Effectiveness of a condom use educational program developed on the basis of the Information–Motivation–Behavioral Skills model

Authors


Yoshiko Kudo, College of Nursing Art & Science, University of Hyogo, 13-71 Kitaoji-cho, Akashi, Hyogo 673-8588, Japan. Email: yoshiko_kudou@cnas.u-hyogo.ac.jp

Abstract

Aim:  To assess a condom use educational program developed using the Information–Motivation–Behavioral Skills (IMB) model to enable Japanese young people to use condoms.

Methods:  The program was implemented in two sessions of approximately 90 min each over a 2 day period. The participants underwent the program in small same-sex groups. The study participants were divided into the intervention group (n = 94) and the control group (n = 186). Knowledge relating to HIV/sexually-transmitted diseases (STDs), attitudes towards contraception and sex, the risk perceptions of pregnancy and HIV/STD infection, anticipated reaction to proposed condom use, self-efficacy of condom use, and sexual behaviors were assessed. Data were obtained at baseline, immediately after the program, and 1 and 3 months after the program or baseline survey.

Results:  Intervention group participants with experience of sexual intercourse (n = 14) had significantly higher scores for knowledge relating to HIV/STDs, self-efficacy of condom use, and relief/confusion associated with condom use than the control group (n = 13) at the 1 and 3 month surveys. Intervention group participants with no experience of sexual intercourse (n = 14) had significantly higher scores for knowledge relating to HIV/STDs and self-efficacy of condom use, and significantly lower scores for condom avoidance, than the control group (n = 116). Condom use behavior was not changed in either group.

Conclusion:  The program used in this study did not promote condom use, but it increased knowledge relating to HIV/STDs and self-efficacy of condom use and reduced resistance to condom use.

INTRODUCTION

Young Japanese people are engaging in sexual activity at increasingly younger ages, with the result that infection with HIV or other sexually transmitted diseases (STDs) and abortions are becoming a problem (Healthy Parents and Children 21 Study and Evaluation Committee, 2010). In 2010, there were 1075 new HIV cases in Japan (AIDS Trend Committee of the Japan Ministry of Health and Welfare, 2011), with cases between the ages of 15 and 24 years accounting for approximately 14% of all cases. In 2009, the rates of induced abortion in the Japanese female population aged 15–19 years and 20–24 years were 7.1 and 15.1, respectively, per 1000 (Mothers' & Children's Health & Welfare Association, 2011). Approximately 90% of Japanese youth engaging in sexual activity use a condom for contraception (Kishida & Kitamura, 2005). However, the percentage of students who reported having had sexual intercourse without using a condom in the previous 6 months was 74.3% for female university students and 70.4% for male university students in Japan (Imai et al., 2004). It has also been shown that the rate of condom use is lower with non-regular than regular partners, and it is lower in people who have had a greater number of sexual partners over the past year (Behavioral Science Research Group of the HIV Epidemiological Survey Team of the Ministry of Health and Welfare, 2001b; Kihara & Kihara, 2002). Because young Japanese people have not consistently used condoms, they have a high risk of STDs and unplanned pregnancies. In Japan, various sexual education programs have been carried out with the aim of preventing STDs and unplanned pregnancies among young people, but their effectiveness has not been verified.

Sexually transmitted diseases and unplanned pregnancies can be prevented by consistent condom use during sexual intercourse. Interventions in the USA designed to prevent STDs and unplanned pregnancies emphasize change of sexual risk-taking behaviors, either by delaying or decreasing sexual activity or by increasing condom use. Effective programs were based on theoretical approaches that have been demonstrated to be effective in influencing other health-related risky behaviors (Kirby, 2000; Lyles et al., 2007; Pedlow & Carey, 2004). One theory and model that conceptualizes the principal factors in preventive behavior is the Information–Motivation–Behavioral Skills (IMB) model. The model asserts that information, motivation, and behavioral skills are the fundamental determinants of preventive behavior (Fisher, J.D. & Fisher, W.A, 1992, 2000; Fisher, W.A. & Fisher, J.D., 1993). Information that is relevant to HIV/STD transmission and prevention is an obvious prerequisite for preventive behavior (e.g. condom use). Motivation to engage in acts to prevent HIV/STDs is an additional determinant of preventive behavior. The motivation to practice a preventive act is assumed to be a function of one's attitudes toward the preventive act (personal motivation), relevant subjective norms regarding the preventive act (social motivation), and perception of personal vulnerability to HIV/STD infection and becoming pregnant. Behavioral skills for performing specific preventive acts are an additional prerequisite for preventive behavior. The behavioral skills component of the IMB model is composed of an individual's objective ability and his or her perceived self-efficacy concerning performance of the sequence of preventive behaviors involved in the preventive practice. The IMB model, the model that most comprehensively integrates these various processes, allows implementation of appropriate behavior by: (i) providing young people with the necessary information; (ii) motivating them through their attitudes, norms, and perceived vulnerabilities; and (iii) developing their behavioral skills (Marsh, Johnson, & Carey, 2003). The IMB model has been used to test prediction of HIV preventive behavior and to develop and evaluate a HIV prevention program, and recently to provide guidance for the design, implementation, and evaluation of diabetes self-care programs and interventions to promote adherence among patients undergoing coronary artery bypass grafting surgery and among HIV-positive patients on antiretroviral therapy (Osborn et al., 2010; Osborn & Egede, 2010; Starace, Massa, Amico, & Fisher, 2006; Zarani, Besharat, Sadeghian, & Sarami, 2010).

