Introduction
Uptake of HIV testing is central to the diagnosis of HIV infection and early initiation of antiretroviral therapy (ART), which are necessary to halt the spread of the disease and prevent HIV-related mortality (Girardi et al., Reference Girardi, Sabin and Monforte2007). Yet many HIV infections remain undiagnosed, especially in sub-Saharan Africa and particularly among adolescents and young adults (White et al., Reference White, Warren, Scribner and Frazee2009; Deblonde et al., Reference Deblonde, De Koker, Hamers, Fontaine, Luchters and Temmerman2010; Staveteig et al., Reference Staveteig, Croft, Kampa and Head2017; Wanyenze et al., Reference Wanyenze, Kamya, Fatch, Mayanja-Kizza, Baveewo and Sawires2011). There is evidence that most recently diagnosed people present with late-stage disease (Girardi et al., Reference Girardi, Sabin and Monforte2007; Burns et al., Reference Burns, Johnson, Nazroo, Ainsworth, Anderson and Fakoya2008; Wanyenze et al., Reference Wanyenze, Kamya, Fatch, Mayanja-Kizza, Baveewo and Sawires2011; Camoni et al., Reference Camoni, Raimondo, Regine, Salfa and Suligoi2013). Delaying testing for HIV results in late diagnosis, which has deleterious consequences not only for the patient, but also for the population at large (Girardi et al., Reference Girardi, Sabin and Monforte2007). In other words, individuals who are diagnosed late are often ill, have a high risk of mortality, tend to respond poorly to treatment and cost more to treat. Individuals who remain untested aid the furtherance of the pandemic nature of the disease by failing to adopt preventive measures and the behavioural patterns required of them to curtail onward transmission of the disease (Marks et al., Reference Marks, Crepaz, Senterfitt and Janssen2005).
Despite having the largest burden of HIV, uptake of HIV testing – especially among young adults – in many sub-Saharan African countries remains suboptimal (Kalichman & Simbayi, Reference Kalichman and Simbayi2003; Vermeer et al., Reference Vermeer, Bos, Mbwambo, Kaaya and Schaalma2009; Ogaji et al., Reference Ogaji, Oyeyemi and Ibrahim2013). Only an estimated 15% and 10% of young men and women, respectively, know their HIV status in sub-Saharan Africa (World Health Organization, 2013). Specifically, the prevalence of HIV is 3.2% in Nigeria, but due to its huge population the country has the second largest number of people living with HIV (PLHIV) worldwide – an estimated 3.2 million (Awofala & Ogundele, Reference Awofala and Ogundele2018). Yet the uptake of HIV testing is relatively low in Nigeria, especially among adolescents and youths (Ogaji et al., Reference Ogaji, Oyeyemi and Ibrahim2013; Amu et al., Reference Amu, Olatona and Onayade2014; Olowokere, et al., Reference Olowokere, Adelakun and Komolafe2018). Only 34% of people living with HIV in Nigeria know their status compared with 75% globally (UNAIDS, 2014).
It has been documented previously that perceived physical health deterioration or death of a loved one, roll-out of various HIV testing initiatives, availability of treatment, social network and support and guarantee of anonymity and confidentiality are all enabling factors for the uptake of HIV testing (Deblonde et al., Reference Deblonde, De Koker, Hamers, Fontaine, Luchters and Temmerman2010; Musheke et al., Reference Musheke, Ntalasha, Gari, Mckenzie, Bond, Martin–Hilber and Merten2013). In contrast, perceived low risk of contracting HIV, fear of stigmatization, lack of access and the perceived psychological burden of living with HIV are reported barriers to the uptake of HIV testing (Meiberg et al., Reference Meiberg, Bos, Onya and Schaalma2008; Deblonde et al., Reference Deblonde, De Koker, Hamers, Fontaine, Luchters and Temmerman2010; MacKellar et al., Reference Mackellar, Hou, Whalen, Samuelsen, Sanchez and Smith2011; Turan et al., Reference Turan, Bukusi, Onono, Holzemer, Miller and Cohen2011; Prestage et al., Reference Prestage, Brown and Keen2012; Musheke et al., Reference Musheke, Ntalasha, Gari, Mckenzie, Bond, Martin–Hilber and Merten2013). Although the cost of HIV testing is relatively low in Nigeria, access to the test is limited, especially in rural areas (Federal Ministry of Health, 2013; NACA, 2014). For instance, over 30% of adolescents and young adults do not know where to access HIV testing (Federal Ministry of Health, 2013).
