CSIRO PUBLISHING & MINNIS COMMUNICATIONS
Sexual Health, 2006, 3, 255–260
Erection loss in association with condom use among young men
attending a public STI clinic: potential correlates
and implications for risk behaviour
Cynthia A. GrahamA,B,C,J, Richard CrosbyC,D, William L. YarberB,C,E,F, Stephanie A. SandersB,C,F,
Kimberly McBrideG, Robin R. MilhausenHand Janet N. ArnoC,I
AOxford Doctoral Course in Clinical Psychology, Oxford, UK.
BThe Kinsey Institute for Research in Sex, Gender, and Reproduction, Bloomington, IN, USA.
CRural Center for AIDS/STD Prevention at Indiana University, Bloomington, IN, USA.
DCollege of Public Health at the University of Kentucky, Lexington, KY, USA.
EDepartment of Applied Health Science, Indiana University, Bloomington, IN, USA.
FGender Studies, Indiana University, Bloomington, IN, USA.
GDepartment of Adolescent Medicine, Indiana University School of Medicine, Indianapolis, IN, USA.
HSocial Justice and Sexual Health Research Lab, Department of Sociology and Anthropology,
University of Windsor, Windsor, Ontario, Canada.
IDepartment of Infectious Diseases, Indiana University School of Medicine, Indianapolis, IN, USA.
JCorresponding author. Email: firstname.lastname@example.org
erection loss among men attending a sexually transmissible infections (STI) clinic. Methods: Men (n=278)
attending an STI clinic responded to an anonymous questionnaire aided by a CD recording of the questions. The
sample was screened to include only men who had used a condom during penile–vaginal sex at least three times
in the past 3 months. Erection loss was assessed for ‘the last three times a condom was used’. Results: The
mean age of the participants was 23.7 years (s.d.=4.1); 37.1% of the men reported condom-associated erection
loss on at least one occasion. Men who had reported condom-associated erection loss were also reported having
more frequent unprotected vaginal sex (P=0.04) and were less likely to use condoms consistently (P=0.014)
than men without erection loss. Men with erection loss were also more likely to remove condoms before sex was
over (P=0.001). Age and race/ethnicity were not associated with erection loss. In multivariate analysis, three
significant statistical predictors were identified: low self-efficacy to use condoms (P=0.001); problems with ‘fit or
feel’ of condoms (P=0.005); and having more than three sex partners during the previous 3 months (P=0.02).
Conclusions: Condom-associated erection loss may be common among men at risk for STIs. This problem may
lead to incomplete or inconsistent condom use. Men may be more likely to experience condom-associated erection
loss if they lack confidence to use condoms correctly, if they experience problems with the way condoms fit or feel,
and if they have sex with multiple partners.
Background: To assess prevalence of condom-associated erection loss and to identify correlates of