Intervention programs that seek to change behavior by providing information alone without arousing additional motivation seem to be especially ineffective in populations that have already received sufficient information, and programs that only provide information about condoms with no skills training have been shown to be ineffective (Marsh et al., 2003). Conversely, sexual risk behavior has been clearly shown to decrease when sexual communication skills, assertiveness skills, and negotiation skills training are introduced into programs (Kim, Stanton, Li, Dickersin, & Galbraith, 1997; Marsh et al., 2003; Pedlow & Carey, 2004). These skills require the use of formal operational thinking, which adolescents are just beginning to develop (Huszti, Hoff, & Johnson, 2003). Sexual education programs for youth also need to include accurate information and skill training in Japan.

The purpose of this study was to assess the effectiveness of the condom use educational program developed using the IMB model (Fisher & Fisher, 1992) for young Japanese people in their late teens and early 20s.

METHODS

Study design

The present study was a quasi-experiment using a convenience sample. Changes in sexual behavior and psychosocial mediators of condom use behavior in an intervention group that attended a condom use educational program and a control group that did not attend the program were measured through a baseline survey and follow-up surveys 1 and 3 months later. The psychosocial mediators of condom use behavior in the intervention group were measured through a post-test survey immediately after the intervention. This study was approved by the Research Ethics Committee of the College of Nursing Arts and Sciences, University of Hyogo.

Condom use education program

The condom use education program was developed from information obtained by focus group interviews and a published work review (Kudo, 2009). The focus group interview members were similar to young people targeted by the intervention. The education program consisted of two approximately 90 min sessions. The program provided one session each day and was administered to same-sex groups of 2–9 participants. In this study, the second session was conducted 1.7 ± 1.8 days (range, 1–8 days) after the first session. The program was provided to 7 groups of males and 12 groups of females, with 5.0 ± 2.4 participants per group. A total of 38 sessions was performed by the author alone.

The first session gave information about why condom use is effective in preventing HIV/STDs and unplanned pregnancy and considered the individual's responsibility for his or her own behavior with regard to sexual intercourse. In this session, participants were equipped with knowledge relating to HIV infection and how it disrupts immunity, routes of HIV infection, the current status of HIV/AIDS, the effects of HIV infection on the body and everyday life, how contraception works, the period for abortion, and the effects of abortion. This is essential information for implementing condom use behavior and for avoiding unprotected sexual behavior. Having this information leads to affirmative attitudes and motivation toward carrying out the behavior. Furthermore, presenting messages from HIV-positive people and showing participants that HIV is something that can occur within one's immediate environment increases their perception of the risk. Finally, affirmative attitudes toward condom use and avoiding unprotected sexual behavior were formed through a discussion of the possible outcomes of sexual intercourse and the meaning of taking responsibility for one's own behavior.

In the second session, participants were given knowledge and skills to correctly use a condom. Participants were shown how to use both male and female condoms via computer graphic images, helping them not only to acquire correct knowledge and skills relating to condom use, but also to lose their resistance to condoms and be motivated to use them. Brainstorming and role-plays were used for participants to acquire sexual communication and negotiation skills relating to condom use, which helped them to gain the strategies to ensure condom use and avoid unprotected sex. Participants were given messages from HIV-positive people at the end of this session as well, increasing their perception of the risk of HIV infection and motivating them to use condoms.

Subjects

The subjects were Japanese males and females in their late teens and early 20s who consented to participate in the study. The need for parental consent was waived by the Research Ethics Committee. The intervention group that attended the condom use educational program was recruited in one university and three high schools that consented to provide their students with the condom use education program through the display and distribution of notices asking for participation in the study. The educational program was provided to 94 people who desired to attend the program. The control group (n = 186) who did not attend the program was recruited from among groups that allowed contact through presentation of written and verbal explanations of the study in two universities and two high schools that consented to cooperate with the study as controls.

Based on an effect size on condom use of 0.43 (Johnson, Carey, Marsh, Levin, & Scott-Sheldon, 2003), the sample size with a 5% significance level on either side and 80% statistical power was calculated to be a sample of 86 young people in each group according to G-power. The dropout rate after 3 months from interventions relating to condom use has been reported to be 3–16% (Jemmott, Jemmott, & Fong, 1992, 1998; Jemmott, Jemmott, Fong, & MaCaffree, 1999; Rotheram-Borus, Murphy, Fernandez, & Srinivasan, 1998), so that the number of subjects required for the study was 178–205.

Measures

The survey was a self-administered questionnaire consisting of demographics and variables relating to the IMB construct. All questionnaires were completed anonymously.

Demographics

The participants reported their age, sex, school grades, and family composition. They were also asked if they currently had a boyfriend/girlfriend, and if they had ever had sexual intercourse.

Information

The information was knowledge relating to HIV/STDs using 20 items of HIV/STD-related knowledge developed by the Behavioral Science Research Group of the HIV Epidemiological Survey Team of the Ministry of Health and Welfare (2001a). This measurement tool consists of true/false questions relating to HIV infection routes and tests, types of STDs and their infection routes, and interactions between STDs and HIV infection. Their internal validity has been confirmed. Primary to tertiary pilot studies (Behavioral Science Research Group of the HIV Epidemiological Survey Team of the Ministry of Health and Welfare, 2001a) have been carried out on males and females aged 18–59 years, with the answers found to be consistent.

Motivation

Motivation was evaluated using tools to measure attitudes towards contraception and sex, and the risk perception of pregnancy or HIV/STD infection. Attitudes towards contraception included 22 items (Fukumoto & Morinaga, 2005). These items were avoidance of condom use (7 items), proactive attitude toward condom use and trust of people who use condoms (7 items), trust in extravaginal ejaculation and people to practice it (5 items), and feeling of rejection of extravaginal ejaculation (3 items), which were all measured on 5 point scales. The coefficient-α for each subscale was 0.69–0.71.

