COLLEGE STUDENTS IN NIGERIA UNDERESTIMATE THEIR RISK OF CONTRACTING HIV/AIDS INFECTION

Kayode T. Ijadunola1, MBChB, MSc, FWACP; Titilayo C. Abiona1, MBChB,
FMCPH;

Olusola O. Odu2, MBChB, FMCPH; Macellina Y. Ijadunola1, MBChB

1Department

of Community Health, College of Health Sciences, Obafemi

Awolowo University, Ile-Ife, Nigeria

2Department

of Community Medicine, Faculty of Clinical Sciences,

Ladoke Akintola University of Technology, Osogbo, Nigeria

Address for Correspondence: Dr. Kayode T. Ijadunola Department of Community Health College of Health Sciences Obafemi Awolowo University P.O. Box 2064, OAU Post Office Ile-Ife, Osun State, Nigeria E-mail: kijadun@yahoo.com Phone: 234 803 725 0846

Running title: College students and HIV/AIDS risk perceptions Key words: College students, risk perceptions, HIV/AIDS infection, Nigeria

Abstract Objectives: This study aimed at assessing the perceptions of personal risk of acquiring HIV/AIDS infection among students of selected tertiary institutions in Osun State, Nigeria and determining the correlates of perceptions of personal risk of infection.

Methods: The study employed a cross-sectional descriptive design. An interviewer-administered questionnaire was applied to the 405 study participants and correlates of perceptions of personal risk of HIV infection were evaluated using multiple logistic regression analysis, confidence intervals and odd ratios.

Results: Only 15% of the students perceived themselves to be at moderate-to-high risk of acquiring HIV infection compared with 85% who perceived themselves to be at little or no risk. Investigators’ assessment of risk status of the participants revealed that 77% of the participants were actually at high risk of infection compared with only 33% that were at low risk. The sensitivity, specificity and predictive value negative of the students’ self perception of personal risk compared with their assessed risk were 7%, 58% and 16% respectively, with a Kappa statistic of 0.178. The only significant correlate of self perception of personal risk was recent symptoms of sexually transmitted infections (STIs).

Conclusions: College students exhibited high rates of HIV risk indicators but low levels of perceived personal risk of infection in favour of an “optimism bias”. This has implications for HIV/AIDS control in Nigeria.

2

Introduction The Nigerian HIV/AIDS epidemic has crossed into the explosive phase, already advancing well beyond high risk groups and into the general population. The official adult prevalence rate is about 5%, but unofficial estimates range as high as 10% - which represents about 4-6 million infections. Infections are most numerous among men aged 20 to 24 years, but some experts caution that infection rates are rising quickly in young women1. While HIV/AIDS education programs for schools and youth groups have proliferated in recent years, the scientific foundation for such programs is quite limited and an evaluation of their impact has been rare, in both industrialized and developing countries2. Identification of the correlates of perceptions of personal risk (threat) of contracting HIV infection among young people is likely to better inform a behaviour change communication intervention more than a stopgap education program that is not based on empirical evidence of the perspectives of the youth.

The construct of perceived threat is an important component in behavioural theories that are applied to understanding and changing sexual risk behaviours that lead to sexually transmitted infections including HIV/AIDS3. For example, the Information-Motivation-

Behavioural Skills model, as applied to HIV infection, posits that the perceived threat of HIV infection contributes to individual motivation to adopt protective behaviour4. However, the perceived threat is still an understudied construct in the field of HIV infection and sexually transmitted infections (STIs)5-6. Perceived threat has been conceptualized as a product of perceived risk and perceived severity relative to a given disease or event7. Adolescents and young people are likely to

underestimate both the severity and their risk of adverse health outcomes5,8-10. This underestimation may, in part, be a mental defense mechanism. The tendency to systematically underestimate personal risk 3

has been termed “optimism bias”7. Quadrel et al11 found that high school students perceived themselves to be at less risk compared with peers for alcoholism and unwanted pregnancy. Moore and Rosenthal found that college students are optimistically biased in their perceptions of relative risk for STIs and AIDS.12

