Condom use in the real world

Perhaps the most crucial question to ask about condoms is not whether they work - but whether they are used.

This distinction is the difference between efficacy – whether an intervention works in ideal circumstances – and effectiveness – whether it reduces disease incidence in the general population.

As we have seen, when condoms are used consistently, their efficacy in preventing HIV and bacterial STIs is in the order of 85%.

Given that the seroconversion rate between ‘sometimes’ and ‘never’ users is very similar, the effectiveness of condoms then becomes their efficacy multiplied by the fraction of the relevant population who use them. So, for instance, in a population where about 50% are ‘always’ or consistent condom users – and recent studies have shown that this appears to be the case amongst many groups of gay men – then the effectiveness is 50 x 0.85 or 42.5%. That is, condoms are stopping 42.5% of HIV infections that would have happened if no-one used them. In a population where 95% use condoms – as in clients of Thai sex workers – then the effectiveness rate is just over 80%.

Condom efficacy in individual cases may not necessarily translate into effectiveness on a population level. The success of condoms as an HIV prevention method depends on how HIV is being transmitted in a particular population. The highest success levels have been reported from countries where HIV is being spread through commercial sex work. Here, campaigns to encourage condom use in sex workers and their clients have proved extremely effective.

This is for several reasons. Firstly, in many situations a mobile but relatively small population of sex workers may service a much larger client group. This provides a ‘pinch point’ whereby, if condom use and insisting upon it becomes normal behaviour in sex workers, there are few other channels by which HIV can be spread. Secondly, the fact that sex work is usually legally restricted and sometimes, formally or informally, monitored and supervised by the police and local politicians means that pressure can be put upon the operators of sex-work establishments to  enforce condom use. Thirdly, although sex workers may be a stigmatised and disenfranchised group, the power differential between men and women in the actual process of negotiating sexual safety may not be as stark as it is in societies where married women effectively have no rights over sexual choice. Fourthly, what is being sold is sex, not love, and issues of trust do not apply; whereas in marriage the insistence upon condom use may be experienced as symbolic of a lack of closeness or as evidence of infidelity, in a commercial sex work encounter it is simply good sense.

Conversely, in heterosexual and gay relationships alike, condom use tends to be the exception rather than the rule in long-term relationships, and even in relationships where partners know they have different HIV statuses condom use is often rarer than it is in casual sex situations. In generalised epidemics where HIV is being largely spread within marriage and longer-term, emotionally committed relationships, condoms and their promotion may make less difference to the spread of HIV than an emphasis on the ‘A’ (abstinence) and ‘B’ (being faithful) of prevention.

These two positions have generated a polarised debate within HIV prevention in the last ten years, but in reality all prevention tools have their place, and what works best for one epidemic situation may not be the best solution for others. The point was made at the 2006 PEPFAR implementers’ meeting by David Stanton of USAID,¹ who said that the expansion of condom use and education amongst sex workers remained a priority to control HIV in countries like India.

On the other hand, in the different conditions of Botswana, Stanton pointed out that rising condom usage had not had an impact on HIV prevalence. The country has concentrated heavily on rolling out national opt-out voluntary counselling and testing and distributing condoms as the main thrust of its HIV prevention strategy. In 1993, one million condoms were sold in the country and HIV prevalence in adults aged 16 to 45 was 18% in rural areas and 27% in urban centres. In 2001, over twice as many condoms were sold (three million), but HIV prevalence had doubled, to 30% in rural areas and 45% in urban centres.²

This may, in part, be because the other part of Botswana’s HIV strategy – the mass provision of antiretrovirals – has meant less attrition of the population due to AIDS.

Stanton attributed the failure of condom prevention efforts in this setting to the fact that it is very difficult to get people in long-term relationships to use them. In Botswana, there has been little change in the uptake of condoms by married women, and similar observations have been made in other countries. “When you get to married couples, it is difficult to get condoms in there,” he said. “Condoms have a hard time crossing the marriage and the relationship barrier.”