Attitudes towards sex (Fukumoto & Morinaga, 2005) measured norms toward sex and were made up of double standards regarding sexual relations (5 items) and openness to sexual relations (5 items), which were measured on 5 point scales. The coefficient-α was 0.71 and 0.69, respectively.

Vulnerability to risk was measured by risk perception of pregnancy and risk perception of HIV/STDs (Fukumoto & Morinaga, 2005). These were measured by reporting a probability from 0–100% of becoming pregnant or being infected by HIV/STDs when using either a condom or extravaginal ejaculation alone. Also, a probability from 0–100% was reported for becoming pregnant or being infected by HIV/STDs when having sexual intercourse without protection.

Behavioral skills

Behavioral skills were evaluated using two measurement tools: anticipated reaction when condom use is proposed and a self-efficacy scale. The anticipated reaction when condom use is proposed (Fukumoto & Morinaga, 2005) requires respondents with previous sexual experience to envisage the next sexual encounter with their most recent sexual partner and measures the response to proposed or actual condom use with 9 items on 5 point scales. This measurement tool is made up of one's own repugnance to or refusal of condom use (3 items), the partner's repugnance to or refusal of condom use (3 items), and relief/confusion associated with condom use (3 items). The coefficient-α was 0.79, 0.82, and 0.68, respectively. The self-efficacy scale (Fukumoto & Morinaga, 2005) was a scale of self-efficacy of condom use (6 items) on 5 point scales, with a coefficient-α of 0.81.

Sexual behavior

Sexual behavior measured lifetime sexual intercourse, age at first intercourse, use of condom at first intercourse, lifetime sexual partners, and frequency of condom use, sexual intercourse over the previous month, frequency of condom use over the previous month, frequency of insisting that the partner use a condom over the previous month, purchase or carrying of condoms over the previous month, and use of a condom at the most recent sexual intercourse.

Procedures

Study participants in both the intervention group and the control group received verbal and written explanations of the details of the survey. The intervention group completed a baseline survey, a post-test survey immediately after the program, and follow-up surveys 1 and 3 months after the program. The control group completed a baseline survey and follow up surveys 1 and 3 months after baseline. Each group participant was given three or four different envelopes containing the relevant questionnaire forms, each form marked with the same ID. The baseline survey included questions relating to demographics, information, motivation, behavioral skills, and sexual behavior. The post-test survey (intervention group only) included questions about whether or not the participant attended the sessions, information, motivation, and behavioral skills. The follow-up surveys included questions about whether or not the participant attended the sessions (intervention group only), information, motivation, behavioral skills, and sexual behavior. The participants wrote their names and addresses on the envelopes for the post-test (intervention group only) and follow-up surveys. The envelopes were then collected by a researcher. The participants then each filled out the baseline survey anonymously, sealed it in the envelope, and gave it to a researcher. The intervention group attended the first session of the condom use educational program immediately after completing the baseline survey. After the end of the second session, the envelopes containing the questionnaire forms for the post-test survey were either mailed or handed over to the intervention group participants, who returned their responses using an envelope addressed to the researchers. The questionnaires for the follow-up surveys were mailed to the intervention group participants 1 and 3 months after the program, and were mailed or handed over to the control group participants 1 and 3 months after the baseline. The post-test and follow-up surveys were mailed or handed over by part-time research assistants, all of whom signed a confidentiality agreement. The participants returned their responses using envelopes addressed to the researchers. Participant flow through this study is shown in Figure 1.

Figure 1.

Diagram of participant flow through the study by group.

Analysis

The characteristics of the intervention (n = 94) and control (n = 186) groups were analyzed based on their responses to the baseline survey. The effectiveness of the program was analyzed using the responses of 157 participants (intervention group, 28; control group, 129) who returned all survey questionnaires (baseline, 1 month, and 3 months); three people gave contradictory responses regarding sexual intercourse and were excluded (Fig. 2).

Figure 2.

Sexual experience of subjects who returned all survey questionnaires.

The 20 items of HIV/STD-related knowledge were analyzed using the number of correct responses. The risk perception of pregnancy or HIV/STD infection was analyzed using the numerical values given as a probability from 0–100%. The other motivation and behavioral skills measurement tools were analyzed using the total scores of the scales.

The characteristics of the intervention group and the control group were compared using the χ2-test and Student's t-test. The effectiveness of the program was determined using the χ2-test, a paired Student's t-test, and repeated measures anova. The level of significance was P < 0.05. All analyses were performed using SPSS Statistics 17.0 (SPSS, Chicago, IL, USA).

RESULTS

Participants' characteristics

There were no significant differences in sex ratio, average age, or proportion of university and high school students between the groups (Table 1). However, there was a significant difference between the groups in lifetime sexual intercourse (χ2[1] = 19.04, P < 0.001).

Table 1. Participants' baseline characteristics
CharacteristicsInterventionControl
SexMale37 (39.4%)59 (31.7%)
Female57 (60.6%)127 (69.3%)
Age (mean) (years) 16.8 (SD, 1.6)16.9 (SD, 2.0)
SchoolUniversity14 (14.9%)46 (27.4%)
High school80 (85.1%)140 (75.3%)
Sexual experience***Yes44 (48.4%)40 (21.5%)
No50 (51.6%)146 (78.5%)
Age at first intercourse (mean) (years)16.0 (SD, 1.7)16.0 (SD, 1.9)
Lifetime sexual partners (mean)2.1 (SD, 1.9)2.8 (SD, 2.5)
Condom use behaviorN%N%
  • ***

    P < 0.001. SD, standard deviation.