Condom use is an important primary prevention strategy
that can substantially reduce the likelihood of transmitting
and acquiring sexually transmissible infections (STIs).1−3
Thus, one longstanding challenge in public health practice
has involved motivating men to use condoms consistently
and correctly.4The challenge has proven formidable
and therefore requires continued effort. For example,
accumulating evidence suggests that men may experience a
host of problems related to the correct use of condoms.5−8
One difficulty that may be especially important to men,
and yet has received surprisingly little attention, is the loss
of erection while applying or using condoms. Indeed it is
reasonable to predict that men who experience condom-
associated erection loss may be reluctant to use condoms and
more likely to report incomplete use of condoms (not using
that identify the correlates of condom-associated erection
loss could be an important starting point for behavioural
Several studies have explored the relationship between
loss of erection and condom use. Among a sample of
© CSIRO 200610.1071/SH060261448-5028/06/040255
C. A. Graham et al.
HIV-positive gay men who had recently seroconverted,
erection problems in association with condom use were a
common concern.9Further, reported inability to achieve
or maintain an erection sometimes led to a decision not to
use condoms, either occasionally or altogether. In another
study involving both HIV-positive and HIV-negative gay
men, there was a relationship between reporting erectile
problems and sexual risk taking (indexed by number of
casual partners with whom condoms were not used).10
Studies involving college students have consistently found
that a proportion (12–32%) of men report erection loss
in association with condom use.5,7,11,13Male college
students who reported loss of erection during condom use
were also more likely to report condom slippage.6In a
recent investigation of incarcerated adolescent males who
were interviewed during their intake physical examination,
the prevalence of erection loss ‘before condom removal’
was 18% during ‘usual use’ and 7% for the last sexual
encounter.13Last, in a study that assessed the prevalence
of condom-related erection problems in the previous year
among men attending STI clinics, erection loss during
condom application or during intercourse using a condom
(before ejaculation) was commonly reported (e.g. less than
between erection loss and condom breakage or slippage
was found. Unfortunately, however, these researchers did
not assess whether erection problems were related to
Clearly, men attending an STI clinic are an important
population for the study of condom-associated erection
loss. Sexually transmissible infection burden may be higher
among relatively younger men.15Given the advanced and
costly sequelae from STIs that occur among women,15
assessment of men who have sex with women is particularly
important. Accordingly, this study was conducted using a
sample of relatively young men who have sex with women
(MSW) attending an STI clinic. The primary purpose was
to assess the prevalence of condom-associated erection
loss. Second, the study determined whether erection loss
was significantly associated with incomplete use (not using
condoms throughout the sexual encounter) or with less
consistent use of condoms. Finally, the study identified
Data were collected at a large, urban, midwestern, public STD clinic
from October 2004 to September 2005. Men attending the clinic were
Inclusion criteria were: (1) 18–35 years of age; (2) English speaking;
and (3) reporting male condom use at least three times in the past
3 months for sex (penis in vagina) with a female. Five hundred and
314 (89.5%) agreed to participate and completed a questionnaire.
A questionnaire refined through use in several studies involving
morethan 800men, including
used to comprehensively assess men’s condom-use errors and
problems.5,7,8,16,17Original questionnaire development was informed
by widely cited condom-use guidelines.18,19Because accuracy of recall
was considered vital,10the recall period was limited to the last three
times condoms were used within the past 3 months. For each question,
men indicated whether the error or problem occurred zero to three
times. Sex was defined as ‘sexual intercourse, or penis in vagina’.
Two measures assessed erection loss: (1) ‘For the last three times
after sex had begun while using the condom?’ Men indicated whether
erection loss had occurred at all and, if so, whether it had occurred on
one, two or three occasions.
Association with risk behaviour
Two measures were assessed to determine if erection loss was
associated with non-use or incomplete condom use. The first was the
frequency of unprotected penile–vaginal sex (UVS) during the past
then asked how many times a condom was used. The latter measure was
then subtracted from the former to produce the measure of UVS. The
second was the behaviour of removing condoms from the penis before
you used a condom, did you start having sex with it on and then take
it off before sex was finished?’ Response alternatives were ‘no’, or on
one, two or three occasions.
Other history-related correlates
Nine potential correlates were assessed. The first two were age and
race/ethnicity (white/non-white). Next, three problems were assessed
using the recall period of the last three times condoms were used. These
were: (1) breakage, (2) slippage and (3) problems with the ‘fit or feel’
of condoms. Men also answered an item asking if they had ever been
taught how to use condoms. They also indicated how many sex partners
they had had sex with in the past 3 months. Men’s motivation to use
condoms was assessed by the following item, ‘I am highly motivated
to use condoms correctly’. Responses to this item were provided using
a scale ranging from ‘1’ (strongly agree) to ‘5?(strongly disagree).
Finally, an eight-item index was used to assess men’s self-efficacy for
the correct application of condoms.20These items asked men how
‘easy or difficult’ it would be for them to perform behaviours related
to condom use. For example, one item was: ‘How easy or difficult
would it be for you to apply condoms correctly?’ Response alternatives
ranged from ‘1’ (very easy) to ‘5?(very difficult). The index produced
a satisfactory Cronbach’s α of 0.70, suggesting adequate reliability
of the measure.