In Uganda, HIV incidence and prevalence fell well before condoms started to be widely used. An analysis by Maria Wawer,⁹ co-lead investigator of the Rakai Cohort Study, found that HIV incidence peaked at around 30,000 cases a year in 1989, at which point the proportion of men reporting sex with a non-regular partner in the last year was 35% and the proportion of women 16%. No condoms were being socially-marketed in the country at this point.

HIV prevalence peaked in 1991, with 32% of women aged 20 to 24 in antenatal clinics found to be HIV-positive. That year, one million condoms were socially marketed in the country, up from zero in 1989. By 1995, HIV incidence had fallen to 7500 a year, HIV prevalence to 20% and the proportion of men reporting sex with a non-regular partner in the last year was 15% and the proportion of women 6%. Ten million condoms were socially marketed. By 2001, 23 million condoms were being sold and HIV prevalence was down to about 8%, with HIV incidence at about 5000 a year.

But the graphs of incidence and prevalence make it clear that incidence started dropping precipitously before a single condom was socially marketed (it dropped threefold in a single year, 1990) and the decline in HIV prevalence started before widespread social marketing. The decline in prevalence is largely due to mortality from AIDS and a shrinking pool of HIV-positive people remaining to infect others.

The sudden decline in incidence, however, may also be at least partly due to the ‘Zero Grazing’ campaign, grassroots community anxiety about AIDS, and resultant partner reduction.¹⁰ Researchers for the Rakai cohort study have found evidence that, in a climate where sexual norms are relaxing again and the average number of partners is slowly creeping up, increased condom use seems to be containing incidence at that steady figure of about 5000 or 1.5% a year.

There is a general perception that condom use amongst high-risk groups in the developed world, especially men who have sex with men, has declined since combination antiretroviral therapy became available in 1995-97. The evidence, however, is conflicting and depends what measures are taken of risk behaviour and how precisely the question is asked: whether condom use, unprotected sex, or unprotected sex with definitely or possible serodiscordant partners is counted, and whether the question is asked positively (“Have you always used a condom during intercourse in the last year?” or negatively (“Have you had intercourse without a condom over the last year?”).

In the UK, according to one early survey by Sigma Research, while about 90% of gay men consistently did not use condoms for anal intercourse before 1980, the proportion of ‘rarely or never’ users was down to 22% by 1988.¹¹

After the advent of antiretroviral therapy, however, the proportion of gay men who sometimes did not use condoms started to increase. In their Gay Men’s Sex Survey for 2000,¹² Sigma feature a synthesis of two series of condom use surveys (both, importantly, restricted to London): their own proto-Gay Men’s Sex surveys, conducted at Gay Pride festivals between 1993 and 1995,¹³ and between 1996 and 1999, a series of surveys of gay men in commercial venues in London conducted by Julie Dodds and her team from the Royal Free & University College Medical School.¹⁴

This analysis needs to be prefaced by the caveat that the two series of surveys had different methodologies and sampled different populations; nonetheless, taken together they show that the percentage of men who did not consistently use condoms – in other words who had had at least one episode of unprotected sex during the previous year – remained stable at about 33% between 1993 and 1996 but then increased linearly to 46% by 1999, a 40% increase (see graph).

Note that the one-third of men who “did not consistently use” condoms is a different quantity from the one-in-five (22%) figure quoted above, which is the figure for the proportion who “consistently did not use” (i.e. never or rarely used) condoms.

Is this increase in unprotected anal sex amongst gay men during this period mirrored in other surveys? Certainly: a survey in San Francisco, for instance,¹⁵ almost exactly mirrored these findings: it found that the proportion of gay men reporting that they did not always use condoms increased from 31% in 1994 to 47% in 1999.

