Introduction

Sexual transmission is now the most common method of HIV infection in Spain (Belza et al. 2006; National Center for Epidemiology / Carlos III Institute of Public Health 2014; Secretariat for the National AIDS Plan 2008). Furthermore, it has been observed that the rates of HIV infections are stabilizing instead of decreasing regarding heterosexual transmission: 30.5 % in 2011, 29.5 % in 2012 and 30.1 % in 2013 (National Center for Epidemiology / Carlos III Institute of Public Health 2014), which demonstrates the need for more efforts in its prevention.

It has been clearly shown that condom use in heterosexuals couples is less common among primary, steady or main partners, compared with other types of partners, such as casual or concurrent (Galavotti et al. 1995; Grimley et al. 1995; Lauby et al. 1998; Morrison-Beedy and Lewis 2001; Santelli et al. 1996). Similarly, perceptions about the frequency with which adolescents use condoms also varies by partner type, such that adolescents reported a lower frequency of condom use with steady partners compared with other type of partners (Ellen et al. 1996). In line with this, young Spaniards use condoms less consistently with a romantic partner than with an occasional partner (Faílde et al. 2008) and a lower level of condom use has also been found with steady partners than with occasional partners (Ballester et al. 2009). Thus, a more detailed knowledge about the factors related to condom use with a primary, steady or main romantic heterosexual partner could be an important tool to help develop better prevention programmes and campaigns.

The Transtheoretical Model of change offers a useful framework to understand why people engage (or not) in preventative behaviours (Grimley et al. 1997; Prochaska and DiClemente 1983; Prochaska et al. 1992). One of the central constructs from the Transtheoretical Model is the concept of decisional balance, which is based on the conflict model of decision making (Janis and Mann 1977). Decisional balance involves weighing up the pros and the cons of engaging in a particular behaviour.

The relationship between a health-related behaviour and their perceived pros and cons was initially explored in smoking cessation (Velicer et al. 1985). It has also been analysed in a wider context, with both the cessation of risky behaviours and the commencement of preventive behaviours, including condom use (Prochaska et al. 1994).

Limiting ourselves to the literature concerning heterosexual partners, several studies have been carried out with different types of samples, such as: college students (Grimley et al. 1993, 1995; Kwon et al. 2008; Parsons et al. 2000); women at high risk of HIV (Galavotti et al. 1995; Lauby et al. 1998); and single urban women (Morrison-Beedy and Lewis 2001; Morrison-Beedy et al. 2002), in order to examine the relationship among the decisional balance constructs and condom use. This research has consistently demonstrated that the pros are related to condom use (Galavotti et al. 1995; Grimley et al. 1993, 1995; Kwon et al. 2008; Lauby et al. 1998; Morrison-Beedy et al. 2002; Morrison-Beedy and Lewis 2001; Parsons et al. 2000). Thus, people who consistently use condoms perceive higher pros of their use than those who do not. On the other hand, although the results have not been as consistent regarding cons (Grimley et al. 1993, 1995; Kwon et al. 2008), the reverse pattern has generally been reported (Galavotti et al. 1995; Kwon et al. 2008; Lauby et al. 1998; Morrison-Beedy and Lewis 2001; Parsons et al. 2000).

Nevertheless, the majority of the studies within this theoretical framework have focused on the influence of the perceived pros and cons in general, but they have not examined whether any pros or cons are more important than others in determining condom use. In a study conducted in Taiwan, significant differences, by stage of change, were found in a number of the advantages and disadvantages of condom use among undergraduate students (Tung et al. 2009). However, unfortunately they only assessed a small number of pros and cons and they did not specify partner type. Therefore, it is interesting to investigate whether there are particular advantages and disadvantages which are more influential in determining condom use with a heterosexual romantic partner. Thus, once identified, these influential advantages and disadvantages could be used to encourage condom use.

