Condom Use Intentions and Behavior Among Older Adolescents: Application of the Health Belief Model through Factor Analysis

A significant amount of progress has been made over the past decade to reduce the impact of risky sexual behavior among adolescents; however, rates of unwanted pregnancy and contraction of sexually transmitted infections remain high. Finding ways to mitigate the consequences of risky sexual behavior continues to be a focus of many working with adolescents engaging in risky behavior. This study performed a secondary analysis of data collected during a 5-year evaluation of a teen pregnancy program targeting youth ages 17-19 who were at high risk for dropping out of high school. 

The goals of this study were to test the applicability of the Health Belief Model (HBM) for assessing both adolescent intentions to use condoms and condom use behavior. Findings indicated that participants who reported greater intentions to use condoms were more than twice as likely to report using a condom in the past three months. Factor analyses guided by the Health Belief Model framework identified a 26-item, six factor model that predicted 63% of the variance in the model. This was followed by ordinal and logistic regressions to detect the relationship between each of the EFA factors and condom use intentions, as well as between each of the EFA factors and condom use behavior. The results demonstrated that five of the newly identified factors (i.e., Partner Efficacy, Interpersonal Barriers – Partner Trust, Structural Barriers – Accessibility, Benefits, and Physical Barriers – Comfort) significantly predicted intentions to use condoms (p < .05). Three of the factors (i.e., Interpersonal Barriers – Partner Trust, Structural Barriers – Accessibility, and Physical Barriers – Comfort) also significantly predicted a positive relationship between the factor and condom use behavior (p < .05).

These findings confirm that some of the factors developed guided by the Health Belief Model did fit well for this population comprised of older academically at-risk adolescents, while some of the factors exhibited limitations. Suggested revisions to items, inclusion of new items, and removal of weaker items that may lead to an improved model will be discussed.