Worldwide, adolescents and youths (aged 15–25 years) account for the highest proportion of new HIV infections (UNAIDS, 2012; World Health Organization, 2013). What is more, this age cohort are more likely not to have tested for HIV, have the lowest level of adherence to ART and viral load suppression and are more likely to die from HIV/AIDS-related causes. To reverse this trend and eliminate the AIDS epidemic by 2030, 90% of PLHIV have to know their HIV status, 90% of those who have received an HIV diagnosis have to receive ART and 90% of those on ART have to be virally suppressed (Joint United Nations Programme on HIV/AIDS, 2014). Understanding factors that predict the uptake of HIV testing among adolescents and young adults is critical to developing relevant interventions for this cohort.
Risk perception is a core construct of the Health Belief Model (HBM), which is widely utilized in health behaviour research (Glanz et al., Reference Glanz, Rimer and Viswanath2008; Painter et al., Reference Painter, Borba, Hynes, Mays and Glanz2008; Hayden, Reference Hayden2013). According to the HBM, individuals who have a higher perceived risk of contracting an illness are more predisposed to risk reduction behaviours, including treatment utilization (Glanz et al., Reference Glanz, Rimer and Viswanath2008; Painter et al., Reference Painter, Borba, Hynes, Mays and Glanz2008). In other words, perceived susceptibility to an illness, perceived severity of the illness and perceived benefits of treatment all influence risk reduction behaviours and treatment utilization (Glanz et al., Reference Glanz, Rimer and Viswanath2008; Painter et al., Reference Painter, Borba, Hynes, Mays and Glanz2008). However, empirical evidence on the association between HIV risk perceptions and the uptake of HIV testing is mixed (Khawcharoenporn et al., Reference Khawcharoenporn, Chunloy and Apisarnthanarak2016). While there are studies that show that low HIV risk perception is associated with not testing for HIV (Marcus et al., Reference Marcus, Gassowski and Drewes2016; Olowokere et al., Reference Olowokere, Adelakun and Komolafe2018), another study did not find this association (Sisay et al., Reference Sisay, Erku, Medhin and Woldeyohannes2014). Thus, there is a gap in our understanding of the association between HIV risk perceptions and the uptake of HIV testing. Many studies have defined HIV risk perception as an individual’s perception of his/her likelihood of contracting HIV (Marcus et al., Reference Marcus, Gassowski and Drewes2016; Olowokere et al., Reference Olowokere, Adelakun and Komolafe2018). The shortcoming of this definition is that an individual may have concerns about contracting HIV, despite rating himself/herself as unlikely to contract HIV. Studies have shown that people tend to underestimate their HIV acquisition risk (Khawcharoenporn et al., Reference Khawcharoenporn, Chunloy and Apisarnthanarak2016). While people may underestimate their risk of contracting HIV (an often stigmatized disease), the chances are that they will accurately report whether they are worried or concerned about contracting HIV. The relationship between concerns about contracting HIV and uptake of HIV testing is less understood.
Discussion of HIV and STIs with a sexual partner represents an important step in self-risk assessment, or at least serves as an important indication of the occurrence of risk assessment. In addition, insisting on knowing a sexual partner’s HIV status could facilitate couple testing and promote positive behaviour among adolescents. Drawing from the psychosocial constructs of the Health Belief Model, this paper examines the effects HIV risk perceptions, discussion of HIV with partners and knowing partners’ status on HIV testing uptake among adolescents and young adults in two Nigerian universities. It also assesses the progress towards achieving the first of the three 90–90–90 UNAIDS targets by focusing on university students in Nigeria, characterized as engaging in risky sexual behaviours including unprotected sex with multiple sexual partners (Imaledo et al., Reference Imaledo, Peter-Kio and Asuquo2012; Noubiap et al., Reference Noubiap, Nansseu, Ndoula, Wang, Jingi and Bigna2015; Ajayi et al., Reference Ajayi, Akpan, Goon, Nwokocha and Adeniyi2016, Reference Ajayi, Nwokocha, Akpan, Adeniyi and Goon2018).