According to the subsequent Gay Men’s Sex Surveys conducted by Sigma Research, however, the rate of increase of unprotected anal sex slowed down considerably. Because the Gay Men’s Sex Surveys have not, over the years, asked precisely the same questions about condom use of precisely the same groups of men (some surveys included men who had not had sex over the previous year, but intended to) it is difficult to provide a consistent set of figures for the proportion of gay men in these surveys who had unprotected sex. However, by analysing the figures a reasonable approximation for the proportion of men who did not use condoms every time over the previous year can be ascertained.

The proportion of all respondents who had at least one episode of unprotected anal sex increased from 42.3% in 2000 to 48.8% two years later, but after this it hardly changed, while the proportion amongst men who had anal sex (excluding ones who had none) Increased from 45% in 2000 to 51.3% in 2002 but after this showed little if any increase.

After 2002, realising that unprotected anal intercourse (UAI) in itself does not imply HIV transmission, Sigma stopped asking simple questions about condom use and started asking whether unprotected intercourse involved serodiscordant partners or not; and/or whether partners were regular or causal; and/or which role (insertive or receptive) men took. In the most recent Gay Men’s Sex Survey report (data collected in 2007) 53% of those who had had receptive anal sex and 52% who had had insertive anal sex had not ‘always’ used a condom in the last year.

In a similar way, a 2004 survey from San Francisco¹⁶ found some evidence of a levelling-off of unprotected sex after 2001. More importantly, the amount of unprotected sex which was or could potentially be between partners of different HIV status (serodiscordant) showed a more distinct decline since that date.

Figure 2 above, from successive Gay Men’s Sex Surveys, shows the proportion of men who had anal sex without a condom - regardless of HIV status, relationship status or sex role - at least once during the previous year, both as a proportion of the men who had any sexual contact with another man at all and as the proportion of men who had anal sex. (The year 2004 is missing because few quantitative questions about sexual behaviour were asked that year.)

While the proportion of gay men who do not maintain 100% condom use has increased slightly and has certainly not decreased, what is perhaps more striking it how little it has changed in a decade, from about 50 to 55% of the proportion of men who have anal sex. The proportion amongst men with any kind of sexual contact with men during the year has increased slightly more, from 42 to 50%, because during this time the percentage of men who have had anal sex at least once in the year has slightly increased, from 89 to 95%.

Interestingly, an internet survey of comparable size, hosted by the website www.fridae.com, which looked at the sexual behaviour of 8000 gay men mainly living in Asia, also found that about half its respondents (46%) inconsistently or never used condoms. (See www.msmsexsurvey.com. As with the Sigma survey, the Fridae survey has now received funding to conduct a large pan-Asian survey, which should provide a lot more information about sexual risk behaviours in a multi-country sample of gay men.)

The relative lack of movement in the proportion of men practising safer sex consistently may suggest that in the absence of strong motivators to use condoms, such as the visible death from AIDS of peers, or the removal of factors that make condom use impossible, such as restrictions on supply, condom use is a relatively unchanging behaviour which is difficult to strongly influence.

Are these trends in condom use mirrored in the general population? The only data we have for the UK general population are the one-a-decade National Surveys of Sexual Attitudes and Lifestyles (NATSAL), which were conducted in 1990 and 2000. A third NATSAL is underway now but results will not be published till 2013.

A comparison of the 1990 and 2000 NATSALs¹⁷ found that consistent condom use among the sexually active population as whole increased from 17 to 24% during the 1990s.

However the rate of STIs also increased during this time, and NATSAL found that the effect of increased condom use had been more than cancelled out by other demographic changes – to which the increased condom use was probably consequent.

• The mean number of lifetime partners had increased from 8.6 to 12.7 in men and from 3.7 to 6.5 in women.

• Concurrent relationships – which are an extremely important factor in the spread of STIs, and which are cited as an important contributing factor to the HIV rate in Africa – increased from 12 to 14% in men and from 5 to 8% in women.

• Age at first sex declined from 21 in women and 17 in men in the 1990 survey to 16 for both sexes in the 2000 survey.

• The proportion of British men who had a male partner increased from 3.6 to 5.4%.

