Examining the applicability of the Information-Motivation-Behavior skills model of HIV preventive behavior in Uganda
Download as PPT, PDF1 like1,226 views
1) The study examined the applicability of the Information-Motivation-Behavioral Skills model to predict condom use among sexually active adolescents in schools in Mbarara, Uganda. 2) Structural equation modeling found that the model predicted condom use for boys but not girls. For boys, subjective norms were predictive, while for girls only non-significant trends were found. 3) Implications are that females may require more tailored programs, as they are a more heterogeneous group. Also, providing more HIV prevention information may help strengthen the model, as levels of information were low.
Examining the applicability of the Information-Motivation-Behavior skills model of HIV preventive behavior in Uganda
- 1. Examining the Applicability of the Information-Motivation-Behavior Skills Model of HIV Preventive Behavior in Uganda Michele L. Ybarra, MPH, PhD1 Josephine Korchmaros PhD1 Julius Kiwanuka, MD2 1 Internet Solutions for Kids, Santa Ana, USA 2 Mbarara University of Science and Technology, Mbarara, Uganda AIDS Impact Gabarone, Botswana Thursday, September 24, 2009 * Thank you for your interest in this presentation. Please note that analyses included herein are preliminary. More recent, finalized analyses may be available by contacting CiPHR for further information.
- 2. Acknowledgement The project described was supported by Award Number R01MH080662 from the National Institute of Mental Health. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institute of Mental Health or the National Institutes of Health We would also like to thank the CyberSenga research team, especially Dr. Kimberly Mitchell and her team: Dennis Nabembezi, Ruth Birungi, and Tonya Prescott, for their rigorous implementation of the data collection and data entry activities.
- 3. Background: HIV in Uganda • HIV/AIDS is a major contributor to morbidity and mortality in Uganda, with an estimated 6% prevalence in the population (UNAIDS, 2005) • Recent data suggest a concerning increase in incidence (Shafer et al., 2006; Kamali, et al., 2002) • Among young people who had sex with a non- cohabitating partner, 38% of young men and 56% of young women reported not using a condom (UNAIDS, 2005)
- 4. Background: IMB Model The information-motivation-behavioral skills (IMB) Model of HIV Preventive behavior (Fisher & Fisher, 2005) Posits that a trilogy of one’s: • Information about how to prevent HIV, • Motivation to engage in non-risky behaviors, and • Skills and abilities in acting out these behaviors together predict HIV preventive behavior over time. IMB predicts HIV preventive behavior among adolescents and young adults in the United States (Fisher et al., 2002) and non-Western countries (Bryan, et al., 2000; Linn et al., 2001; Kalichman, 2005).
- 5. IMB Model HIV Preventive Behavioral Skills HIV Preventive Behavior HIV Preventive Information HIV Preventive Motivation Motivation includes subjective norms, behavioral intentions, and attitudes
- 6. The research question What is the applicability of the IMB model to predicting condom use among adolescents who are sexually active and attending secondary school in Mbarara, Uganda?
