly displace,
observations and records made by the participants.
Conflicts of interest
There are no conflicts of interest for all authors during
the study period.
Acknowledgments
We thank all research nurses, interviewers and data
entry staff who collected and entered the data. We thank
all participants who gave their time to participate in the
trial.
References
[1] Gallo MF, Kilbourne-Brook M, Coffey P. A review of the
effectiveness and acceptability of the female condom for dual
protection. Sex Health 2012;9(1):18–26,
http://dx.doi.org/10.1071/SH11037.
[2] The Female Health Company Annual report 2013. femalehealth.
investroom.com/download/FHC02013final.pdf [accessed Nov 17th
2014]. [3] UNFPA. HIV prevention gains momentum. UNFPA978-
0-89714-933-4; 2011.
[4] Reproductive Health Supplies Coalition. Caucus on new and
underused reproductive health technologies product brief: female
condom. http://www.rhsupplies.org/working-groups/caucus-
onnewunderused-rh-technologies.html.
[5] Beksinska M, Smit J, Joanis C, Potter WD. New female condoms in
the pipeline. Reprod Health Matters 2012;20(40):188–96,
http://dx.doi.org/ 10.1016/S0968-8080(12)40659-0.
[6] Beksinska M, Smit J, Mabude Z, Vijayakumar G, Joanis C.
Performance of the Reality® polyurethane female condom and a
synthetic latex prototype: a randomised cross-over trial among
South African women. Contraception 2006;73:386–93.
[7] Beksinska M, Piaggio G, Smit J, Wu J, Zhang Y, Pienaar J, et al.
Performance and safety of the second-generation female condom
versus the Woman’s, the VA worn-of-women, and the Cupid female
condoms: a randomised controlled non-inferiority crossover trial.
Lancet Glob Health 2013;1:e146–52, http://dx.doi.org/10.1016/
S2214-109X(13)70054-8.
[8] ISO.ISO 25841:2011. Female condoms — requirements and test
methods. http://www.iso.org/catalogue_detail.htm?
csnumber=43282 [accessed July 6, 2012].
[9] Beksinska M, Joanis C, Manning J, Smit J, Callahan M, Deperthes
B, et al. Standardized definitions of failure modes for female
condoms. Contraception 2007;75(4):251–5.
[10] Williams EJ. Experimental designs balanced for the estimation of
residual effects of treatments. Aust J Sci Res 1949;2(3):149–68.
[11] StataCorp. Stata Statistical Software: Release 13. College Station,
TX: StataCorp LP.STATA; 2013.
[12] Beksinska M, Joanis C, Smit J, Pienaar J, Piaggio G. Using scratch
card technology for random allocation concealment. Clin Trials
2012;10:125–30, http://dx.doi.org/10.1177/1740774512465496
[15].
[13] Schwartz JL, Barnhart K, Creinin MD, Poindexter A, Wheeless A,
Kilbourne-Brook M, et al. Comparative crossover study of the
PATH Woman's Condom and the FC Female Condom®.
Contraception
2008;78:465–73,http://dx.doi.org/10.1016/j.contraception.2008.07.020.
[14] Smita J, Neelam J, Rochelle DY, Kumar BK, Callahan M, Suniti S,
et al. Comparative acceptability study of the Reality female condom
and version 4 of modified Reddy female condom in India.
Contraception 2005;72:366–71.
[15] Joanis C, Beksinska M, Hart C, Tweedy K, Linda J, Smit J. Three
new female condoms: which do South African women prefer?
Contraception 2011;83:248–
54,http://dx.doi.org/10.1016/j.contraception.2010.08.002.
[16] Taylor D. Issues in the design, analysis and interpretation of
condom functionality studies. Contraception 2009;80:237–44.
[17] Gray A, Smit J, Manzini N, Beksinska M. Systematic Review of
Contraceptive Medicines. “Does choice make a difference?”. http://
archives.who.int/eml/expcom/expcom15/applications/sections/
ContraChoiceReview.pdf 2006.
[18] Jain AK. Fertility reduction and the quality of family planning
services. Stud Fam Plann 1989;20:1–6.
[19] http://www.prb.org/Publications/Articles/2003/TheFemale
CondomWomenControlSTIProtection.aspx [accessed 18th April
2015].