In addition, the relationship between condom use and a number of demographic and descriptive variables, as well as relationship length, have been found in several studies. In particular, among young people, consistent condom use decreases with age (Grimley et al. 1997; Planes et al. 2012; Pleck et al. 1991). Similarly, lower condom use has been found for those in longer relationships (Aalsma et al. 2006; Bankole et al. 1999; Brady et al. 2009; Corneille et al. 2008; Polacsek et al. 1999; Sturdevant et al. 2001; Xiao et al. 2010).

The goals of the present study were:

  1. 1.

    To measure the frequency of condom use, by gender, with a heterosexual romantic partner in the last month and its use the last time they had sexual intercourse.

  2. 2.

    To investigate whether there are significant differences in the pros and cons of condom use for those who use condoms consistently and those who do not.

  3. 3.

    To examine whether the pros and cons of condom use are able to predict condom use, after controlling for age and relationship length, and whether the predictors were different for men and women.

  4. 4.

    To identify which advantages and disadvantages predicted consistent condom use in the last month with a heterosexual romantic partner (separately for men and women), after controlling for age and relationship length.

Methods

Participants

Random sampling within strata was used for participant selection. Strata criterion was the field of knowledge: (1) humanities and social sciences; (2) natural sciences and health; (3) technical studies. A third of the degrees in each stratified sample were randomly selected.

The sample was comprised of 1502 undergraduate students from the University of Girona, with the majority being female (53.4 %) and the average age was 21.2 (4.73). From this sample, only those who a) were 25 years or less, b) identified themselves being heterosexual or bisexual, c) had had sexual intercourse, d) were not attempting to get pregnant, and e) had a heterosexual partner were analysed. The final sample consisted of 619 students (64.9 % females). The mean age of men was 20.75 (S.D. = 2.01) and 20.08 (S.D. = 1.71) for women. The vast majority identified themselves as heterosexual (97.9 %). The mean relationship length (in months) was 26.5 (S.D. = 21.22).

Measures

The data were collected using a self-administered questionnaire. The following brief story of a heterosexual romantic couple was presented: Charles and Eve have been dating for 6 months. They feel a lot of attraction for each other and are very happy with their relationship. They even make future plans. Since they began to have sexual relationships they have almost always been using condoms. They also used one frequently with their former partners.

Participants were then asked to identify themselves with the character of their own gender. Next, they were asked to rate how important each item was when deciding whether or not to use condoms on a 5-point likert scale (from 1 = not important to 5 = extremely important). The list of items to be rated was randomly ordered and based upon CIVIUP (Planes et al. 2012).

Although some items were dropped from the scales, in order to increase their internal consistency, when the predictive capability of the individual items was investigated, all items were analysed. The items which were not included in the scales are indicated in Table 1. The Cronbach’s alpha coefficients were .58 for the pros scale and .72 for the cons scale. The sum of the pros and cons scales were transformed into Z scores (M = 0; SD = 1). More thorough analyses of the reliability and construct validity of the scales are reported elsewhere (Prat et al. 2012).

Table 1 Item means (SD) for the pros and cons of condom use with a heterosexual romantic partner
Full size table

The questionnaire included questions regarding gender, age, sexual orientation, whether they had previously engaged in sexual intercourse, whether they had a romantic partner, whether they were actively attempting to get pregnant, the length of their current relationship, the frequency of condom use in the last month with their romantic partner (never, hardly ever, sometimes, almost always and always), and whether they had used a condom the last time they had intercourse with their romantic partner. The use of short reporting periods, such as a month, has been recommended for assessing the frequency of sexual behaviours (Catania et al. 1990a, b). Other measures regarding intentions to use condoms in the future and how long they had been using condoms consistently were also included.