This is a reminder that many other risk factors, some of them modifiable and others difficult or unethical to change, produce changes in the rates of STIs.

Sexually transmitted infection diagnoses are amongst the best indicator of sexual risk. They are usually better indicators than self-reported sexual behaviour, which is subject to ‘social desirability bias’, defective recall, and other confounders. Of the STIs, gonorrhoea is seen as the most reliable indicator of unprotected sex in the absence of HIV data, although because it is much more contagious infections are not so closely correlated with anal intercourse and therefore not concentrated so much amongst gay men.

Gonorrhoea diagnoses, which have been collected in England and Wales since 1925, show that the decade during which AIDS was both widespread and untreatable in the developed world – roughly 1985 to 1995 – marked a historic low point in diagnoses of STIs.¹⁸ Peaks occurred during the second world war and the years 1965-85. Gonorrhoea started to increase again in the late 1990s but reached a peak in 2003 and in the last few years appears to have reached a plateau, at levels higher than the mid 1990s but historically quite low.

What figures we have for HIV incidence follow this trend: there was a sharp fall in incidence in many populations after the first phase of the epidemic; it increased somewhat, in step with the introduction of antiretroviral combination therapy, in the late 1990s, but amongst most groups now appears to have reached a steady state.

HIV incidence increased among gay men in many areas. For instance, according to UNAIDS, HIV incidence (new diagnoses) in gay men increased from 0.6% a year in the late 1990s in Vancouver to 3.7% a year in 2000; from 1.16% in 1996 in Madrid to 2.16% in 2000.

In other areas, however, an increase in incidence was not noted. In the UK, for instance, annual HIV incidence amongst gay men attending STI clinics was 3.0% in 1995 and 2.45% in 2001. Within London it remained static at about 3% a year (3.02% in 1995 and 3.06% in 2001); outside London, it actually declined from 2.96% in 1995 to 0.97% in 2001, though this was not statistically significant.¹⁹

Similarly, in the San Francisco study cited above,¹⁵ despite an 87.5% increase in unsafe sex amongst gay men, annual HIV incidence amongst men seeking care at the STD clinic appeared to be declining: from 7.3% in 1994 to 4.7% in 1999, though this was not significant (p=0.26). At a non-clinical anonymous community testing site it increased from 2.2% in 1996 to 4.2% in 1999 and this was significant (p=0.015), but more recent survey²⁰ HIV incidence showed signs of a community-wide decline (see HIV treatment as preventionfor more on this).

There are multifactorial reasons why a decrease in condom use may not always lead to an increase in HIV incidence. Firstly, HIV incidence is very difficult to measure unless large populations are surveyed and sensitive incidence assays are used. An observed increase in the proportion of people who are diagnosed in early infection, for instance, may be due to more people coming forward for testing and testing more frequently, so that HIV infections are detected sooner.

Secondly, as we point out in Serosorting, sexual harm reduction and disclosure, an increase in unprotected sex is not necessarily an increase in unsafe sex, if people are preferentially choosing people of their own status as unprotected-sex partners. Even in the Dodds study cited about from the late 1990s,¹⁴ although the proportion of gay men who had unprotected sex increased from 33 to 46% between 1996 and 1999, the proportion of those who had unprotected anal sex with people of opposite or unknown HIV status – the true measure of unsafe sex – did not increase, and the proportion who (by self report) only had it with men of their own HIV status remained static at 38 to 42% of all the men who had unprotected sex.

Thirdly, even if there is an increase in the community prevalence of HIV, as long as there is also an increase in the proportion of people with HIV in that community who are on treatment and have an undetectable viral load, HIV incidence will not increase and may even start to decline.

In is important to add, however, that HIV incidence can start to increase in real terms in high-risk communities whenever any of these factors, which serve to hold back incidence despite increasing rates of unprotected sex, starts to have a weaker influence.