After providing written informed consent, volunteers completed a
brief self-administered written questionnaire lasting 15–20min. To
reduce problems with literacy, the questions were recorded to a CD
that men could choose to play using a portable headset to assist
them in providing written answers to the orally presented questions.
Each question constituted a single track; thus, men could easily
replay a question, just as they would a track of music. Responses
were anonymous. Men who completed the questionnaire were paid
$10. The Institutional Review Board at Indiana University approved
Erection loss in association with condom use
The measure of erection loss was dichotomised as one or more events
versus none. Associations between erection loss and UVS were then
assessed. First, UVS was treated as a continuous variable and an
independent groups t-test was used. Next, UVS was dichotomised
into consistent condom use (100%) versus less than consistent use.
Contingency table analysis was then applied, using prevalence ratios,
95% confidence intervals and their respective P-values. Contingency-
table analysis was also used to test the association between erection loss
times during the last three times condoms were used).
To identify correlates of condom-associated erection loss, bivariate
associations were determined followed by a multivariate test of
associations. Age was preserved as a continuous measure and its
bivariate association with erection loss was determined by a t-test
comparing mean age for those reporting erection loss with that for
those not reporting erection loss. The remaining correlates were
tested using contingency-table analysis. Of note, skewness necessitated
dichotomisation of the index measure pertaining to self-efficacy. Scores
achieve a median split, men with scores of 15 and less (high efficacy)
were compared with men with scores of 16 or more (low efficacy).
Correlates achieving bivariate significance (P<0.05) were entered into
a forward stepwise logistic regression model. Multivariate significance
was defined by 95% confidence intervals and P-values of less than 0.05.
Characteristics of the sample
Despite screening attempts, 36 men provided questionnaire
responses that indicated that they were in fact not eligible,
thereby leaving an analytic sample of 278 men (88.5% of
the 314). The mean age was 23.7 years (s.d.=4.1). About
two-thirds (67.6%) identified as Black or African American,
nearly one-quarter (23.7%) as white and the remainder as
other minority groups.
Prevalence of condom-associated erection loss
Nearly three of every 10 men (28.1%) reported they had
lost their erection on one of the last three times that they
applied condoms; none indicated that this happened more
than once. Thirty-seven men (13.4%) reported they had lost
their erection once while using a condom during vaginal–
penile intercourse. Twenty-six men (9.4%) reported that this
happened twice and 10 men (3.6%) reported that it had
happened on all three occasions. About one of every six
men (17.3%) reported both events (i.e. erection loss during
application and during use). Combining the two measures,
37.1% of the men reported condom-associated erection loss
on at least one occasion. Specifically, 47 men (16.9%)
reported experiencing one event of erection loss, 25 (9.0%)
reported two events, 24 (8.6%) reported three events, and
seven (2.5%) reported four events.
Association of reports of erection loss with risk
Those reporting any erection loss engaged in UVS more
frequently (mean=10.6 times) over the past 3 months than
men indicating no erection loss (mean=7.0 times) (t=2.0
more likely to report inconsistent condom use (i.e. less than
100% of the time). Inconsistent condom use was reported by
95% CI=1.04–1.34; P=0.014). Men reporting erection
loss were also significantly more likely to remove condoms
before sex was over. Condoms were removed prematurely on
at least one (of the past three) occasions by 40.8% of the
men reporting erection loss compared with 21.3% of men
not reporting this problem (PR=1.92; 95% CI=1.33–2.77;
Other history-related correlates of condom-associated
Age was not significantly associated with erection loss.