Condom use at first intercourseUsed3886.43075.0
Did not use511.41025.0
Do not know12.300.0
Frequency of condom useEvery time2147.71538.5
Most of the time1534.11333.3
Some of the time36.8615.4
Rarely36.837.7
Never24.525.1
Having sexual intercourse over the previous monthNo1738.61640.0
Yes2761.42460.0
Frequency of condom use over the previous monthEvery time1763.01354.2
Most of the time414.828.3
Some of the time27.428.3
Rarely311.1312.5
Never13.7416.7
Purchase of condoms over the previous monthYes518.5313.0
No2281.52087.0
Carrying of condoms over the previous monthYes1140.7729.2
No1659.31770.8
Condom use at the most recent sexual intercourseUsed3170.52870.0
Did not use1125.01230.0
Do not know24.500.0

No significant difference was found between the intervention and control groups in average age at first intercourse and lifetime sexual partners (Table 1). Furthermore, no significant difference was found between the intervention and control groups in condom use behavior among people with sexual intercourse experience (Table 1). Average scores and standard deviations of the measurement tools for information, motivation, and behavioral skills in the intervention and control groups are shown in Table 2 and Table 3. Among people with sexual intercourse experience, a significant difference was found between the groups in knowledge relating to HIV/STDs (information) and one's own repugnance to or refusal of condom use (predicted reaction toward proposed condom use), with both scoring lower in the intervention group (Table 2). No significant difference between the intervention and control groups was found for information, motivation, and behavioral skills among people without sexual intercourse experience (Table 3).

Table 2. Information, motivation, and behavioral skills of people with sexual experience in the intervention group (n = 44) and the control group (n = 40)
VariableScaleInterventionControl
MSDMSD
  1. *P < 0.05. Responses only from people with sexual experience. M, mean; SD, standard deviation; STDs, sexually transmitted diseases.

InformationKnowledge relating to HIV/STDs*10.74.212.94.2
Motivation
Attitudes towards contraceptionAvoidance of condom use15.65.816.17.2
Proactive attitude toward condom use and trust toward people27.15.727.05.3
Trust in extravaginal ejaculation and people to practice it11.84.212.13.6
Feeling of rejection of extravaginal ejaculation6.62.76.92.6
Attitudes towards sexDouble standards regarding sexual relations14.13.712.73.6
Openness toward sexual relations15.33.114.73.1
Risk perception of pregnancyBecoming pregnant when using a condom19.632.325.332.8
Becoming pregnant when using extravaginal ejaculation34.923.745.726.8
Becoming pregnant when having sex without protection68.731.274.023.4
Risk perception of HIVInfection when using a condom11.822.616.126.0
Infection when using extravaginal ejaculation48.931.252.131.9
Infection when having sex without protection74.030.482.123.2
Risk perception of STDsInfection when using a condom11.719.715.121.9
Infection when using extravaginal ejaculation49.929.952.132.1
Infection when having sex without protection70.829.775.024.8
Behavioral skills
Anticipated reaction when condom use is proposedOne's own repugnance to or refusal of condom use*3.81.54.72.4
Partner's repugnance to or refusal of condom use5.02.35.62.4
Relief/confusion associated with condom use11.82.711.23.3
Self-efficacySelf-efficacy of condom use19.45.120.84.6
Table 3. Information, motivation, and behavioral skills of young people with no sexual experience in the intervention group (n = 50) and the control group (n = 146)
VariablesScalesInterventionControl
MSDMSD
  1. M, mean; SD, standard deviation; STDs, sexually transmitted diseases.

InformationKnowledge relating to HIV/STDs9.94.210.24.9
Motivation
Attitudes towards contraceptionAvoidance of condom use16.16.415.45.8
Proactive attitude toward condom use and trust toward people26.05.224.55.5
Trust in extravaginal ejaculation and people to practice it13.83.513.13.3
Feeling of rejection of extravaginal ejaculation7.92.27.72.1
Attitudes towards sexDouble standards regarding sexual relations14.23.713.63.6
Openness toward sexual relations14.23.113.43.1
Risk perception of pregnancyBecoming pregnant when using a condom16.626.323.926.8
Becoming pregnant when using extravaginal ejaculation41.829.941.827.9
Becoming pregnant when having sex without protection74.527.773.725.6
Risk perception of HIVInfection when using a condom13.019.718.824.1
Infection when using extravaginal ejaculation45.028.546.729.9
Infection when having sex without protection67.533.269.129.3
Risk perception of STDsInfection when using a condom16.123.619.723.7
Infection when using extravaginal ejaculation45.731.545.429.1
Infection when having sex without protection67.332.269.127.7
Behavioral skills
Self-efficacySelf-efficacy of condom use18.65.117.64.6

Characteristics of subjects who returned all survey questionnaires

The subjects were 157 people from whom responses were received for the baseline and 1 and 3 month follow-up surveys. In the comparison of the attributes of these persons between groups, there were 11 males (39.3%) and 17 females (60.7%) in the intervention group, and 50 males (38.8%) and 79 females (61.2%) in the control group, with no significant difference. The average age was 17.9 ± 2.1 years in the intervention group and 16.2 ± 1.4 years in the control group; the intervention group was significantly older (P = 0.001). Comparison by place of study showed 10 university students (35.7%) and 18 high school students (64.3%) in the intervention group, and 10 university students (7.8%) and 119 high school students (92.2%) in the control group, with a significantly higher percentage of university students in the intervention group than in the control group.