Recently, for instance, rising HIV incidence has been noted in young gay men in Amsterdam.²¹

The Amsterdam Cohort Study is a longitudinal cohort study of HIV infection among a group of initially HIV-negative gay men, recruited at regular intervals ever since 1984, at an average age of 29, and tested for HIV every six months thereafter. For this study, the average length of follow-up was six years.

Out of 1627 men included in this study, 215 acquired HIV during follow-up. Incidence was calculated by assuming that the time of HIV infection was midway between a positive test result and the previous negative test. HIV incidence was initially 7.4% a year in 1985 but fell to 1.3% by 1990 and stayed at that level until around 1995-97. Since then it has slowly increased to 2% a year, though the increase does not pass the test of statistical significance (p = 0.1).

However, among gay men aged 30 and under, annual HIV incidence, which had been 0.9% in 1997, increased to 3.8% by July 2009. This result was due to a sudden doubling in the level in 2009 (it had been about 2% in 2004-8) and the increase observed is statistically significant (p = <0.01).

Sexual risk behaviour mirrored the incidence pattern, with an increase in the proportion of men reporting unprotected sex with casual partners from 12% in 1992 to 30% in 2008.

The strongest factors associated with becoming HIV positive were unprotected sex with casual partners (relative risk [RR] 4.74), having over five partners (RR 2.5), having gonorrhoea (RR 5.76), and having had no tertiary education (RR 2.11). It was estimated that three-quarters of new infections were due to sex with casual partners and only a quarter in steady partners.

However, among older gay men, unprotected sex with steady partners was more significant, and three times as many infections were acquired from primary partners in over-50 year olds between 2003 and 2008 than in the late 1980s. The incidence pattern in gay men aged 25 and gay men aged 50 were mirror images of each other: for younger gay men, the year 1995 featured the maximum risk of infection by a steady partner, but by 2009 this risk had declined considerably. Conversely, in 50-year-olds the maximum risk from a steady partner was reached in 2005.

The researcher commented that the Netherlands had a lower HIV testing rate than many other developed countries: although 88% of sexual health clinic attendees test for HIV, but rates are much lower in gay men who do not attend sexual health clinics.

Condoms as a method both of contraception and of reducing the transmission of HIV have always been controversial. Opponents of condoms, when not simply against their use for religious reasons, consistently suggest (either explicitly or implicitly) that the free or subsidised provision of condoms will simply encourage people, particularly young people, to have sex and take sexual risks they might not otherwise have done. We started this chapter with the remarks of Pope Benedict XVI to this effect.

There are few studies that support this view: however, a study from Makerere University in Uganda did suggest that in some circumstances condom provision can increase rather than decrease HIV risk, and has been cited by condom opponents.²²

The researchers took 378 young men aged 18 to 30 from two urban communities near Kampala and split them randomly into two groups. One group attended a three-hour workshop teaching them about how condoms stopped HIV and STIs, how to put on a condom, strategies for negotiating condom use with partners and talking about barriers to having safer sex. They were then given vouchers to redeem for free condoms.

The control group was just given a general lecture on the HIV situation in Uganda and given the free vouchers but no condom tuition.

The men taught how to use condoms certainly used more – 110 per man in the six-month period after the study compared with 13 per man in the control group.

However, the public health benefit of this may have been offset by the fact that they had an increased number of partners, whereas the control group reduced their partners.

Men taught condom use increased their average number of partners from 2.13 to 2.44 in the six months whereas the control group decreased their number of partners from 2.20 to 2.03. This was highly statistically significant.

The control group ended up having fewer regular and casual partners; the condom group reduced their number of casual partners slightly but had considerably more regular partners.

This would not matter if condom use was consistent; but while the amount of unprotected sex the control group had was reduced with both regular and casual partners, the condom group only reduced unprotected sex with all partners slightly and actually slightly increased the amount of unprotected sex they had with casual partners.

After adjusting for the fact that men in the condom group were on the whole somewhat older and more likely to be married, the researchers calculated that providing the men with condom lessons actually led to them having 48% more unprotected sex relative to the ones without lessons.