The mean age of those not reporting erection loss was
23.6 years compared with 24.0 years among those reporting
erection loss (t=0.95 (272); P=0.34). Table 1 displays the
correlates and erection loss. As shown, four of the eight
correlates achieved bivariate significance. Among men
classified as having low self-efficacy for correct condom
use, nearly 50% reported erection loss compared with 24%
among men with high self-efficacy. Erection loss was also
more likely among men who reported at least one condom
breakage (47.1%) than men not reporting breakage (32.5%).
not reporting this problem. Finally, men reporting sex with
three or more partners (during the past 3 months) were more
likely to have erection loss (45.0%) than men having one or
two partners (29.6%).
Table 2 displays the results of a logistic regression model
for erection loss. The model was significant (c2 with
3 d.f.=33.7, P<0.0001), and achieved an excellent fit with
the data (goodness of fit c2 with 6 d.f.=2.88, P=0.82).
Three of the four correlates entered retained significance.
Compared with men with high self-efficacy, those classified
as having low self-efficacy to use condoms were ∼2.8 times
more likely to report erection loss. Men reporting recent
problems with the ‘fit or feel’ of condoms were ∼2.2 times
more likely to report erection loss than men not having these
problems. Finally, men reporting sex with three or more
partners were ∼1.9 times more likely than men having fewer
partners to report erection loss.
The findings suggest that condom-associated erection loss
may be a relatively frequent occurrence among MSW
attending public STI clinics. Evaluating only the last three
times a condom was used during the past 3 months,
C. A. Graham et al.
278 men attending an STI clinic
Correlates% with lossA
Motivation to use condoms
Self-efficacy for correct condom use
Taught how to use condoms
Condom slipped off during sex
Condom broke during sex
Problems with ‘fit and feel’ of condoms
Three or more sex partners (past 3 months)
2.06 1.46–2.91 0.001
ADefined as experiencing at least one of two types of erection loss during the last 3 times condoms were used.
Table 2.Multivariate differences between men who did and did
not report erection loss
Low self-efficacy for correct
Problem with fit or feel of condoms
Three or more sex partners
(past 3 months)
AAdjusted odds ratio – adjusted for the influence of all other variables
in the model.
protected penile–vaginal sex was reported by nearly four of
every 10 men (37.1%). Although erection loss is in itself an
important health issue for men, our findings also suggest that
condom-associated erection loss may be linked to non-use or
incomplete use of condoms. Thus, consistent with previous
studies involving gay men,9,10for MSW who are at high
risk for STIs, condom-associated erection loss may lead to
increased risk-taking behaviour through UVS.
There was no significant association between age and
condom-associated erection loss. Given the strong positive
in general,21one might expect reports of erection problems
during condom use to be more prevalent among older men.
However, the current sample was limited to predominantly
young men, aged 18–35 years, making it less likely that a
significant association would be found.
The findings from this study have important implications
for education and counselling efforts directed towards the
promotion of condom use to men at risk of STIs. First,
it may be that men who lack confidence in their ability
to use condoms correctly would be particularly receptive
to receivingclinic- orcommunity-based
designed to enhance their condom-use skills. This may
be particularly important for men who worry about the
possibility of losing their erection when using a condom.
Such instruction should be interactive22and would ideally
involve guided practice applying condoms to a partially
rigid penile model. In turn, enhanced ability to apply
condoms might reduce the likelihood of erection problems
during condom application. Indeed, if one outcome of an
education program was greater confidence in erectile ability
for male participants, this might lead to more consistent
Second, the findings suggest that problems related to ‘fit
and feel’ of condoms may be critically important to the
issue of condom-associated erection loss. Various negative
Erection loss in association with condom use
and feel’ of condoms in previous studies.8,17The sensations
associated with poorly fitted condoms or those that don’t
feel ‘right’ may directly affect erection. Alternatively, it
may be that ill-fitting (i.e. too tight or too loose) condoms
create anxiety about slippage or breakage and resultant
pleasure and lead to erection loss. Similarly, it is possible
that poorly lubricated condoms become quite dry and cause
irritation for either partner. In turn, this irritation may reduce
sexual sensation, increasing the likelihood of male erection
and may increase the risk of erection loss because of reduced
focus on sexual sensations.
