Predicting preparatory behaviours for condom use in female undergraduate students: a one-year follow-up study

Abstract

Summary

The objective of this study is to investigate whether the Theory of Planned Behaviour (i.e. attitude, subjective norm, perceived behavioural control and intention), fluctuations in motivation over time, and variables from the Prototype-Willingness Model (i.e. behavioural expectation and behavioural willingness to have unprotected sex) predict preparatory behaviours for condom use. Sixty-two female undergraduates completed baseline and one-year follow-up questionnaires. Having condoms at home and carrying condoms were predicted by behavioural willingness to have unsafe sex at baseline. Having bought condoms was predicted by the behavioural expectation to use condoms with new partners at baseline. Intention and fluctuations in motivation did not emerge as significant predictors of preparatory actions. Female undergraduates, who are more willing to have unprotected sex under risk-conducive circumstances, are also less likely to prepare adequately for condom use, and thereby increase their chances of encountering such situations. Overall, the findings are in support of the Prototype-Willingness Model.

Keywords

condom use preparatory behaviour young females prospective study behavioural willingness

INTRODUCTION

In Western society, the incidence of sexually transmitted infections (STI) is high in young adults and in young women in particular. For example, in the USA¹ as well as in the Netherlands² almost 50% of all newly diagnosed STIs are found in the age group of 15–25 years, and young females account for approximately 75% of the chlamydia and gonorrhoea infections among all females. Correspondingly, the number of sexual partners during adolescence has increased over the years, and many young females do not use condoms consistently.³ This is likely to be at least partly due to insufficient preparations for condom use, such as buying and carrying them.⁴ The focus of the present study is therefore on the prediction of these so-called preparatory behaviours for condom use.

One of the predominant theories in the field of condom use prediction is the Theory of Planned Behavior.⁵ It states that intention to perform a certain behaviour is the most proximal determinant of actual behaviour. Intention, in turn, is influenced by attitude towards the behaviour, the perceived norm of important others regarding whether or not one should perform the behaviour (i.e. subjective norm), and the extent to which one is convinced that one is capable of performing the behaviour (i.e. perceived behavioural control). The usefulness of this model has been largely supported,⁶ however, only approximately 20% of condom use can be ascribed to intention.⁷

The relative lack of correspondence between intention and behaviour may be due to the fact that intentions may vary over time, and that this affects their strength in predicting behaviour. Indeed, research indicates that intentions that remain stable over time have a far stronger effect on subsequent behaviour than intentions that fluctuate.⁸ Furthermore, temporal stability of intention has been shown to mediate the effect of other moderators, such as past behaviour, on behaviour.⁹ Therefore, in the present study, next to constructs from the Theory of Planned Behaviour, fluctuations in intention over time will be assessed.

In addition, from the Prototype-Willingness Model,^10,11 it is argued that ‘behavioural expectation’ in which the likelihood of performing a certain behaviour is estimated, is a better predictor of future behaviour than intention because it incorporates factors such as fluctuations in motivation and control limitations. Although the empirical support for this notion is equivocal,¹² research indeed indicates that behavioural intention and behavioural expectation are conceptually distinct and both should be considered as behavioural predictors.^13,14 From the model it is furthermore suggested that behaviour is related to the extent to which young people are ‘open’ to having unprotected sex, or the so-called behavioural willingness. In other words, young females may, in response to environmental cues, show risky behaviour without having intended to do so and they may be willing to take certain risks under such risk-conducive circumstances.

Our study, which is based on a prospective design with a one-year follow-up measurement, incorporates these different, non-exclusive perspectives on reasons that may account for the intention-behaviour gap. It is expected that those who (a) have low intentions to use condoms with future partners, (b) fluctuate in their motivation over time, (c) have low expectations of actually using condoms with future partners, and (d) are willing to have unprotected sex under certain risk-conducive circumstances; are also less likely to be adequately prepared for condom use. Behavioural willingness and behavioural expectation are hypothesized to have a stronger influence on behaviour than (fluctuations in) intention, because they both appear to be more realistic representations of what individuals expect themselves to be doing in the future.