According to previous research (Brady et al. 2009; Manlove et al. 2008) and clinical recommendations (Medical Institute for Sexual Health 2012), condom use during the last month was collapsed into two categories for further analyses (consistent users and inconsistent users), Those who used condoms every time they had sex in the last month constituted the consistent condom users, whereas those who used it never, hardly ever, sometimes or almost always were categorized as inconsistent users. The rationale behind this decision was that those who had used condoms less than always may have exposed themselves to the risk of contracting HIV and other sexually transmitted infections. In other words, those in the inconsistent group had all engaged in unsafe sex in the last month.

Procedures

Firstly, permission from the coordinators and professors of the chosen degrees was needed in order to carry out the study. Students who were in their first, second and third years were asked to fill out the questionnaire in class. The questionnaire was completely voluntary and the students were assured of anonymity and confidentiality of their data. Participants received no compensation for their participation in the study. Almost all students who were asked agreed to complete the questionnaire. The questionnaire took about 10–15 min to complete and data were collected from March to May in 2010.

Data Analysis

T-tests were used to analyse gender-related difference on the pro and con items ratings, as well as to compare consistent and inconsistent condom users, and compare those who used a condom the last time they had sexual intercourse with those who did not. Chi-square tests were performed to explore the relationship between gender and consistent condom use in the last month and the last time they had sexual intercourse. Two hierarchical logistic regressions (one using the scores in the pros and cons and the other using the whole pool of individual items) were performed for each gender to predict consistent condom use in the last month. In the first step, age and relationship length were simultaneously entered using the “enter” method. In the second step, the pros and cons scale scores were also simultaneously entered in the case of the first equation, whereas all the individual items were entered using the forward Wald stepwise procedure in the case of the second equation.

Results

The item means and standard deviations are shown in Table 1. For both males and females, all means for the pros items were above the midpoint, as were the majority of the con item means. Males and females both rated protection: against HIV, other sexually transmitted infections and pregnancies as the most important pros. With respect to the cons, both genders rated needing their partner’s agreement to use a condom as the most important con. This was followed by the reduction of pleasure for their gender (i.e. themselves) and for their partner. In other words, the reduction in their own pleasure was rated as the second most important con by both genders, while the third was the reduction of their partner’s pleasure. The fourth most important con, for both genders, was the interruption and cooling off from foreplay caused by condom use. There were also a number of statistically significant gender differences in the items ratings. Regarding the pros, females deemed protection against sexually transmitted infections to be more important than males. Females also rated more highly the pros that using condoms demonstrates that they are a responsible person and that they are interested in their partners’ health, as well as feeling safe during and after sexual intercourse. Regarding the cons, higher ratings of importance were given by females to the difficulty in keeping an erection when using condoms and the need for their partner’s agreement to use them. In contrast males deemed the reduction of pleasure for males to be more important than females did.

Condom Use by Gender

Table 2 shows the frequency of condom use in the last month with their romantic partner, by gender. Less than half the sample (42.1 %) had used condoms every time they had sexual intercourse with their romantic partner during the last month. There were no gender differences in the frequency of condom use in the last month (Χ2 (4) = 7.29; p = .12). In terms of the last time the participants had engaged in sexual intercourse with a romantic partner, 54.4 % of the participants reported using a condom. Again there were no gender differences in condom use for their most recent engagement in sexual intercourse (Χ2 (1) = 2.26; p = .13).

Table 2 Frequency of condom use with a heterosexual romantic partner in the last month, by gender (%)
Full size table

Pros and Cons of Condom Use in the Last Month and During their Last Intercourse

Consistent condom users reported higher pros (t = −4.86; p < .001) and lower cons (t = 3.48; p < .005) than inconsistent condom users. A similar pattern was found for those who used a condom the last time they had sexual intercourse, in that those who used a condom reported significantly higher pros (t = −4.96; p < .001) and lower cons (t = 3.79; p < .001) than those who did not use a condom. The pros mean for consistent condom users was 40.77 and the cons mean was 22.18, whereas they were 39.29 and 24.04, respectively, for inconsistent condom users. Likewise, the pros and the cons means were 40.64 and 22.33 for those who used a condom the last time they had sexual intercourse, whereas these means were 39.07 and 24.34 for those did not use one.