Finally, the study found that men having sex with three or
more partners had a greater risk of experiencing condom-
associated erection loss. It is possible that men who are
experiencing erection difficulties may engage in more risk
behaviour, and with more female partners, in an effort
to heighten their sexual arousal. Alternatively, it may be
that negotiating sex with a new partner combined with the
challenge of using condoms may be particularly anxiety
provoking, thereby making erection loss more likely. As
this was a cross-sectional study, we cannot establish which
of these two explanations is correct. If having numerous
partners does increase the likelihood of experiencing erectile
problems, the implication is that education or counselling
efforts might target this group of men by teaching them
how to discuss condom use with a new partner. Such
discussions could be constructed to allow men some delay
in the progression of erotic events to pause for condom
application or (if condoms become dry) to briefly interrupt
penises fluctuate in rigidity, that temporary loss of erection
is a common experience (maybe especially when applying
condoms) and that if their erection subsides, they should
allow time for it to return. Another suggestion might be for
which might be experienced as greater interruption of the
‘flow’ of sexual activity. Indeed, recent evidence suggests
that women may commonly apply condoms to their male sex
partners,11although whether this reduces the likelihood of
male partners experiencing erectile problems is not known.
Lastly, the findings also have implications for future
research. Assessing erection problems associated with
condom use should be a critical component of studies that
investigate reasons for non-use or incomplete condom use.23
This study compared men who reported erection loss during
the last three times they used condoms with men who did
not report erection difficulties. Because we did not obtain
information about erection loss during sexual encounters
when condoms were not used, the data collected do not allow
any causal conclusions. Additionally, one of our correlates,
‘problems with fit and feel’ of condoms, may be a proxy for
erection difficulties to some extent and thus causal in both
As is true for any study of sexual behaviour, the findings
are limited by the validity of the self-reported data. Although
have minimised recall error. Generalisability is also limited
by the use of a convenience sample.
The findings suggest that more in-depth research should
condom-associated erection difficulties. Indeed a qualitative
study may be an important ‘next step’ in learning more about
condom-associated erection loss.
To our knowledge, this is the first published study that
has reported correlates of condom-associated erection loss.
Among MSW attending a public STI clinic, condom-
associated erection loss may be a relatively common
occurrence. This problem may lead men to use condoms
less often or to remove condoms before sex is over. Men
loss if they lack confidence to use condoms correctly, if they
experience problems with the way condoms ‘fit or feel’ and
if they have sex with numerous partners. These correlates
are amenable to brief, clinic-based interventions, thereby
suggesting that adding such programs to clinic protocols is
have an impact on STI incidence rates. Future studies should
include assessment of men’s confidence in their erectile
ability, because men who experience condom-associated
erection loss may be those who are more vulnerable to
experience erection difficulties in situations not involving
Conflicts of interest
Support for this project was provided by the Rural Center for
AIDS/STD Prevention, a joint project of Indiana University,
University of Colorado, and University of Kentucky, and
the Office of the Associate Dean of Research, School
of Health, Physical Education, and Recreation, Indiana
Rose Hartzell, Martha Payne, Lauri Legocki, Lindsay Brown
and Alexis Rothring.
C. A. Graham et al.
1 Holmes KK, Levine R, Weaver M. Effectiveness of condoms
in preventing sexually transmitted infections. Bull World Health
Organ 2004; 82: 454–61.
2 Warner L, Stone KM, Macaluso M, Buehler JW, Austin
HD. Condom use and risk of gonorrhea and chlamydia: a
systematic review of design and measurement factors assessed
in epidemiologic studies. Sex Transm Dis 2006; 33: 36–51.
3 Centers for Disease Control and Prevention. Sexually transmitted
diseases treatment guidelines 2002. MMWR Morb Mortal Wkly
Rep 2002; 51: 1–60.
