Behavioral and subjective health changes in US and Mexico border residing participants in two promotora-led chronic disease preventive interventions.

Chronic diseases are the primary health burden among Mexican-origin populations and health promotion efforts have not been able to change negative population trends. This research presents behavioral and subjective health impacts of two related community health worker (CHW) interventions conducted in the US-Mexico border region. Pasos Adelante (United States) and Meta Salud (Mexico) are 12-13 week CHW-led preventive interventions implemented with Mexico-origin adults. Curricula include active learning modules to promote healthy dietary changes and increasing physical activity; they also incorporate strategies to promote social support, empowerment and group exercise components responsive to their communities. Questionnaire data at baseline (N = 347 for Pasos; 171 for Meta Salud), program completion and 3-month follow-up were analyzed. Results showed statistically significant improvements in multiple reported dietary, physical activity and subjective health indicators. Furthermore, at follow-up across both cohorts there were ≥10% improvements in participants' meeting recommended physical activity guidelines, consumption of whole milk, days of poor mental health and self-rated health. While this study identifies some robust health improvements and contributes to the evidence base for these interventions current dissemination, the lack of change observed for some targeted behaviors (e.g. time sitting) suggests they may have stronger overall impacts with curricula refinement.

Psychosocial predictors of delay of first sexual intercourse by adolescents.

This investigation predicted adolescents' delay of intercourse onset from attitudes, social norms, and self-efficacy about refraining from sexual intercourse. Age, gender, ethnicity, and parental education were also examined as predictors and moderators of the relationships among the 3 psychosocial determinants and onset. The participants (N = 827), part of a cohort initially surveyed in the 9th grade, reported at baseline that they had never engaged in intercourse. The multivariable proportional hazards regression model suggested that adolescents with more positive attitudinal and normative beliefs, as well as those with a parent who graduated from college, were less likely to engage in intercourse in the follow-up period (up to approximately 2 years). Interventions that include an objective to delay onset may benefit from addressing psychosocial determinants, especially attitudes and norms about sexual intercourse.

PIP: This study predicted adolescent's delay of intercourse onset from attitudes, social norms, and self-efficacy about refraining from sexual intercourse. Age, gender, ethnicity, and parental education were also examined as predictors and moderators of the relationships among the three psychosocial determinants and onset. The longitudinal data for the study were obtained from 827 participants in the US who were part of a cohort initially surveyed in the 9th grade. These participants reported at baseline that they had never engaged in intercourse. Utilizing the multivariable proportional hazards regression model, findings suggested that adolescents with more positive attitudinal and normative beliefs were less likely to engage in intercourse in the follow-up period (up to approximately 2 years). This was also the case for those students with a parent who graduated from college. Attitudes and norms were the most robust predictors of intercourse. In addition, a relatively modest increase in either scale was predictive of a 30% reduction in the onset of future intercourse in the most conservative analytic model. Interventions that include an objective to delay onset may benefit from addressing psychosocial determinants, especially attitudes and norms about sexual intercourse.

Psychosocial determinants of the onset and escalation of smoking: cross-sectional and prospective findings in multiethnic middle school samples.

PURPOSE: To investigate a broad range of social influence-related and global determinants of smoking to aid in the design of comprehensive multiethnic interventions by testing the most important factors of initiation and escalation of smoking across various subgroups. METHODS: Cross-sectional (N = 2546) and cohort (N = 736) samples of multiethnic middle school students near a large Southwestern metropolis were surveyed through self-report questionnaires. The confidential questionnaires included information on demographics, risk factors, and smoking behavior and were administered in class by trained data collectors. Multivariable logistic regression analysis was used to examine the statistical significance and strength of the factors. RESULTS: Those lower in self-esteem and higher in social assertiveness appeared to be most at risk for the onset of smoking, whereas those low in optimism appeared to be the most at risk for the escalation of smoking. Attitudes, friends' norms, parents' norms, perceived behavioral control, and perceived prevalence were consistent predictors of all smoking status outcomes. CONCLUSIONS: The behavioral-specific determinants of smoking appear to be important predictors of smoking status outcomes in all demographic subgroups. The relationships of the global determinants were more dependent on the smoking outcome variable and subgroup examined. The findings may serve to help facilitate the targeting of comprehensive interventions aimed at reducing adolescent smoking in multiethnic and ethnic group-specific populations.

Older boyfriends and girlfriends increase risk of sexual initiation in young adolescents.