Pros and Cons Scores as Predictors of Consistent Condom Use

The results of the hierarchic logistic regression for predicting consistent condom use are presented in Table 3. This shows that age was a significant predictor of condom use for both males and females, as age increased condom use decreased. In contrast, relationship length did not make a significant contribution to the prediction of consistent condom use in females or males. Both the pros and cons were significant predictors for being a consistent condom user among females, after controlling for the effects of age and relationship length, whereas only the pros were a significant predictor in males. Nevertheless, among both genders the pros were more strongly associated with being a consistent condom user than were the cons. Men who scored one standard deviation above the pros mean were 97.4 % more likely to be a consistent condom user, while for cons those who scored one standard deviation below the mean were 15.1 % less likely to be a consistent user. In women, these percentages were 60.7 and 36.8 %, respectively.

Table 3 Hierarchical logistic regressions for predicting consistent condom use with a heterosexual romantic partner in males and females
Full size table

Pros and Cons Items as Predictors of Consistent Condom Use

The results of using the individual advantages and disadvantages of using condoms to predict their consistent use, after controlling for age and relationship length, are presented in Table 4. Among males, the only item which made a significant contribution to the prediction of consistent condom use was that they feel safe during and after sexual intercourse, as it was the only item retained in the model. Among females, however, there were four significant predictors. The only advantage retained in the model was that the condom gives them a contraceptive method without any side-effects, whereas the disadvantages were that: condoms diminish the sensations of pleasure, that it interrupts and cools off foreplay and that they find it difficult to ask for condom use when they are already using the pill. As the perceived importance of advantages increased, participants were more likely to end up being consistent condom users, while the converse were true for the disadvantages of condom use with the exception of the item “It’s difficult to use a condom when you are already using the pill”. In this case, those who perceived it to be more important were more likely to be consistent users.

Table 4 Hierarchical logistic regressions for predicting consistent condom use with a heterosexual romantic partner using the individual pros and cons items
Full size table

Discussion

In this sample, a third of the participants had never used condoms with their current romantic partner in the last month and nearly a quarter of participants did not use condoms consistently. Just over two fifths had used condoms consistently. Surprisingly, the majority of the participants reported having used a condom the last time they engaged in sexual intercourse. Thus, assessing condom use through its use during the last time they had sexual intercourse appears to result in the overestimation of this preventive behaviour, at least with this type of sexual partner.

In the present study condom use during the last engagement in sexual intercourse was lower than that reported by Spanish undergraduate students, when they were not asked to distinguish between partner type (Belza et al. 2006). Therefore, the dissimilar finding in the present study could be explained by the fact that this study specified partner type, while previous research did not. Moreover, in contrast to Belza et al. (2006), who found higher condom use among men, no gender differences in condom use during the last experience of sexual intercourse were found in the present study. As can be seen, many young people do not engage in consistent condom use with their partner, which could place them at risk of contracting HIV and other sexually transmitted diseases.

Both genders gave very high ratings to the importance of the protection provided by condom use, the need to negotiate its use, as well as the reduction in pleasure that condom use involves. However, there were also gender differences, with females rating several pros and two cons as more important than did males, while males gave significantly more importance to one of the cons. The results regarding the pros are in line with those found by Grimley et al. (1995) and Parsons et al. (2000), who reported differences by gender at the scale level, with women generally attributing higher importance to the pros of using condoms. Nonetheless, they also reported that male college students rated the cons of condom use to be higher than females (Parsons et al. 2000; Grimley et al. 1995). Although we did find males placed a greater importance on the reduction of pleasure for men, females gave more importance to the need to negotiate condoms use and the difficulties in keeping an erection. In agreement with our research, some previous research has found that women report more agreement with several barriers to using condoms (Crosby et al. 2013). All in all, gender related-differences on views of condom use are not unusual, and research has also found female college students have more positive attitudes towards condom use (Muñoz-Silva et al. 2009; Sacco et al. 1993), which highlights the need for always taking into account gender when dealing with the issue both in research and prevention.