Methods
This cross-sectional study was conducted between February and April 2018 in two universities in North–Central Nigerian: the University of Ilorin, run by the federal government, and Nasarawa State University, run by the state government. These universities were selected purposively. Nasarawa State not only has a high prevalence of HIV (7.5%), but shares boundaries with states that have the highest prevalence of HIV in Nigeria (Federal Ministry of Health, 2013). Most students in Nasarawa State University are from Nasarawa and neighbouring states. In contrast, the University of Ilorin has a highly diverse group of students. The two universities combined have an approximate total number of 45,000 students.
A self-designed validated questionnaire was administered to 800 male and female students using a face-to-face interview method. A pilot study was conducted among 20 students at the University of Ibadan, Nigeria, prior to commencement of the study. Each student completed the questionnaire independently and was debriefed in order to get their feedback on the clarity of the instrument. Four graduate students were recruited as research assistants and trained to administer the questionnaire at the two study universities. The University of Fort Hare and Ondo State Ministry of Health Ethical Review Committees approved the study protocol. The participants voluntarily granted their consent after being convinced of confidentiality and anonymity.
Study participants and sampling design
All students currently enrolled at the two universities were eligible for the study. The study sample size was based on an estimated 50% prevalence of HIV testing uptake, as previously reported by Ogaji et al. (Reference Ogaji, Oyeyemi and Ibrahim2013). The sample size was determined using the sample size calculator at a confidence level of 95%, with a confidence interval of ±5. The appropriate sample size needed to detect 5% error at each of the universities after adjusting for possible missing responses was 400 participants. A stratified random sampling technique was employed for sample selection. To ensure representativeness, students were stratified by sex, year of study and faculty of study. Probability proportion to the size of each stratum was used to determine the number of samples required from each stratum. Overall, 800 students took part in the study, but sixteen questionnaires were discarded due to incomplete responses, giving a final sample of 784 (402 males and 381 females).
Measures
Dependent variables
HIV testing uptake was measured using two variables based on the answers to two questions: (1) ‘ever tested for HIV’, with a dichotomized yes/no response to the question ‘Have you ever tested for HIV?’; and (2) ‘recently tested for HIV’, based on the question ‘What month and year was your most recent HIV test done?’ Recent HIV test was defined as an HIV test done within one year of the survey.
Independent variables
The main independent variables were ‘HIV risk perception’, ‘perceived severity of HIV’, ‘knowing partners’ HIV status’ and ‘discussed HIV/STIs with partners’.
‘HIV risk perception’ was based on the HBM assumption that illness perception influences risk reduction behaviour. The study thus hypothesized that HIV risk perception is significantly related to HIV testing – which is a risk reduction behaviour. It was measured using three questions derived from previous studies (Corneli et al., Reference Corneli, Mckenna, Headley, Ahmed, Odhiambo and Skhosana2014; Sisay et al., Reference Sisay, Erku, Medhin and Woldeyohannes2014; Marcus et al., Reference Marcus, Gassowski and Drewes2016): (1) ‘Do you think you are at risk of contracting HIV?’, with a dichotomized yes/no response; (2) How would you rate your chances of contracting HIV?’, on a scale of ‘highly likely’ to ‘highly unlikely’, re-coded into ‘unlikely’ (merging of unlikely and highly unlikely) and ‘likely’ (merging of likely, highly likely and neutral) for the bivariate and multivariate analyses; (3) ‘Do you have concerns about contracting HIV from your sexual partners?’, with the three mutually exclusive responses ‘concerned’, ‘somewhat concerned’ and ‘not concerned’, re-coded into ‘concerned’ and ‘somewhat/unconcerned’ for the bivariate and multivariate analyses. It was assumed that some of the participants currently had more than one sexual partner.
‘Perceived severity of HIV’ was assessed by asking participants whether they considered HIV to be a dangerous disease, with a dichotomized yes/no response. ‘Knowing partners’ HIV status’ was assessed by asking participants whether they knew the HIV sero-status of their partners, with a dichotomized yes/no response. ‘Discussed HIV/STIs with partners’ was assessed by asking participants whether they had discussed HIV/STIs with their sexual partners, with a dichotomized yes/no response.
Other independent variables included in the study were age, sex, ever engaged in sex and being sexually active in the past year.