Effectiveness of the condom use education program

Behavior relating to condom use at the baseline, 1 month, and 3 month surveys was compared between the intervention and control groups (Table 4). At each survey, there were significantly more people with sexual intercourse experience in the intervention group. The number of people engaging in first sexual intercourse at the 3 month survey was one in the intervention group and five in the control group (Fig. 2). There were no differences between the intervention and control groups in sexual behavior: condom use in lifetime sexual intercourse, condom use over the previous month, condom use or extravaginal ejaculation in the most recent sexual intercourse, purchase or carrying of a condom over the previous month, and frequency of insisting that the partner use a condom.

Table 4. Sexual behavior in the intervention and control groups
Behavior Baseline1 month3 months
InterventionControlInterventionControlInterventionControl
N%N%N%N%N%N%
Lifetime sexual intercourseNo1450.011689.91346.411488.41346.411186.0
Yes1450.01310.11553.61511.61553.61814.0
Frequency of condom useEvery time642.9650.0533.3853.3426.7950.0
Most of the time642.9433.3640.0533.3746.7633.3
Some of the time00.0216.7426.716.7320.015.6
Rarely214.3014.300.016.717.100.0
Never00.000.000.000.006.7211.1
Having sexual intercourse over the previous monthNo535.7753.8533.31066.7320.0844.4
Yes964.3646.21066.7533.31280.01055.6
Frequency of condom use over the previous monthEvery time666.7350.0440.0480.0541.7660.0
Most of the time222.2116.7220.000.0216.700.0
Some of the time111.1116.7330.0120.0216.7440.0
Rarely00.0116.7110.000.0216.700.0
Never00.000.000.000.018.300.0
Frequency of insisting partner use a condom over the previous monthEvery time444.4350.0444.4360.0650.0333.3
Most of the time111.100.0444.400.0216.700.0
Some of the time222.2116.700.0120.018.3111.1
Rarely111.100.0111.1120.000.0222.2
Never111.1233.300.000.0325.0333.3
Condom use at the most recent sexual intercourseUsed1178.61292.31178.61392.91173.31688.9
Did not use321.417.7321.417.1320.015.6
Do not know00.000.000.000.016.715.6
Purchase of condoms over the previous monthYes111.1116.7222.2125.0541.7110.0
No888.9583.3777.8375.0758.3990.0
Carrying of condoms over the previous monthYes444.4350.0444.4125.0650.0222.2
No555.6350.0555.6375.0650.0777.8

Because there was a significant difference between the intervention and control groups in the proportion of subjects who had sexual intercourse experience, information, motivation, and behavioral skills at the baseline, 1 month, and 3 month surveys were compared between the intervention and control groups among people with sexual experience and among people without sexual experience. Table 5 shows the means and standard deviations of the scores for the scales measuring information, motivation, and behavioral skills at each survey (baseline, 1 month, 3 months) for people with sexual experience at the baseline survey (intervention group, 14; control group, 13). A repeated measures two-way anova showed interactions between responses for knowledge relating to HIV/STDs (information) (P < 0.01), possibility of infection when using a condom (the risk perception of HIV infection; P < 0.05), relief/confusion associated with condom use (behavioral skills, anticipated reaction to proposed condom use; P < 0.05), and self-efficacy of condom use (P < 0.001). In the repeated measures analysis of each group, scores for knowledge relating to HIV/STDs in the intervention group were significantly higher at the 1 month follow-up survey than at the baseline survey. In the intervention group, scores for relief/confusion associated with condom use were significantly higher at the 3 month follow-up survey than at the baseline survey, and scores for self-efficacy of condom use were higher at the 1 month and 3 month follow-up surveys than at the baseline survey. Furthermore, while no interaction was found, scores for avoidance of condom use were significantly lower at the 1 month and 3 month follow-up surveys than at the baseline survey in both the intervention and control groups.

Table 5. Information, motivation, and behavioral skills of people with sexual experience (intervention group, n = 14; control group, n = 13)
ValuablesTimeInterventionControlRepeated measures two-way anova
TimeGroupTime × group
NMSDNMSDF P F P F P
  1. *Repeated measures anova for each group, Bonferroni method for multiple comparisons: difference from baseline survey tested, P < 0.05. Responses only from people with sexual experience. M, mean; NS, not significant; SD, standard deviation; STDs, sexually transmitted diseases.