The pros and cons were both significantly related to engaging in condom use, as would be expected by the decisional balance theory (Janis and Mann 1977), as well as from the results of earlier studies on condom use (Galavotti et al. 1995; Grimley et al. 1995; Kwon et al. 2008; Lauby et al. 1998; Parsons et al. 2000; Planes et al. 2012) and research on other health-related behaviours (Prochaska et al. 1994). In this study, those who consistently used condoms perceived greater advantages and lower disadvantages of their use. More specifically, those who used a condom during their last engagement in sexual intercourse perceived greater advantages and lower disadvantages than those who did not.

Age was also shown to be a significant predictor of condom use, which is consistent with previous research (Grimley et al. 1997; Planes et al. 2012; Pleck et al. 1991). Nonetheless, unlike previous research (Aalsma et al. 2006; Bankole et al. 1999; Brady et al. 2009; Corneille et al. 2008; Polacsek et al. 1999; Sturdevant et al. 2001; Xiao et al. 2010) relationship length was not a significant predictor of condom use. This inconsistent finding could be explained by fact that the mean relationship length in this sample was greater than in most previous studies (Aalsma et al. 2006; Brady et al. 2009; Sturdevant et al. 2001), while parity in the rates of condom use between new partners and established partners appears to occur in a short period of time (Fortenberry et al. 2002). However, these previous studies were not based on college student samples, and relationship length may be differentially related to condom use in different sample types.

The pros and cons of condom use are significant predictors of consistent condom use in the last month among women, even after controlling for age and relationship length. This relationship was such that those women who perceived higher pros and lower cons were more likely to be consistent condom users. However, only the pros were a significant predictor among men. In that same vein, Noar et al. (2001) found that the cons were not related to condom use among American male college students. Moreover, in both genders the impact of the perceived pros was stronger than for the cons, as indicated by the odds ratio, with this difference being greater in men than for women. The greater importance of the pros has been reported in previous research (Grimley et al. 1993, 1995; Grossman et al. 2008; Morrison-Beedy et al. 2002).

Therefore, these results suggest that in order to promote condom use with a romantic partner, prevention campaigns and programmes should highlight the perceived pros and try to diminish the perceived cons of this preventive behaviour. Bearing in mind that the pros have repeatedly been shown to have a greater influence, pros should be the main focus for preventative actions, without ignoring the role of the cons. Despite this, messages directed towards prevention must refer specifically to the advantages of condom use, rather than just too general advantages or disadvantages.

When the predictive validity of the advantages and disadvantages were tested, different significant predictors emerged for women and men. Only one advantage was found to be a significant predictor among men (the feeling of safety during and after intercourse when using a condom). The more men perceived the safety sensation to be important, the more likely they were to report using condoms. Furthermore, those who rated this item one standard deviation above the mean were 207.3 % more likely to have used condoms consistently in the last month. It is interesting to point out that it was these safety feelings, rather than the protection against HIV and other STIs, that was predictive of consistent condom use with a romantic partner. This relationship was also found in another Spanish study conducted among adolescents (Lameiras et al. 2004).