Data analysis
The obtained data were coded and captured into SPSS version 24 for Windows (Chicago, IL, USA). Data were cleaned to remove possible errors during capturing. This was done by running a frequency count for all variables and re-visiting all errors and discrepancies observed in the results. Three levels of analysis were employed: univariate, bivariate and multivariate. Frequency and mean distributions were run for all variables of interest at the univariate level of analysis. The relationship between each independent variable and the dependent variable was examined using cross-tabulation, chi-square and unadjusted logistic regression models. To examine the net effects of knowing partners’ HIV status, having discussed HIV testing with sexual partners, perceived susceptibility to HIV and perceived severity of HIV on recent and ever uptake of HIV testing, a multivariate logistic regression model was computed at 95% confidence interval. Alpha values less than 0.05 were considered statistically significant. Possible confounders were adjusted for in the model.
Results
The demographic characteristics of the study participants by sex are presented in Table 1. Most participants were Christian, resided off campus and shared their rooms with at least one person. Their median age was 22 years.
Table 1 Demographic characteristics of participants by sex, Nigerian university students, 2018
Table 2 shows the distribution of participants by their perceptions of HIV risk and awareness of partners’ sero-status by sex. The majority of students did not think they could contract HIV (87.5%), but more males (15.2%) than females (9.5%) considered themselves to be at risk of contracting the disease (Table 2). Likewise, most participants (81.2%) rated their chances of contracting HIV as either highly unlikely or unlikely, again with significant differences between the sexes. The vast majority (92.1%) considered HIV a dangerous disease, with significant sex variations. When asked whether they were concerned about contracting HIV from their sexual partners, most (63.9%) responded affirmatively, with no significant sex differences. Nonetheless, the majority of participants (79.1%) reported that they were currently doing something to mitigate their risk of contracting HIV, with correct and consistent condom use (29.6%), faithfulness (25.0%) and abstinence commonly being reported (37.4%). Only a few (8%) reported that they did nothing to protect themselves from contracting HIV. However, most participants (61.2%) were unaware of their partners’ HIV sero-status and only 37.6% had ever discussed HIV with their sexual partners, with no significant sex differences.
Table 2 Distribution of participants by HIV risk perception, knowing partners’ HIV sero-status and discussing HIV testing with sexual partners, by sex
Tables 3 and 4 show the logistic regression analyses of the effects of HIV risk perception, knowing partners’ HIV sero-status and discussing HIV testing with sexual partners on ‘ever tested for HIV’ and ‘recently tested for HIV’, respectively. Only 50.6% and 30.6%, respectively, of participants had ever tested, or recently tested, for HIV, with no significant differences between the sexes (Figure 1). In the case of ‘ever tested for HIV’, the unadjusted logistic regression model showed an independent association with being aged 20 or over, knowing partners’ HIV status, discussing HIV with sexual partners, perceived susceptibility to HIV, rating of chances of HIV susceptibility, concerns about contracting HIV and being sexually active (Table 3). However, in the adjusted model, after controlling for other covariates (age, sex and being sexually active), knowing partners’ HIV status, having discussed HIV with partners, perceived likelihood of contracting HIV and being very concerned about contracting HIV were significantly associated with ‘ever tested for HIV’. Individuals who knew their partners’ HIV sero-status had increased odds of ‘ever tested for HIV’ compared with those who did not. Likewise, individuals who had discussed HIV testing with their sexual partners had increased odds of ‘ever tested for HIV’ compared with those who had not. Those who considered themselves to be likely to be susceptible to contracting HIV were twice as likely to have ever tested for HIV compared with those who considered themselves to be unlikely to be susceptible to contract HIV.
Table 3 Adjusted and unadjusted logistic regression model showing the effects of HIV risk perception, knowing partners’ HIV sero-status and discussing HIV testing with sexual partner on ‘ever tested for HIV’
UOR: unadjusted odd ratio; AOR: adjusted odd ratio; Ref: reference category.
*** p<0.005
Table 4 Adjusted and unadjusted logistic regression model showing the effects of HIV risk perception, knowing partners’ HIV sero-status and discussing HIV testing with sexual partner on ‘recently tested for HIV’
UOR: unadjusted odd ratio; AOR: adjusted odd ratio; Ref: reference category.
*** p<0.005; *p<0.05.
Figure 1 Uptake of HIV testing among Nigerian university students by sex (N=784).