InformationKnowledge relating to HIV/STDsBaseline 11.25.4 12.94.41.58NS0.38NS7.33<0.01
1 month1314.9*4.81211.65.4      
3 months 13.56.6 11.35.8      
Motivation
Attitudes towards contraceptionAvoidance of condom useBaseline 14.96.5 15.35.916.41<0.0010.10NS0.12NS
1 month1311.8*4.51212.3*5.6      
3 months 11.3*4.6 12.3*5.3      
Proactive attitude toward condom use and trust toward peopleBaseline 29.44.6 27.44.80.98NS4.75<0.051.50NS
1 month1229.35.11225.93.8      
3 months 29.65.4 24.26.3      
Trust in extravaginal ejaculation and people to practice itBaseline 11.43.9 12.84.70.57NS0.05NS0.91NS
1 month1411.44.31011.43.3      
3 months 11.93.9 11.65.5      
Feeling of rejection in relation to extravaginal ejaculationBaseline 7.52.6 6.42.31.15NS0.58NS0.42NS
1 month137.71.7137.42.5      
3 months 7.73.0 7.22.7      
Attitudes towards sexDouble standards regarding sexual relationsBaseline 13.43.6 13.13.90.39NS0.11NS0.21NS
1 month1413.53.81112.64.2      
3 months 13.64.3 13.44.3      
Openness toward sexual relationsBaseline 15.62.5 14.03.30.03NS1.15NS0.50NS
1 month1415.32.51214.43.3      
3 months 15.63.4 14.34.2      
Risk perception of pregnancyBecoming pregnant when using a condomBaseline 11.226.7 18.526.90.84NS1.44NS0.24NS
1 month1311.227.01235.337.7      
3 months 18.034.5 21.228.5      
Becoming pregnant when using extravaginal ejaculationBaseline 28.219.7 36.429.20.30NS0.31NS0.18NS
1 month1331.321.61142.529.0      
3 months 33.227.4 37.220.5      
Becoming pregnant when having sex without protectionBaseline 56.630.0 73.622.01.48NS1.11NS1.13NS
1 month1355.230.71158.933.1      
3 months 54.835.2 69.931.5      
Risk perception of HIVInfection when using a condomBaseline 3.57.4 11.010.71.19NS10.03<0.013.67<0.05
1 month130.61.41222.625.5      
3 months 4.28.3 10.613.9      
Infection when using extravaginal ejaculationBaseline 49.033.4 42.335.22.03NS0.03NS0.40NS
1 month1352.137.71157.329.6      
3 months 42.837.8 38.032.8      
Infection when having sex without protectionBaseline 66.435.5 87.911.83.83<0.051.15NS1.39NS
1 month1363.929.61274.228.0      
3 months 59.234.9 58.737.8      
Risk perception of STDsInfection when using a condomBaseline 4.27.5 12.312.20.21NS5.01<0.052.31NS
1 month132.25.51217.320.1      
3 months 7.314.1 11.811.7      
Infection when using extravaginal ejaculationBaseline 50.434.7 48.335.61.85NS0.00NS0.05NS
1 month1352.737.41254.430.7      
3 months 41.634.7 41.832.2      
Infection when having sex without protectionBaseline 68.932.9 76.723.13.42<0.050.14NS0.21NS
1 month1365.426.31268.726.0      
3 months 56.930.8 56.840.0      
Behavioral skills
Anticipated reaction when condom use is proposedOne's own repugnance to or refusal of condom useBaseline 4.22.1 5.02.42.12NS0.84NS0.21NS
1 month143.81.9134.62.3      
3 months 3.91.8 4.42.5      
Partner's repugnance to or refusal of condom useBaseline 5.12.1 5.62.22.88NS0.00NS1.52NS
1 month145.02.3114.62.4      
3 months 4.82.0 4.62.7      
Relief/confusion associated with condom useBaseline 11.62.8 11.52.70.48NS3.02NS5.70<0.05
1 month1412.43.01310.53.7      
3 months 12.9*2.5 9.24.2      
Self-efficacySelf-efficacy of condom useBaseline 18.84.4 22.13.310.90<0.0010.03NS10.38<0.001
1 month1422.1*3.61221.23.0      
3 months 24.0*4.3 22.34.4      

Table 6 shows the means and standard deviations of the scores for the scales measuring information, motivation, and behavioral skills at each survey (baseline, 1 month, 3 months) for people without sexual experience at the baseline survey (intervention group, 14; control group, 116). Repeated measures two-way anova showed interactions among responses for knowledge relating to HIV/STDs (information; P < 0.001), avoidance of condom use (motivation, attitudes towards contraception; P < 0.05), and trust in extravaginal ejaculation and people to practice it (motivation, attitudes towards contraception; P < 0.05). In the repeated measures analysis of each group, scores for knowledge relating to HIV/STDs in the intervention group were significantly higher at the 1 month and 3 month follow-up surveys than at the baseline survey. Scores for avoidance of condom use were significantly lower in the intervention group at the 1 month and 3 month follow-up surveys than at the baseline survey. No significant difference between groups was found for trust in extravaginal ejaculation and people to practice it in the repeated measures analysis.

Table 6. Information, motivation, and behavioral skills of people without sexual experience (intervention group, n = 14; control group, n = 113)
ValuablesTimeInterventionControlRepeated measures two-way anova
TimeGroupTime × group
NMSDNMSDF P F P F P
  • *

    Repeated measures ANOVA for each group, Bonferroni method for multiple comparisons: difference from baseline survey tested, P < 0.05. M, mean; NS, not significant; SD, standard deviation; STDs, sexually transmitted diseases.