4 Crosby R, DiClemente RJ, Holtgrave DR, Wingood GM. Design,
relative to condom effectiveness against non-viral STIs. Sex
Transm Infect 2002; 78: 228–31. doi: 10.1136/sti.78.4.228
5 Crosby R, Sanders S, Yarber WL, Graham CA. Condom-use
errors and problems: a neglected aspect of studies assessing
condom effectiveness Am J Prev Med 2003; 24: 367–70.
6 Yarber WL, Graham CA, Sanders SA, Crosby RA. Correlates
undergraduates. Int J STD AIDS 2004; 15: 467–72. doi: 10.1258/
7 Crosby RA, Sanders S, Yarber WL, Graham CA, Dodge B.
Condom use errors and problems among college men. Sex Transm
Dis 2002; 29: 552–7.
8 Crosby R, Yarber WL, Sanders SA, Graham CA. Condom
discomfort and associated problems with their use among
university students. J Am Coll Health 2005; 54: 143–8.
9 Richters J, Hendry O, Kippax S. When safe sex isn’t safe. Cult
Health Sex 2003; 5: 37–52.
mood, and sensation seeking. Arch Sex Behav 2003; 32: 555–72.
11 Sanders SA, Graham CA, Yarber WL, Crosby RA. Condom use
errors and problems among young women who put condoms
on their male partners. J Am Med Womens Assoc 2003;
12 Warner L, Clay-Warner J, Boles J, Williamson J. Assessing
condom use practices. Implications for evaluating method and
user effectiveness. Sex Transm Dis 1998; 6: 273–7.
13 Bortot AT, Risser WL, Cromwell PF. Condom use in incarcerated
adolescent males: knowledge and practice. Sex Transm Dis 2006;
33: 2–4. doi: 10.1097/01.olq.0000187195.51056.2b
14 Richters J, Gerofi J, Donovan B. Why do condoms break or slip
off in use? An exploratory study. Int J STD AIDS 1995; 6: 11–8.
15 Eng TR, Butler WT, editors. The hidden epidemic: confronting
Academy Press; 1997.
16 Crosby R, Salazar LF, DiClemente RJ, Yarber WL, Staples-
Horne M. Accounting for failures may improve precision:
condom use. Sex Transm Dis 2005; 32: 513–5. doi: 10.1097/
17 Crosby RA, Graham CA, Yarber WL, Sanders SA. If the condom
fits, wear it: a qualitative study of young African American men.
Sex Transm Infect 2004; 80: 306–9. doi: 10.1136/sti.2003.008227
18 Centers for Disease Control and Prevention. Facts about condoms
GA: Department of Health and Human Resources; 1998.
19 Warner DL, Hatcher RA. Male condoms. In: Hatcher RA,
Trussell J, Stewart F, Cates W Jr, Stewart GK, Guest F, Kowal D,
editors. Contraceptive technology, 17th edn. New York, NY:
Irvington Publishers Inc; 1999. pp. 325–352.
20 Crosby R, DiClemente RJ, Wingood GM, Sionean C, Cobb
BK, Harrington K, Davies S, Hook EW 3rd, Oh MK. Correct
condom application among African American adolescent
females: the relationship to perceived self-efficacy and the
association to confirmed STDs. J Adolesc Health 2001; 29:
194–9. doi: 10.1016/S1054-139X(01)00273-7
21 Feldman HA, Goldstein I, Hatzichristou DG, Krane RJ,
McKinlay JB. Impotence and its medical and psychosocial
correlates: results of the Massachusetts male aging study. J Urol
1994; 151: 54–61.
et al. Efficacy of risk-reduction counseling to prevent human
immunodeficiency virus and sexually transmitted diseases:
a randomized controlled trial. JAMA 1998; 280: 1161–7.
23 Graham CA, Crosby RA, Sanders SA, Yarber WL. Assessment
of condom use in men and women. Annu Rev Sex Res 2005; 16:
Washington, DC: National
Received 26 April 2006, accepted 13 October 2006