Previous studies have however varied widely in their measurement of perceived threat, with constructs such as perceived vulnerability, susceptibility, and worry being assessed by reserachers6. An important aspect of perceived threat is whether the perception of threat creates worry (an indication of dissonance). Dissonance may be a strong motivating factor for behaviour change, particularly if the individual perceives control over the risk behaviour13. Alternatively, intellectual perceptions of susceptibility without feelings of dissonance are less likely to motivate behaviour change. Few published data exist regarding correlates of young peoples’ STI and HIV worry. Some of the studies recruited subjects from STI and reproductive health clinics, and thus, potentially limited the generalizability of the findings. Little is therefore known about the determinants of perceptions of personal risk for HIV/AIDS and STIs among young people generally. These may be related to their past condom use, their number of lifetime sexual partners and their beliefs that they can prevent and or recognize individuals who are infected with STI or HIV/AIDS12,14.

This study was conducted among undergraduate students of selected tertiary educational institutions in Osun State of Nigeria with the aim of assessing the self perceptions of risks for acquiring HIV infection among the general population of young people, determining the validity of their perceived risk status and investigating the reproductive health correlates of self perceptions of risks of HIV infection among the respondents.

4

Methods Study location: Study was carried out in two tertiary educational institutions located in Ile-Ife and Ilesa in Osun State, Southwest, Nigeria. One was a State College of Education with a student enrollment of about 5,000 and the other was a private polytechnic with a student enrollment of about 4,000. Students of both institutions make private arrangements for accommodation as the schools are non-residential.

Study design: The study employed a cross-sectional descriptive design.

Sample size determination: Sample size was estimated from the Computer Programs for Epidemiologists (PEPI) version 3.01 by employing the sample size formula for estimation of a single proportion as described by Armitage and Berry and cited in Abramson and Gahlinger (1999)15. Assuming a 95% level of confidence, an estimate of true proportion of perception of personal risk of HIV among young people of 50% (no previous estimate found in literature), and a maximum acceptable difference from true proportion of 5%, the minimum sample size estimated was 385. This was eventually upgraded to 400, but 405 participants were eventually interviewed.

Sampling technique: This was by multi-stage sampling procedure. The sampling frame consisted of all 6 tertiary educational institutions in the state. Two of these were randomly selected by balloting. From each selected institution, two faculties were randomly selected from the lists of constituent faculties. From each selected faculty, two departments were randomly selected from the list of constituent departments. From each selected department, two years of study were randomly selected and all students in each selected year of study were interviewed in a classroom setting.

5

Data collection methods: A semi-structured questionnaire was applied to each participating student face-to-face by final year medical students who were recruited and trained on the instrument. The questionnaire specifically asked questions on respondents’ biodata, their perceptions of HIV/AIDS and knowledge of the modes of transmission, the sexual practices and indicators of risks of contracting HIV. The ethics and research committee of the local university teaching hospital approved the research protocol and study instruments. A written consent was obtained from each participating student prior to data collection.

Data analysis: Data were analyzed with the use of the SPSS statistical software package. Descriptive statistics was used to present respondents’ demographic variables, reproductive health history, selected indicators of risk of acquiring HIV and perceptions of HIV risk status. We conducted multiple logistic regression analyses to examine the odds of perceived susceptibility to acquiring HIV infection that was associated with selected indicators of risk for HIV infection namely: recent history of unprotected sex, history of multiple sexual partners, past history of STIs, recent symptoms of STIs. The indicators of risk used in this study were similar to those described by Ethier et al. for sexually transmitted infections16. We calculated the validity (sensitivity, specificity and predictive value negative) of the respondents’ perceived risk statuses against the assessed risks and the degree of agreement (Kappa statistic) with the investigators’ assessment of risk.

OUTCOME MEASURES Unprotected sex Participants were categorized as engaging in unprotected sex in the month preceding the survey if their reported episodes of condom use were less than 100%.

6

Lifetime sexual partners Participants were categorized as having >4 or 4 sexual partners (n=405)