METHODS

Participants and procedure

First-year psychology students at Leiden University were invited to participate in the study via a webpage on ongoing experiments at the Department. A total of 112 students accepted our invitation and received information on the purpose of the study. They were informed that participation in the study was on a voluntary basis, and that their answers would be treated confidentially. Furthermore, they were instructed that they could ask questions before, during or after completion of the questionnaire, and that they were free at any time to withdraw from the study without having to explain their motives. All participants signed the informed consent form and filled out a paper-and-pencil questionnaire on ‘Relationships, sexuality and condom use’, which took them an average of 20 minutes. After completion of the questionnaire, they received course credits.

After 12 months of the first measurement, the respondents were asked via email to participate in the follow-up study. The second questionnaire was shorter than the first one, and could be completed through the internet. Among participants of this follow-up study, seven gift vouchers of €25 were distributed by a lottery. The procedure was in line with the guidelines installed by the Institutional Review Board of the Faculty of Social Sciences at Leiden University.

Eight of the participants could not be located for the second time. Of the remaining 104 respondents at baseline, 70 (62 females and eight males) took part in the follow-up study. Student's t-tests indicated that there were no significant differences between those who completed the second questionnaire and those who did not with regard to age or any of the predictor variables (i.e. attitude towards condom use, subjective norm, perceived control over condom use, perceived control over condom management, intention, behavioural expectation and behavioural willingness). In the following, we will focus on the female respondents only. At baseline, their mean age was 18.86 (standard deviation [SD] = 1.42) ranging from 18 to 24 years. They had had on average three sexual partners in the past (M = 2.96, SD = 2.94; range 1–14), and almost one-third (n = 22) had had experience with casual sex. Five participants reported at baseline to have ever had a STI, and three had been pregnant. Twelve participants had been tested in the year between measurements, of whom none was diagnosed with a STI. Three respondents reported having used the morning-after pill and one other had been pregnant in the year between measurements.

Baseline measures

Attitude towards condom use^15,16 (α = 0.74) was measured by a five-point semantic differential scale reflecting whether respondents thought that using condoms when having sex (in all sections of the questionnaire involving items referring to ‘having sex’, it was indicated that the term was defined as ‘having sexual intercourse’.) with a new partner is ‘useless/useful’, ‘unpleasant/pleasant’, ‘bad/good’, ‘unhealthy/healthy’, ‘uncomfortable/comfortable’ and ‘unsafe/safe’.

Subjective norms^15,16 (α = 0.70) were assessed by two items with a five-point scale, referring to the extent to which respondents believed that (i) peers and (ii) parents thought that they should use condoms when having sex with a new partner.

Perceived control:¹⁷ Ten items, which could be answered on a five-point scale referred to the perceived capability of using condoms. Factor analyses yielded two different factors, which were interpreted as perceived control over condom use (i.e. being able to use condoms under difficult situations, including ‘when being drunk or high’, ‘when being highly aroused’, ‘when it is dark’, ‘when being nervous’, ‘when being very much in love’ and ‘when sex is unexpected’; α = 0.87), and as perceived control over condom management (i.e. being able to use condoms smoothly, without interrupting the process, e.g. ‘without spoiling the atmosphere’, ‘without making a mess of it’, ‘without the other noticing much of it’ and ‘without reducing sexual excitement’; α = 0.71).

Intention^15,16 (α = 0.89) to use condoms with new partners in the future was measured by three items on a five-point scale (‘Do you intend to always use condoms when having sex with a new partner?’, ‘Do you have plans to always use condoms when having sex with a new partner?”, and ‘Is it your aim to always use condoms when having sex with a new partner?’).

Behavioural willingness to have unprotected sex¹⁷ (α = 0.77) was assessed by four items that could be answered on a five-point scale (‘Would you be willing to have sex without condoms: ‘… if your partner indicates that he or she has only very rarely had unprotected sex?”, ‘… if you would use the pill?, ‘… if your partner indicates that he would rather have sex without condoms?’ and ‘… if you and your partner both did not have condoms available?’).

Behavioural expectation¹⁷ (α = 0.95) is a combined measure of two items that could be answered on a five-point scale (‘How likely is it that you will always use a condom when having sex with a new partner?’, and ‘How high are the chances that you will always use a condom when having sex with a new partner?’).