The study had many limitations. It was small and the two groups compared were not identical. And of course it does not show that providing condoms makes you have unsafe sex. But it may demonstrate that condom provision in the absence of other measures encouraging behaviour change is an incomplete answer to HIV transmission, at least in the African situation.

Meanwhile, however, despite all the above reservations and complexities of behaviour, it is important to remember that, aside from the provision of effective antiretroviral therapy, condoms remain the most effective and most widely used single HIV prevention method among sexually active people,and that, in high-risk populations where condoms are widely available, half of all acts of sexual intercourse take place with a condom.

It is also important to know, as we said above, that properly implemented HIV prevention programmeshave been shown to consistently increase condom use. For more on this see Behaviour change.

The degree to which people have taken up the use of condoms during the HIV epidemic varies hugely according to a whole number of different factors. These include the following:

• The degree to which people know that an activity carries an HIV transmission risk.
• Whether they think their partner is likely to have HIV.
• Whether they are having casual sex, commercial sex or are in a steady relationship.
• Whether they are HIV-positive themselves.
• Whether risky sex is linked to using alcohol or drugs, which impair people’s ability to make healthy decisions.
• Whether their behaviour is influenced by chronic mental health problems such as depression, low self-esteem or learned behaviour due to sexual abuse.
• Whether they are in a position to insist on the use of condoms.

The main factors that influence condom use (other than drugs and mental health) can be illustrated by taking a selection of condom usage figures from different population groups and exploring the differences between them.

The same pattern applies in both heterosexual men and gay men: men in steady relationships, whether of the same HIV status or not, are far less likely to use condoms. This gives us a clue as to the primary psychological driver behind unsafe sex and the decision to use, or not to use, condoms.

Take two examples. A study from the Netherlands in 2000³⁰

A London study³¹ stratified the same results by HIV status of participants. It found that, in HIV-negative men, 28.5% had unprotected sex within relationships but only 5% with casual partners. HIV-positive men, by contrast, were just as likely to have unprotected sex with regular and casual partners (22.2 vs 20.6%). The researchers argued that HIV-negative men cannot be sure of the HIV status of partners without mutual testing.  HIV-positive men, on the other hand, can find out their partners’ HIV status by the simple act of mutual disclosure. However it was not ascertained whether disclosure was responsible for higher rates of casual unprotected sex in positive men.

We will look at evidence like this in the next section to understand how gay men are using disclosure to minimise HIV transmission risk. For the time being we are looking at the psychology of what condoms symbolise and why they tend not to get used in primary relationships.

An interesting insight into this was provided by a study from New York³² which examined whether HIV-positive women had safer sex and if so, whether they did so more often in primary relationships. The authors hypothesised that women would be more likely to maintain condom use in steady relationships in order to protect partners.

They found the opposite to be the case. Forty-six per cent of women maintained condom use in all sex (in this study oral sex without a condom was counted as ‘unprotected’). But of the remainder, 61% had had at least one episode of unprotected sex in the past 90 days with a steady partner, compared with 16% who had done it with a casual partner. Women in steady relationships were three times more likely to have unprotected sex with a steady than with a casual partner.

Was this because steady partners were more likely to be known to be HIV-positive themselves? No, because unprotected sex was just as common with HIV-negative male partners as HIV-positive ones.

On further investigation, condom use had a bipolar distribution. Condom use was significantly higher in women who had casual partners – but also within the most committed relationships, when these were defined by length (over one year), by being within a legalised marriage, or by partners living together. Condom use was a lot lower with primary partners who were new or who did not live with the women.

The researchers theorised: “Women in our study who were married and in the longest, most supportive relationships may have possessed the power to broach or insist upon consistent condom use.”

Conversely, they add: “Perhaps in [more recently established] steady partnerships, condom use implies, not primarily protection, but mistrust, suspicion, lack of emotional and physical intimacy, or denial of potential motherhood.”

