Individual Submission Summary
Share...

Direct link:

Poster #107 - Parental Attachment and Patterns of Adolescents’ Health-Risk Behaviors

Thu, March 21, 4:00 to 5:15pm, Baltimore Convention Center, Floor: Level 1, Exhibit Hall B

Integrative Statement

Late adolescence is a critical time for establishing lasting health-behavior patterns that are associated with long-term disease risk (Kang et al., 2014). However, little research has considered patterns of adolescents’ engagement across different forms of health-risk behavior, which is problematic, as the co-occurrence of multiple health-risk behaviors increases the risk of chronic disease and mortality (Mays et al., 2012). In addition, recent research has emphasized the importance of identifying precursors to adolescents’ engagement in health-risk behaviors to inform potential interventions (Laska et al., 2009). Previous research has identified adolescents’ insecure attachment to parents as a predictor of greater engagement in specific health-risk behaviors examined in isolation, such as substance use (Khallad, 2010; Labrie & Sessoms, 2012). The current study extended previous research by using a person-centered approach to assess patterns of adolescent health-risk behaviors and whether different risk-behavior profiles are differentially associated with attachment to parents.

Participants were 580 incoming college students (Mage=18.13, SD=.93; 96.7% White; 30.3% male) attending a large mid-Atlantic university. During the summer before freshman year, participants completed questionnaires assessing avoidant and anxious attachment to mothers and fathers (α = .91-.93) and engagement in seven health-risk behaviors: binge drinking, marijuana use, tobacco use, physical activity, body mass index (BMI), unprotected vaginal intercourse, and unprotected anal intercourse.

Health-risk behaviors were submitted to a latent class analysis (LCA), which yielded five latent classes based on an examination of AIC, BIC, SSABIC, and VLMR. Class 1 (2% of sample) scored high on unprotected anal intercourse and BMI, Class 2 (60%) scored low on substance use, unprotected anal and vaginal intercourse, and BMI, Class 3 (14%) scored high on substance use and physical activity, Class 4 (10%) scored high on substance use, unprotected anal and vaginal intercourse, and BMI, and Class 5 (14%) scored high on marijuana use, tobacco use, and unprotected vaginal intercourse (Figure 1).

A one-way between-subjects Multivariate Analysis of Variance (MANOVA) examined associations between classes of health-risk behaviors and attachment to parents while controlling for socioeconomic status (SES). A significant Pillai’s trace indicated a main effect for classes of health-risk behaviors, F(16, 2084)=3.59, p =.007, partial 2=.029. The follow-up univariate ANOVA indicated that Class 4 reported higher levels of maternal avoidant attachment than Classes 2, 3, and 5 (Table 1).

The current findings suggest that adolescents, even before entering college, engage in diverse patterns of health-risk behaviors that vary by maternal avoidant attachment. Researchers are urged to use person-centered approaches to examine risk behaviors, as our findings suggest that some adolescents may be experimenting in a smaller range of risk behaviors, whereas others are engaging in a wide range of risk behaviors. Additionally, the findings suggest that avoidant attachment, wherein individuals have difficulty relying on their mothers for support, is associated with engagement in health-risk behavior for adolescents engaging in numerous risk behaviors compared to adolescents who do not engage in any risk behaviors or those who experiment in specific types of risk behaviors only.

Authors