Follow-up measures

Intention and behavioural expectation were assessed again at the one-year follow-up. Temporal instability of intention⁸ was calculated by adding the absolute difference scores between each of the corresponding three pair of items at baseline and at the follow-up measurement. Scores on temporal instability of intention ranged from 0 to 12. A higher score on this variable reflects a less stable intention, i.e. the answers to the various items showed larger differences between the two measurement points in time.

Preparatory behaviours for condom use:¹⁷ At the follow-up measurement, participants were asked whether they had ever bought condoms in the past year (yes/no). They also indicated if they (i) had condoms stored at home, (ii) carried condoms with them in general and (iii) carried condoms with them when they thought that they might have sex. These three items could be answered on a five-point scale ranging from ‘never’ to ‘always.

RESULTS

Preliminary analyses: relationships between predictor variables

Table 1 shows the intercorrelations between the various predictor scales. All scales were relatively independent. There was a significant relationship between intention and temporal instability of intention. Behavioural expectation, however, was more strongly related to temporal instability of intention. Additional crosstab analyses with subgroups based on baseline scores revealed that those who were highly stable over time (i.e. had an instability score of 0), were far more likely to have the highest baseline score (i.e. a score of 5) than all others, which implies that those who are stable in their intention also have the highest onset scores.

Table 1

Scales descriptives and intercorrelations (n = 55–62)

	Mean (SD)	1^†	2	3	4	5	6	7	8	9	10	11
1. T2^‡ Having bought condoms	44% yes
2. T2 Having condoms available at home	3.44 (1.54)	0.57***
3. T2 Carrying condoms in general	1.49 (0.99)	0.41***	0.40***
4. T2 Carrying condoms when intercourse may occur	2.51 (1.49)	0.51***	0.62***	0.61***
5. T1 Attitude	4.13 (0.49)	0.26*	0.15	0.00	0.09
6. T1 Subjective norm	4.43 (0.74)	0.15	0.01	−0.12	−0.05	0.32**
7. T1 Perceived control over condom use	4.01 (0.84)	0.03	0.09	−0.11	0.18	0.27*	0.39***
8. T1 Perceived control over condom management	3.33 (0.81)	0.20	0.12	0.19	0.23*	0.27*	0.29*	0.46***
9. T1 Intention	4.07 (0.99)	0.18	0.12	0.10	0.11	0.52***	0.37***	0.19	0.28*
10. T1 Behavioural expectation	3.72 (1.11)	0.38**	0.32**	0.16	0.31*	0.56***	0.27*	0.27*	0.32**	0.78***
11. T1 Behavioural willingness	2.68 (0.89)	−0.41***	−0.50***	−0.30*	−0.41***	−0.43***	−0.20	0.30*	−0.39***	−0.46***	−0.70***
12. Temporal instability of intention	2.52 (2.37)	−0.18	−0.13	−0.07	−0.17	−0.25*	−0.33**	0.36**	−0.03	−0.31*	−0.41***	0.27*

*P (one-tailed) < 0.05

**P (one-tailed) < 0.01

***P (one-tailed) < 0.001

^†Spearman' ρ correlation coefficients; all others: Pearson correlation coefficients

^‡T1 = baseline measurement; T2 = follow-up measurement

SD = standard deviation

Stepwise regression analysis indicated that attitude (β = 0.52) explained 25% of the variance of intention (F = 21.31; P < 0.001). The other Theory of Planned Behaviour variables did not add significantly to the amount of explained variance of intention. Similarly, attitude (β = 0.56) explained 31% of the variance of behavioural expectation (F = 27.01; P < 0.001). For behavioural willingness to have unprotected sex, attitude (β = −0.36) and perceived control over condom management (β = −0.29) together explained 24% of the variance (F = 10.27; P < 0.001). Finally, perceived control over condom use (β = −0.36) explained 12% of the variance of temporal instability of intention (F = 7.75; P < 0.01).

