Predicting condom use in adolescents: a test of three socio-cognitive models using a structural equation modeling approach

BACKGROUND: The theory of planned behavior (TPB), socio-cognitive model (SCM), and information-motivation-behavioral skills (IMB) model are effective in predicting condom use. However, the adequacy of these three theoretical models in predicting the frequency of condom use (FCU) among young people has not been compared. This cross-sectional study tested the applicability and suitability of these three models in predicting the FCU, and analyzed the relationships among the postulated constructs. METHODS: Sexually experienced adolescents (n = 410) aged 13–18 completed a survey assessing the TPB, SCM, and IMB model constructs. Participants were students recruited from 18 high schools, randomly selected from the north, south, east, and southeast of Spain. A structural equation modelling (SEM) analysis was applied to test TPB, SCM and IBM and constructs relationships of each model using R. RESULTS: The results of SEM demonstrated that behavioral skills predict behavior via motivation as hypothesized by the IMB model, but not directly via knowledge about condom use and sexually transmitted infections (STIs). Cognitive factors, such knowledge about condom use and STIs as well as condom use self-efficacy, directly predicted the FCU when modeled as per the SCM. According to the TPB, condom use intention was the best predictor of the FCU, and condom use intention was predicted by attitudes toward condom use and subjective norms related to condom use, but perceived control was not directly or indirectly related to the FCU. Based on the data, the TPB becomes the best-fit model for predicting the FCU among young people compared to the SCM and IMB model. CONCLUSIONS: From a statistical perspective, the TPB seems to be the most suitable model for predicting the FCU among young people compared to the other models. Overall, key direct predictors of the FCU in adolescents included condom use intention, behavioral skills and cognitive factors, such as STIs knowledge and condom use self-efficacy. The next step should be to test integrative models that include personal, contextual, environmental, and social factors. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (doi:10.1186/s12889-016-2702-0) contains supplementary material, which is available to authorized users.