However, the results for females were substantially different from those of men. Four individual items were found to be predictive of consistent condom use among women, over and above that predicted by age and relationship length. The first predictor was the reduction in the sensations of pleasure. Similarly, the reduction of sexual satisfaction associated with condom use was also found to predict inconsistent condom use among female Mexican college students (Robles et al. 2006). Furthermore, research has also found the loss of pleasure caused by using condoms to be a factor associated with non-use (Randolph et al. 2007). Although they found this to be the case for both genders, it was more markedly so among males. There is also research which reported diminished sexual arousal in safe sex to be more strongly related to engagement in unprotected sex among women (Higgins et al. 2009), which would be more in line with our finding. The second individual item was the only advantage that was a significant predictor of consistent condom use in the last month. Women who reported that a condom was a contraceptive method without side-effects were also more likely to be consistent condom users. The lack of side-effects has also been reported as one of the most important advantages in a previous sample of Spanish college students (Ballester et al. 2009). Moreover, this item was the strongest predictor among women, as shown by its odds ratio. Thus, those who scored one standard deviation over the mean on this item were 53.8 % more likely to be a consistent condom user. Another significant predictor was the degree to which women felt condoms interrupted foreplay, with those reporting a higher degree of interruption being less likely to be a consistent condom user. Lameiras et al. (2004) also found that females who agreed that condom use interfered with the romantic atmosphere were less likely to use condoms. Eroticising condom use has been suggested as an important component for safer sex interventions and this has had encouraging results. Therefore, this could be a suitable strategy to counteract the effects of this perception (Scott-Sheldon and Johnson 2006). The perceived difficulty of asking for condom use when the female is taking the pill was also a significant predictor. However, this relationship with consistency of condom use was in the opposite direction to that expected from theory (Janis and Mann 1977), so that those who reported greater importance of this disadvantage were more likely to be consistent users. This result could be explained by the fact that those who used condoms consistently were not also taking the pill. Thus, they would rate using both methods together as being more difficult. The figures concerning double method use would support this hypothesis (Daphne 2011; Berer 2006). However, it could also be that those who have experienced difficulties in asking for condom use, when they were already taking the pill, would be more aware of how difficult it is although they have achieved their purpose.

Notwithstanding that pros are more critical than cons, especially among males, when the effect of the specific advantages and disadvantages were analysed, more cost-related aspects were shown to be predictors of consistent condom use with romantic partners in women. As prevention messages urging this target population to use condoms consistently must refer to concrete advantages or disadvantages of their use, those aspects which have been shown to be predictors of consistent condom use for each gender should be used. Although finding different correlates of condom use according to gender is certainly not a novelty (see, for instance, Helweg-Larsen and Collins 1994), our results at the level of specific aspects associated with consistent condom use underscores the need for tailoring gender-based interventions for young adults engaged in romantic relationships.

Limitations

It should be pointed out that some of the participants may have had a decisional balance which was other than expected, because they may use or not use condoms consistently because of their partners influence. Moreover, although the content covered by the set of disadvantages and advantages used in the present research was wider than previous research, a number of important pros and cons may have been missed.

Social desirability may have affected participant’s responses because of the private and sensitive nature of sexual behaviour. Nevertheless, in order to promote honest responding, participants were informed that surveys were anonymous and voluntary, and the confidentiality of the data was assured. It may be argued that the story presented to the participants describing a romantic couple who almost always used condoms had a priming effect on participants’ responses. However, as always using condoms seems to be the socially desirability response, it seems unlikely that any respondent who actually always used it would have been primed to answer “almost always” due to the story, while the dichotomization of the variable would have resolved this problem for those who answered less than almost always. Another limitation of this research is that the results presented here may not be generalizable to young adults nor other populations of college students, since the selection strategy did not served that purpose but was intended for gathering data from both genders and ensuring diversity in the degree of study, since the issue being researched can be sensitive to area or study, particularly amongst health-related studies (e.g., nursing & psychology). Lastly, because of the cross-sectional design of the present study, causal relationships among the variables cannot be inferred.

Conclusions

The results of the present study show that the perceived pros of condom use are more important than the cons, which is in agreement with earlier research. However, in our view, identifying the advantages and disadvantages whose perceived importance predicts consistent condom use can also be useful for prevention purposes. To sum up, preventive programmes should give priority to the pros, while at the same time take into consideration the individual advantages and disadvantages that have been shown to be significant predictors of condom use. Likewise, it should be pointed out that the items which predicted consistent condom use were dependent on gender.