Predicting preparatory behaviours at follow-up

Almost half of the respondents had bought condoms during the year between measurements. In addition, the group as a whole was most likely to have condoms at home most of the time. The carrying of condoms in general was not a very common behaviour. The carrying of condoms when sex might occur was observed more frequently, but certainly not always.

Logistic regression analysis (forward Wald procedure including all predictor variables) indicated that buying condoms was predicted by behavioural expectation alone (odds ratio [OR] = 2.45; CI 1.28–4.68; Nagelkerke R ² = 22%; χ(1) = 9.19; P < 0.01). In addition, stepwise regression analysis yielded willingness to have unprotected sex as the only predictor to account for a significant unique amount of variance of the availability of condoms at home (β = −0.50, adjusted R ² = 23%, F = 16.43, P < 0.001), of the carrying of condoms in general (β = −0.30, adjusted R ² = 7%, F = 5.09, P < 0.05) and of the carrying of condoms when sex might occur (β = −0.41, adjusted R ² = 15%, F = 10.18, P < 0.01). (Twenty-six participants had had at least one new partner in the year between measurements. Fourteen of them had used a condom during the first sexual intercourse with their most recent partner. Condom use at first intercourse was most strongly related to behavioural willingness to have unprotected sex (Spearman's ρ = −0.39; P < 0.05), followed by behavioural expectation (Spearman's ρ = 0.36; P < 0.05) and by instability of behavioural expectation over time (Spearman' ρ = −0.38; P < 0.05). Logistic regression analysis (forward Wald procedure) yielded behavioural willingness, yet again as single most important predictor of actual condom use during first intercourse (OR = 4.03; CI = 1.17–13.17; Nagelkerke's R ² = 31%; X ² ₍₁₎ = 6.85; P < 0.01). Due to the very limited sample size included in this analysis, however, generalization of these results may not be justifiable.)

DISCUSSION

The present study was based on a prospective design with a long follow-up period. Significant results were found even over such an extended period of time. However, the study involved a relatively small and highly educated sample, and caution is, therefore, warranted when interpreting these results.

The majority of respondents reported to have condoms at home, and thus seemed to show readiness to use condoms, but they did not extend their efforts to taking them along when going outdoors. Preparations for condom use were mainly and consistently dependent on this willingness to have unprotected sex. Clearly, some young women find the risks they are taking acceptable within certain situations, such as when they are highly sexually aroused or the partner is considered to be ‘safe’. As a result, they do not adequately prepare for condom use, which, in turn, increases their chance of encountering exactly such situations in which they are likely to take risks. In such instances it is possible that, for example, the desire to have sex,¹⁸ and its underlying motives^15,19 may over-ride the possible concerns that the person has about the risk accompanying unprotected sex.

Intention, on the other hand, was not related to preparatory behaviours for condom use. It should be noted that the majority of the sample had quite strong intentions, but at the same time they were less likely to expect that they would actually do so. Although most past research has focused on intention as the most proximal determinant of condom use, our findings suggest that preparations for condom use are more dependent on the level of willingness to have unprotected sex. It was also found that for most of the participants motivation indeed varied over time. Only 20% gave similar ratings on the three items for intention. Temporal instability of intention was strongly related to baseline measures of behavioural expectation. This indicates that, as theorized, people take into account motivational fluctuations over time when estimating the likelihood of future behaviour. Moreover, stability of scores occurred more frequently among those with more positive attitudes, a higher subjective norm, higher feelings of perceived control over condom use and stronger intentions. Thus, apparently a subgroup of people can be discerned, who are convinced that they should and will use condoms with new partners, and feel able to do so, even under difficult circumstances, and keep this conviction over time. This may suggest that they have internalized the goal of protected sex and that for them the use of condoms is a function of intrinsic values rather than that of environmental cues.²⁰

Finally, behavioural willingness was related to attitude towards condom use and to feeling confident that one can effectively and smoothly use condoms. Practicing condom use and learning from mastery experience is, therefore, likely to be a means through which this process of preparing could be enhanced. It is plausible that when young females feel confident that they are capable of using condoms even in the face of obstacles and without ‘spoiling’ the sexual experience, they will regard condom use more favourably. They will then probably be less inclined to accept risks by having the partner to decide whether a condom will be used and take an active role in condom use preparation.