InformationKnowledge relating to HIV/STDsBaseline 12.53.6 9.84.97.51= 0.00119.99<0.00116.16<0.001
1 month1217.1*2.7949.15.6      
3 months 17.3*2.2 8.85.5      
Motivation
Attitudes towards contraceptionAvoidance of condom useBaseline 15.85.1 16.05.81.99NS4.41<0.053.87<0.05
1 month1412.65.111116.05.4      
3 months 12.45.0 17.06.0      
Proactive attitude toward condom use and trust toward peopleBaseline 26.76.5 23.75.40.19NS13.53<0.0011.02NS
1 month1427.45.810622.3*4.4      
3 months 27.06.5 22.64.5      
Trust in extravaginal ejaculation and people to practice itBaseline 12.94.7 13.62.82.59NS8.31<0.013.76<0.05
1 month1411.24.710813.82.6      
3 months 11.13.9 13.72.6      
Feeling of rejection of extravaginal ejaculationBaseline 9.12.2 7.92.12.94NS1.36NS1.87NS
1 month147.73.61077.72.0      
3 months 8.42.9 8.01.9      
Attitudes towards sexDouble standards regarding sexual relationsBaseline 14.13.1 13.73.12.88NS0.55NS2.99NS
1 month1412.94.710713.13.3      
3 months 12.14.0 13.92.9      
Openness toward sexual relationsBaseline 14.04.8 13.42.53.55<0.050.64NS0.15NS
1 month1314.33.510913.72.5      
3 months 14.83.5 14.52.0      
Risk perception of pregnancyBecoming pregnant when using a condomBaseline 20.530.1 24.627.30.51NS2.11NS0.55NS
1 month1412.118.211124.827.2      
3 months 16.427.2 26.629.8      
Becoming pregnant when using extravaginal ejaculationBaseline 45.429.6 41.227.40.03NS2.04NS0.54NS
1 month1450.425.610838.124.9      
3 months 48.425.4 39.626.0      
Becoming pregnant when having sex without protectionBaseline 71.732.0 75.123.90.01NS0.28NS1.35NS
1 month1475.128.610871.828.4      
3 months 77.922.6 68.128.2      
Risk perception of HIVInfection when using a condomBaseline 18.327.1 20.425.50.17NS1.78NS1.00NS
1 month1411.519.010724.527.5      
3 months 15.327.5 25.430.1      
Infection when using extravaginal ejaculationBaseline 48.331.7 46.029.40.02NS0.63NS0.40NS
1 month1448.929.210644.730.6      
3 months 52.532.1 42.828.6      
Infection when having sex without protectionBaseline 72.734.6 69.028.30.07NS0.98NS0.98NS
1 month1470.929.810668.230.0      
3 months 76.427.6 62.432.0      
Risk perception of STDsInfection when using a condomBaseline 20.330.5 20.424.30.16NS2.26NS1.94NS
1 month1410.218.810625.928.9      
3 months 12.525.9 24.730.1      
Infection when using extravaginal ejaculationBaseline 52.736.1 45.228.90.02NS2.04NS0.15NS
1 month1453.633.510644.128.5      
3 months 55.636.8 43.627.9      
Infection when having sex without protectionBaseline 74.233.2 69.127.30.20NS1.40NS0.85NS
1 month1473.830.810668.926.5      
3 months 76.430.2 62.231.9      
Behavioral skills
Self-efficacySelf-efficacy of condom useBaseline 18.73.7 17.53.48.54<0.00119.21<0.0016.46<0.01
1 month1422.1*4.29917.33.5      
3 months 22.64.4 18.03.7      

The scores of 28 subjects in the intervention group for information, motivation, and behavioral skills immediately following the condom use educational program were compared to their baseline scores. Compared to the baseline survey, scores of the post-test survey were higher for knowledge relating to HIV/STDs (baseline, 11.9 ± 4.5; post-test, 16.4 ± 3.8; P < 0.001), relief/confusion associated with associated condom use (baseline, 11.6 ± 2.8; post-test, 12.9 ± 2.6; P < 0.01), self-efficacy of condom use (baseline, 19.0 ± 3.9; post-test, 22.6 ± 4.1; P < 0.001), and possibility of becoming pregnant when using extravaginal ejaculation (baseline, 36.0 ± 26.4; post-test, 49.0 ± 29.7; P < 0.05). Compared to the baseline survey, post-test survey scores were lower for avoidance of condom use (baseline, 15.5 ± 5.8; post-test, 12.4 ± 5.0; P < 0.001), trust in extravaginal ejaculation and people to practice it (baseline, 12.0 ± 4.2; post-test, 9.8 ± 3.8; P < 0.01), and possibility of STDs when using a condom (baseline, 12.9 ± 23.5; post-test, 8.8 ± 17.7; P < 0.05). A significant difference was also found between the post-test survey and the 1 month follow-up survey for trust in extravaginal ejaculation and people to practice it (post-test, 9.8 ± 3.8; 1 month, 11.3 ± 4.5; P < 0.05), with the 1 month score significantly higher. No difference was found between the post-test survey and the 1 month follow-up survey in other scales.

DISCUSSION

The condom use educational program conducted for the current study increased knowledge relating to HIV/STDs among people both with and without experience of sexual intercourse. The study provided participants with proper information regarding HIV/STDs, increased their knowledge immediately after the intervention, and sustained this knowledge at 1 and 3 months after the intervention. Knowledge alone does not change behavior, but knowledge is the most important factor in increasing the possibility of behavior being changed (Green, 2002). The present education program was able to provide the necessary knowledge for condom use during sexual intercourse.

In the intervention group, scores for self-efficacy of condom use were significantly higher 1 month after the program among both people with and without experience of sexual intercourse, and they were also significantly higher 3 months after the program among people with experience of sexual intercourse. Furthermore, because the scale scores were significantly higher immediately after the program, the education program developed for the present study increased self-efficacy, which has an effect on condom use. Self-efficacy of condom use includes using a condom correctly, carrying or purchasing a condom without feeling embarrassed, and using a condom regardless of who the sexual partner is. The present educational program provided knowledge about correct use of both male and female condoms via computer graphic images, which facilitated understanding of how to correctly use condoms. Moreover, the program provided participants with a group discussion of reasons for not using condoms and the opportunity to express their ideas on specific strategies to encourage condom use. It would appear that this program increased self-efficacy with regard to condom use.

The scores for avoidance of condom use in the intervention group decreased significantly immediately and 1 and 3 months after the present educational program. Avoidance of condom use indicates that buying, carrying, or using a condom is considered too much trouble, that using a condom is “uncool”, or that using a condom will “spoil the mood”. It appears that the present program was able to engender the recognition that the chain of action of buying, carrying, and using a condom is straightforward to perform. There was also a significant increase among participants with experience of sexual intercourse in the post-test scores for relief/confusion associated with condom use when envisaging the next sexual encounter with the most recent partner, with higher scores for responses indicating that the participant or the partner would be bothered by not using a condom, and that the participant would feel relieved if condom use were proposed. It would appear that the changes engendered by the present educational program were loss of condom use avoidance, a sense of security with condom use, and a feeling of disturbance when not using a condom. The program thus reduced resistance toward condoms.