- 7. Mbarara Adolescent Health Survey Methodology • Mbarara, Uganda is the 6th largest urban center in Uganda • Five participating secondary schools: One Catholic, one Muslim, three government • Participants were randomly identified • Eligibility: • Current student in grades S1-S4 in one of our 5 partner schools • Parental consent was required for day students / Head master consent for boarding students • Youth assent
- 8. Mbarara Adolescent Health Survey Methodology • 1,503 S1-S4 students were surveyed cross-sectionally • Data were collected between September & October, 2008; and March & April, 2009. • On average, the survey took 1 hour to complete • Estimated response rate: 87.5% • Cognitive testing among ‘friends and family’ youth in S1 was done before fielding to ensure appropriate reading level. • Per CAB preferences, questions about condoms were only asked of youth reporting sexual activity
- 9. IMB Measures • HIV prevention information (Guttmacher; Uganda Demographic Health Survey) • 8 items, 1 point for each correct response • Example: HIV is small enough to go through a condom • HIV prevention motivation • HIV prevention subjective norms (Misovich et al.) • 11 items, 0-4 scale • Example: Friends that I respect think I should not have sex until I’m older • HIV prevention behavioral intentions (Misovich et al.) • 5 items, 0-4 scale • Example: I’m planning not to have sex until I’m older
- 10. Measures • HIV prevention behavioral skills (Misovich et al.) • 7 items, 0-4 scale • Example: How hard or easy would it be for you to make sure you do not have sexual intercourse until you’re older • HIV prevention behavior: Condom use during sex (MacPhail et al) • 1 item: How often do you use a condom during sex? 5- point scale • Dichotomized to: At least some of the time (1) vs. never (0)
- 11. Participant characteristics (n=1503) Personal characteristics All youth Abstinent (73%) Had sex (27%) Statistical comparison p- value Sex Male 62% 57% 74% X2(1) = 31.9 <0.001 Female 38% 43% 26% Age (Range: 12-19+) 14.9 14.7 15.4 t = -8.4 <0.001 Class "new" S1 25% 28% 15% X2(3) = 38.9 <0.001 S1 26% 26% 23% S2 25% 24% 27% S3 25% 22% 34% School type Boarding 86% 86% 84% X2(1) = 0.72 0.4 Day 14% 14% 16%
- 12. Condom use among sexually active • How often do you use a condom during sex? • Never: 58% • Less than half the time: 7% • Half of the time: 6.5% • More than half of the time: 3% • Always: 25% • 73% of those sexually active know where they can buy condoms • 48% of those sexually active know where they can get condoms for free
- 13. SEM Analysis: Condom Use HIV Preventive Behavioral Skills Condom use HIV Preventive Information HIV Preventive Subjective Norms HIV Preventive Behavioral intentions Model fit: χ2 (4) = 3.23, p = .519 CFI = 1.0 RMSEA = .000 r = .71*, . 85* β = .08, -.07 β = .42*, .27 β = .33*, .50* β = .27*, .35 β = .19*, -.05 β = -.15, .25 β = -.03, .12 Boys, Girls
- 14. Summary The I is not predictive for either boys or girls The M is somewhat predictive: for boys through subjective norms; similarly strong but non-significant trends for girls neither boys nor girls through behavioral intentions The B is predictive for boys but not for girls Other factors are not predictive of condom use: self- esteem, social support, orientation to the future, physical health
- 15. Limitations Analyses do not reflect recency of sex: the adolescent could have had sex 5 years ago and been abstinent since then Sexual activity is a highly stigmatized behavior in Uganda; likely under-reporting based upon self- report Unable to examine the IMB model to predict abstinence vs. sexual activity due to cultural sensitivities
- 16. Implications o Females appear to be a more heterogeneous group than males when trying to predict condom use. This may require more program tailoring. o HIV information is uniformly low, suggesting that more attention needs to be paid to this area of the IMB model.
- 17. Implications o The IMB model is better able to predict boys’ reports of condom use than girls’. o It is possible that the areas that are not predictive for condom use are the areas that we need to attend to most in the intervention. o E.g., Boost the behavioral skills boost the link to condom use
Editor's Notes
- #6: According to IMB, motivation includes subjective norms, behavioral intentions, and attitudes. We did not measure attitudes, which is why it is not included in our analyses. I changed the text on this slide
- #14: Results of model without including any covariates influencing the ultimate outcome (condom use). All covariates including perceived health did not influence condom use—hence no slide displaying those results Model fit well: Chi square was statistically insignificant, CFI was between .90 and 1.0 and RMSEA was between .10 and 0.0 Blue coefficients are for males, red are for females Only those designated with an asterisk are statistically significant at the .05 level Results: predicting condom use: for both males and females, (1) HIV info and intentions were not reliably related to condom use, (2) HIV info was not reliably related to skills, (3) intentions were positively and strongly related to skills. Males and females differed: (1) for males norms were positively and strongly related to skills and condom use, (2) for males, skills were positively related to condom use The large unreliable (insignificant) beta coefficients for females suggest that there is more variation among females than for males. There may be effects there that we do not have enough power to detect