References

Centre for Disease Control and Prevention. Sexually Transmitted Disease Surveillance, 2005. Atlanta, GA: US Department of Health and Human Services, 2006. See www.cdc.gov/std/stats/toc2005.htm (accessed 6 March 2006)

De Boer

, Op de Coul

ELM

, Koedijk

FDH

, van Veen

, van Sighem

, van de Laar

MJW

. HIV and Sexually Transmitted Infections in the Netherlands in 2005. Bilthoven, The Netherlands: RIVM. (RIVM report 441100024/2006). See www.rivm.nl/bibliotheek/rapporten/441100024.pdf (accessed March 23, 2007)

Bakker

. Neemt het aantal mensen met onveilig seksueel gedrag toe of af? [Is The Number of People Who Show Unsafe Sexual Behavior Increasing or Decreasing?]. De Bilt, The Netherlands: RIVM, 2004

Bryan

, Fisher

. Tests of the mediational role of preparatory safer sexual behavior in the context of the Theory of Planned Behavior. Health Psychol 2002;21:71–80

Ajzen

. From intentions to actions: a theory of planned behavior. In: Kuhl

, Beckman

eds. Action-Control: From Cognition to Behavior. Heidelberg, Berlin: Springer, 1985:pp 11–39

Albarracin

, Johnson

, Fishbein

, Muellerleile

. Theories of reasoned action and planned behavior as models of condom use: a meta-analysis. Psychol Bull 2001;127:142–61

Sheeran

, Abraham

, Orbell

. Psychosocial correlates of heterosexual condom use: a meta-analysis. Psychol Bull 1999;125:90–132

Conner

, Sheeran

, Norman

, Armitage

. Temporal stability as a moderator of relationships in the Theory of Planned Behaviour. Br J Soc Psychol 2000;39:469–93

Sheeran

, Orbell

, Trafimow

. Does the temporal stability of behavioral intentions moderate intention-behavior and past behavior-future behavior relations? Pers Soc Psychol Bull 1999;25:724–34

10.

Gibbons

, Gerrard

, Blanton

, Russell

. Reasoned action and social reaction: willingness and intention as independent predictors of health risk. J Pers Soc Psychol 1998;74:1164–81

11.

Gibbons

, Gerrard

, Lane

. A social-reaction model of adolescent health risk. In: Suls

, Wallston

eds., Social Psychological Foundations of Health and Illness. Oxford, UK: Blackwell, 2003:pp 107–36

12.

Sheeran

. Intention-behavior relations: a conceptual and empirical review. Eur Rev Soc Psychol 2002;12:1–36

13.

Rhodes

, Hunt Matheson

. Discrepancies in exercise intention and expectation: theoretical and applied issues. Psychol Health 2005;20:63–78

14.

Warshaw

, Davis

. Disentangling behavioral intention and behavioral expectation. J Exp Soc Psychol 1985;21:213–28

15.

Gebhardt

, Kuyper

, Greunsven

. Need for intimacy in relationships and motives for sex as determinants of adolescent condom use. J Adolesc Health 2003;33:154–64

16.

Morrison

, Rogers Gillmore

, Baker

. Determinants of condom use among high-risk heterosexual adults: a test of the Theory of Reasoned Action. J Appl Soc Psychol 1995;25:651–76

17.

Van Empelen

, Kok

. Condom use in steady and casual sexual relationships: planning, preparation and willingness to take risks among adolescents. Psychol Health 2006;21:165–81

18.

Abraham

, Sheeran

, Norman

, Conner

, de Vries

, Otten

. When good intentions are not enough: modeling postdecisional cognitive correlates of condom use. J Appl Soc Psychol 1999;29:2591–612

19.

Gebhardt

, Kuyper

, Dusseldorp

. Condom use at first intercourse with a new partner in female adolescents and young adults: the role of cognitive planning and motives for having sex. Arch Sex Behav 2006;35:217–23

20.

Deci

, Ryan

. Intrinsic Motivation and Self-Determination in Human Behavior. New York, NY: Plenum Publishing Co, 1985