Although the educational program was expected to increase the rate of condom use among participants, the number of people actually using a condom did not increase. We therefore investigated why the number of condom users did not increase. Condom use is an act involving sexual intercourse. Of the 157 participants, 33 (21.0%) had sexual intercourse at the 3 month survey. Of these 33, one from the intervention group and five from the control group had their first sexual intercourse after the baseline survey. This means that almost all participants with sexual experience had their first sexual intercourse before the baseline survey. Initiating condom use in a committed relationship can be interpreted as questioning commitment and interpersonal trust, so it may have been difficult to use a condom in a committed relationship (Jaworski & Carey, 2001). Jemmott, Jemmott, Braverman, and Fong (2005) suggested that it was difficult to introduce safer sex practices into existing relationships that had been continuous for 3 and 6 months. One possible explanation for why no increase was found in the number of condom users in the present study is that many of the study participants found it difficult to adopt the new behavior of condom use because they had already been having sexual intercourse within steady relationships. A second possible explanation is that a long-term relationship makes condom use even more difficult. People in long-term relationships are less likely to use condoms than people in short-term relationships, because an increased frequency of sexual intercourse enhances a partner's sense of interpersonal familiarity and similarity, thus leading to judgments of lower HIV risk and the perception of safety (Kordoutis, Loumakou, & Sarafidou, 2000). Because the present study had an evaluation period of 3 months, the steady relationships became long term, and condom use thus became more difficult. The third possible explanation is that the educational program did not provide sufficient practice in skills training to change behaviors. Jemmott et al. (2005) found that condom use was enhanced in a group given skills training that included handling condoms, practicing putting condoms on anatomical models, and role-playing a realistic situation that involves pressure to have unprotected intercourse, while a group that only received information about the skills necessary for condom use changed the mediating variables (e.g. attitudes and belief regarding using condoms) without changes in behavior. The present educational program was able to significantly change the mediating variables of self-efficacy of condom use and avoidance of condom use, but no change was seen in actual condom use. It seems likely that the program only provided information on the skills necessary for condom use. The fourth possible explanation is that the follow-up period for determining the effectiveness of the program was too short, and there were too few subjects for analysis. Jemmott et al. (2005) suggested that, although significant changes in mediating variables such as self-efficacy occurred 3 and 6 months after the intervention, a reduction in unprotected sexual behavior was found at 12 months but not at 3 or 6 months after the intervention. The effects of the intervention might increase at later follow ups. Furthermore, Kipke, Boyer, and Hein (1993) noted that, because sexual behaviors in adolescents are sporadic and erratic, the lack of changes in sexual behavior may reflect the short period for reporting sexual behavior and the small number of sexually active subjects. The present study's evaluations were 1 and 3 months after the program, so it is likely that the period was too short to observe changes in behaviors. The subjects for analyzing the effectiveness of the program were participants with sexual intercourse experience. In the present study, 15 participants in the intervention group and 18 participants in the control group had previous sexual intercourse; hence, the number was small.

Of the participants in the educational program, one (3.6%) had first sexual intercourse in the 3 month period of the survey, which was not statistically significant. In Japan, education regarding specific details relating to sexual intercourse including how to use condoms is not carried out for the reason that it might incite sexual intercourse among young people with no sexual experience. Although the present program illustrated correct condom use and taught participants that correct condom use is effective in preventing HIV/STD infection and for contraception, the sexual activity of the participants did not increase. It appears that teaching high school and university students specific details relating to sexual intercourse does not encourage sexual intercourse among those with no sexual experience.

Generalizing the results of the present study would be problematic because of the small number of participants that attended the program and responded to the whole survey, as well as the small number of participants that had sexual intercourse. Another limitation is that the survey was about self-reported sexual behavior. However, in Japan, no other studies have developed an educational program based on a theoretical model of behavioral change and performed a statistical investigation of the results. Because the present condom use educational program was able to change some of the psychosocial mediators that are involved in the behavior of young people, it will contribute to the development of future strategies to prevent STDs and unplanned pregnancies among young people in Japan.

CONCLUSION

The program used in this study increased knowledge relating to HIV/STDs and self-efficacy of condom use and reduced resistance to condom use. Further research should test the generalizability of these effects in a wider variety of young people in Japan. Because the number of people actually using a condom in the intervention group did not increase, improvement of the current intervention is needed, including handling condoms, practicing putting condoms on anatomical models, and role-playing a realistic situation that involves pressure to have unprotected intercourse, as well as evaluating effectiveness for longer than 3 months and increasing the number of subjects attending the program.

Theoretically-derived and empirically-based HIV preventive intervention can successfully promote the adoption of protective sexual behaviors among young people. This study provides empirical support for the IMB model of condom use behavior change. This model may produce an appropriate, comprehensive behavior change framework for Japanese young people. Furthermore, theory-based sexual education programs should be developed, and further research should be conducted to test the efficacy of such programs among youth in Japan.

ACKNOWLEDGMENTS

The author would like to sincerely thank the young people who took part in this study, as well as the teachers of the high schools and universities who allowed the study to be carried out. The author would also like to express sincere gratitude to Dr Aiko Yamamoto, Executive Director, Research Institute of Nursing Care for People and Community, University of Hyogo, Japan, for all the advice and guidance she gave. This study was supported, in part, by a Grant-in-Aid for Scientific Research (C) (no. 18592384) from the Japan Society for the Promotion of Science. The study is part of a doctoral thesis submitted to the College of Nursing Art and Science, University of Hyogo.

Ancillary