PURPOSE: To explore the prevalence and impact of older boyfriends or girlfriends on sexual behavior in sixth graders (mean age 11.5 years). METHODS: Students in 19 ethnically diverse middle schools in an urban area were surveyed (n = 2829, response rate 68%). Instrument measured demographics, age of oldest boyfriend or girlfriend, unwanted sexual advances, peer norms, and sexual behavior. Students with older, same-age, or no boyfriend or girlfriend were compared on demographic and psychosocial variables using analysis of variance. Separate multivariate logistic regressions for both boys and girls were used to predict sexual behavior from demographics, psychosocial variables, and age categories of boyfriend or girlfriend. RESULTS: One-half of the respondents (56%) had never had a serious boyfriend or girlfriend, 35% reported that their oldest boyfriend or girlfriend was <2 years older than they, and 8.5% reported a partner > or =2 years older. Those reporting an older boyfriend or girlfriend were more likely to be Hispanic, were less acculturated, reported more unwanted sexual advances and more friends who were sexually active, and, among girls, were more likely to have experienced menarche. Overall, 4% of students reported ever having had sex. Students with an older boyfriend or girlfriend were over 30 times more likely than those with no boyfriend or girlfriend ever to have had sex (odds ratio = 33.8 for boys and 44.2 for girls). In the multivariate logistic regressions, peer norms about sexual behavior, having experienced unwanted sexual advances, and having a boyfriend or girlfriend were strongly associated with having had sex. CONCLUSIONS: Having an older boyfriend or girlfriend, although rare, is associated with early sexual onset and unwanted sexual activity in this population of sixth graders.

Schoolwide effects of a multicomponent HIV, STD, and pregnancy prevention program for high school students.

Few studies have tested schoolwide interventions to reduce sexual risk behavior, and none have demonstrated significant schoolwide effects. This study evaluates the schoolwide effects of Safer Choices, a multicomponent, behavioral theory-based HIV, STD, and pregnancy prevention program, on risk behavior, school climate, and psychosocial variables. Twenty urban high schools were randomized, and cross-sectional samples of classes were surveyed at baseline, the end of intervention (19 months after baseline), and 31 months afterbaseline. At 19 months, the program had a positive effect on the frequency of sex without a condom. At 31 months, students in Safer Choices schools reported having sexual intercourse without a condom with fewer partners. The program positively affected psychosocial variables and school climate for HIV/STD and pregnancy prevention. The program did not influence the prevalence of recent sexual intercourse. Schoolwide changes in condomuse demonstrated that aschool-based program can reduce the sexual risk behavior of adolescents.

Health risk behaviors of Texas students attending dropout prevention/recovery schools in 1997.

This study determined prevalence of health risk behaviors of 9th through 12th grade students attending dropout prevention/recovery alternative schools in Texas in 1997. Participants were 470 youth whose health risk behaviors were assessed using the Youth Risk Behavior Survey in an anonymous, self-administered format. Behaviors measured included frequency of weapon-carrying and fighting, suicide-related behaviors, substance use, and sexual behaviors. A substantial percentage of alternative school students reported participating in behaviors that placed them at acute or chronic health risk. Differences in the prevalence of risk behaviors were noted by gender, racial/ethnic, and age subgroups. In addition, alternative school students frequently engaged in multiple risk behaviors. These findings suggest a need for comprehensive school-based health education/intervention programs to reduce the prevalence of risk behaviors in populations of alternative school students.

Multilevel Models and Unbiased Tests for Group Based Interventions: Examples from the Safer Choices Study.

For many large-scale behavioral interventions, random assignment to intervention condition occurs at the group level. Data analytic models that ignore potential non-independence of observations provide inefficient parameter estimates and often produce biased test statistics. For studies in which individuals are randomized by groups to treatment condition, multilevel models (MLMs) provide a flexible approach to statistically evaluating program effects. This article presents an explanation of the need for MLM's for such nested designs and uses data from the Safer Choices study to illustrate the application of MLMs for both continuous and dichotomous outcomes. When designing studies, researchers who are considering group-randomized interventions should also consider the features of the multilevel analytic models they might employ.

Models of the relationship of stress, depression, and other psychosocial factors to smoking behavior: a comparison of a cohort of students in grades 6 and 8.

BACKGROUND: Six specific hypotheses regarding putative mechanisms by which stressful life events might lead to initiation of smoking among adolescents were proposed and tested on a Grade 6 cohort of students in Scarborough, Ontario, Canada. In addition, the data were used to determine the set of risk factors for initiation of smoking most pertinent to the experience of the cohort. METHODS: The same relationships were examined for the 1,543 students when they were in Grade 8 and compared to the earlier Grade 6 results. The hypotheses include the effects of personal resources (coping, self-esteem, social support, and mastery), social conformity, rebelliousness, attitudes, smoking environment factors, and gender differences. RESULTS: The hypotheses were not unequivocally supported, except for the hypotheses about attitudes and smoking environment as well as gender effects. Males and females differ with regard to the variables and interrelationships in both years and in the final models developed. In Grade 6, there are more smoking environment items for males than for females. By Grade 8, male smoking is influenced by mastery, social conformity, and rebelliousness, while for females environmental smoking and rebelliousness are important. CONCLUSION: Male and female students differ in how stress, depression, and smoking are related in the presence of psychosocial factors.