A 2004 study of 78 HIV-positive gay men in London³⁴ found that, while 38% of the men reported some degree of erectile dysfunction, this went up to 51% in the context of trying to use condoms. In other words, more than half of the men experienced difficulty in getting or sustaining an erection when trying to put a condom on. Furthermore, 90% of the 37 men whose erectile dysfunction was associated with condom use reported inconsistent condom use during insertive sex, compared with 28% of those who did not report condom-related erectile dysfunction.

Subsequent studies have strengthened the evidence that performance anxiety and fear of impotence may be important drivers of men’s reluctance to use condoms.

A US-study of 278 young (average age 23.7) heterosexual men attending an STI clinic, for example,³⁵ found that men who reported at least one erection loss, either while putting on condoms or during sex with a condom, in the previous three months had 50% more episodes of unprotected sex during that period (an average of 10.6 episodes, versus 7.0 in men with no erection difficulties). In addition, 40.8% of men who’d experienced erection failure also reported removing condoms before sex was over – twice as many as the 21.3% of men with no erection problems. Erection loss, even in this young group, was common, with 28.1% reporting one experience of erection failure in the three most recent episodes of sex.

One of the other predictors of erection loss was, importantly, having multiple partners. Forty-five per cent of men reporting sex with three or more partners during the previous three months reported erection loss, compared with 30% of those with one or two partners. This suggests a synergy between multiple partners, erectile dysfunction and unprotected sex. Researcher Cynthia Graham suggested that men were more likely to be nervous with new partners, which would make them reluctant to use condoms, even though they were the very people with whom they should be using them.

‘Once bitten, twice shy’ also seemed to apply to gay men when it came to experiencing erection loss during condom use, according to a study presented to the Mexico International AIDS Conference in 2008.³⁶ If gay men experience erection loss when they use condoms for insertive anal sex, they are both less likely to use condoms over the following six months, and less likely to intend to use condoms.

The researchers looked at the relationship between ‘COINED’ (COndom INduced Erectile Dysfunction, defined as loss of erection because of condom use) and deliberately risky sexual behaviour (unprotected insertive anal intercourse, or UIAI, that was planned in advance). They recruited 435 men, 6% of them HIV-positive, taking part in the Amsterdam Cohort Studies of Homosexual Men. Rates of UIAI were 18% and 17% with casual partners and steady partners of discordant or unknown HIV status respectively.

COINED was not associated with UIAI with steady partners. But among casual partners, COINED was significantly associated with deliberate UIAI (odds ratio of 6.57) but not with non-deliberate UIAI. The researchers conclude that COINED is a unique predictor of deliberate UAI between casual partners.

If men experienced COINED they were 2.71 times more likely, six months later, to have risky UIAI (p = <0.05) with casual partners. Men who had experienced COINED were 63% more likely to have UAI over the next six months, and 59% more likely to intend not to use condoms over the following six months, indicating that COINED could be considered as a surrogate measure or predictor of deliberately unprotected sex. HIV status had no relation either to COINED or to deliberately planned unprotected sex.

The researchers suggested that prescription of erectile medication like sildenafil (Viagra) could be an appropriate intervention. Erectile dysfunction drugs have been found, in a number of studies, to be associated with multiple partners, unsafe sex, and HIV seroconversion – but this is probably because Viagra and other erectile dysfunction drugs are often used by gay men who practice sex with multiple partners and take other recreational drugs – whose effects on erection may necessitate the use of Viagra.

If the associations with recreational drugs are ruled out, erectile dysfunction drugs in themselves do not appear to be associated with lower condom usage rates, though one study found that they were associated with more condom breakage.

In this study,³⁷ the researchers used newspaper and internet advertisements to recruit a convenience sample of 440 men who used condoms for vaginal sex.

Just under one in ten of the men had used an erectile dysfunction drug the last time they had sex. For 12% of these men the condom broke, compared to 5% of men who didn’t use an erectile dysfunction drug and, in multivariate analysis, erectile dysfunction drug use remained significant - users were four times more likely to report condom breakage (adjusted odds ratio 4.04, 95% confidence interval 1.06 - 15.41).

However, the same team of researchers published a separate study the same year which did not find the same association.³⁸ A different group of 705 men were recruited via a website selling sexual paraphernalia. Although most men were heterosexual, some were reporting on anal sex. Men who used recreational drugs were excluded. Seven per cent of men reported using an erectile dysfunction drug the last time they had sex.

Condom breakage was rarely reported in this sample, and was no more commonly reported by erectile dysfunction drug users. On the other hand, users of erectile dysfunction drugs were actually more likely to report the condom slipping off or erection problems during sex. The researchers concluded from this that, while Viagra and similar drugs may improve men’s erections, they do not always eliminate all erection problems, especially those associated with condom use.

These studies underline that what has been called ‘fit and feel’ may be crucial when it comes to men’s decisions as to whether to continue to use condoms or not. Failure to supply condoms that fit varying sizes of penis, for instance, may make a difference.

One study found that, although most men do not have problems with condom fit and feel, men with shorter or larger penile dimensions (length and circumference) were more likely to have negative attitudes.³⁹

The Centre for Sexual Health Promotion at Indiana University conducted a study among 1661 men from seven US states. Data were collected on condom fit and feel through self-reporting by the men, who took the measurements using a paper-based measuring device that ranged from 0 to 23cm for length and 0 to 19cm for circumference.

Participants were broadly grouped into three categories of short, medium and large penile sizes. For the penile length, participants categorised as “shorter” had a mean penis length of 11cm (range 4 to 12cm); “medium” a mean length of 14cm (range 13 to 15cm); and “longer” a mean penile length of 18cm (range 16 to 26cm).

For penile circumference, men in the “smaller” category had a mean measurement of 9cm (range 3 to 11cm); the “medium” group had a mean measurement of 13cm (range 12 to 13cm); and the “larger” group had a mean measurement of 14cm (range 14 to 19cm). The mean age of participants was 29 years.

Considering the variables separately, men in the “longer” group were more likely to describe condoms as too short (p = .005), not rolling down enough and feeling too tight along the penile shaft (p = 0.002) than those in “smaller” or “medium” groups.

On the other hand, men in the “shorter” group reported unrolled condom at the base of the penis.

As regards circumference, men in the “larger” group described condoms as being too short and too tight particularly around the glans (head of the penis) and the penis base. When the two variables were combined, shorter penile lengths with smaller circumferences were predictive indicators of condoms being too loose (p = 0.047) or too long (p = 0.004) whereas longer penile lengths combined with larger circumferences were predictive of condoms being reported as too tight and too short (p = 0.001).

Overall, the majority of men did not complain about condom fit and feel. However, results show that those with complaints about these issues had negative perceptions of condom fit and feel (p = 0.024).

Condomi and Pasante manufacture larger, wider condoms that can be bought online in Europe, and Pasante also manufactures a slightly shorter and tighter than average condom. In the United States Durex and Trojan market large condoms that are widely available both online and through pharmacies (these products are also available online in Europe). Durex and Mates manufacture smaller condoms that are widely available through online retailers.

However, the choice of condoms tends to be much more limited when they are supplied free of charge in developing countries, and more work is needed to identify what range of condoms needs to be supplied in different settings in order to increase men’s comfort and satisfaction with using condoms.

Unsatisfactory ’fit and feel’ was also associated with condom breakage. In a second study⁴⁰ of the same group of heterosexual young men cited above,³⁵ condom breakage was twice as likely to be reported by men who felt their condoms did not fit properly. Other predictors of condom breakage were, unsurprisingly, letting sharp objects near the condom and previous experiences of condom slippage: STIs were also significantly associated with condom breakage, showing this is a real risk for HIV.

The vast majority of male condoms are made of latex. However there are condoms designed to be hypoallergenic. These include Durex Avanti, which is made of polyurethane, and Durex Allergy, made out of hypoallergenic latex. Female condoms are generally made of hypoallergenic plastic rather than latex.