In the case of ‘recently tested for HIV’, the unadjusted logistic regression model showed an independent association with being aged 20 or over, knowing partners’ HIV status, having discussed HIV testing with partner, perceived risk of contracting HIV and being sexually active (Table 4). After controlling for other covariates (age, sex and consistent condom use) in the multivariate analysis, knowing partners’ HIV status, having discussed HIV testing with partners, having concerns about contracting HIV and rating oneself as likely to contract HIV were associated with recent HIV testing. Individuals who knew their partners’ status were twice as likely to have recently tested for HIV compared with those who had not. Likewise, individuals who had discussed HIV testing with their partners had increased odds of recent uptake of HIV testing compared with those who had not. Similarly, students who believed they were likely to contract HIV had higher odds of recent HIV testing compared with those who rated themselves as unlikely to contract HIV. Individuals who had concerns about contracting HIV from their sexual partners had increased odds of recent HIV testing.
Discussion
This study examined the effect of HIV risk perception, discussion of HIV with sexual partners and knowing partners’ HIV sero-status on HIV testing uptake among young adults in two Nigerian universities. It was found that most university students had a low perception of their risk of susceptibility to HIV. This was the case despite the fact that the majority of the students are unaware of their own or their partners’ HIV sero-status and had never discussed HIV/STIs with their sexual partners. It is possible that most students underestimated their risk of contracting HIV, perhaps due to perceived HIV stigma, as reported in a previous study by Corneli et al. (Reference Corneli, Mckenna, Headley, Ahmed, Odhiambo and Skhosana2014). These findings corroborate those of a recent study which showed that many people at risk of STIs underestimate their level of risk (Clifton et al., Reference Clifton, Mercer, Sonnenberg, Tanton, Field and Gravningen2018). However, most students have concerns about contracting HIV from their sexual partners. This clearly indicates that while most students in the study believed they were unlikely to contract HIV, they were of course concerned about contracting HIV. Their concerns about contracting HIV may possibly be due to their exposure to the risk of contracting HIV. In other words, they engage in unprotected sex with a partner despite not knowing their partner’s HIV status. Risky sexual behaviour has been reported to be common among Nigerian university students (Ajayi et al., Reference Ajayi, Nwokocha, Akpan, Adeniyi and Goon2018).
The rate of HIV testing uptake found in this study was very low, and is consistent with that reported in previous similar studies (Babalola, Reference Babalola2007; Golub & Gamarel, Reference Golub and Gamarel2013; Ogaji et al., Reference Ogaji, Oyeyemi and Ibrahim2013). The study’s reported rates of ever and recent testing for HIV fell short of the first of the 90–90–90 UNAIDS targets, which is required for the elimination of new HIV infections. This clearly indicates that many cases of HIV infection are undiagnosed in the study population and this could have deleterious consequences. Although the cost of HIV testing is relatively low in Nigeria, access to this is limited (Federal Ministry of Health, 2013), especially on university campuses. Thus, concerted efforts to promote HIV testing on campuses are long overdue and should be a priority strategy for the National Agency for the Control of AIDS (NACA) in Nigeria. Self-testing for HIV has demonstrated impressive results (MacPherson et al., Reference Macpherson, Lalloo and Webb2014), and thus may be an important strategy for promoting uptake of HIV testing on Nigerian campuses.
This study has shown that high HIV risk perceptions increase the odds of HIV testing uptake after controlling for other covariates. This finding is consistent with the literature (De Paoli et al., Reference De Paoli, Manongi and Klepp2004) and lays credence to the assumptions in the HBM (Glanz et al., Reference Glanz, Rimer and Viswanath2008). Consistent with the proposition in the HBM, individuals who are gravely concerned about contracting HIV from their sexual partners are more likely to have ever tested or recently tested for HIV compared with those who are not concerned or neutral about contracting HIV. This finding extends what is already known about HIV risk perceptions and uptake of HIV testing. Individuals who are concerned about contracting HIV, perhaps resulting from their risk exposure, may decide to test for HIV in order to allay their fears and change their risky behaviour. Similarly, individuals who rate their risk of contracting HIV as highly likely/likely have higher odds of uptake of HIV testing compared with those who rate their chances of contracting HIV as highly unlikely or unlikely. It is possible that those who rate themselves as highly unlikely to contract HIV are abstaining from sex or using condoms consistently, and as such feel protected from contracting HIV. Nonetheless, uptake of HIV testing is still important for everyone, irrespective of his or her perceived risk of susceptibility to contracting HIV.
This study also revealed a strong association between knowing partners’ HIV sero-status and the uptake of HIV testing. Previous studies have reported that spousal communication and knowledge of male partner’s HIV sero-status are determinants of uptake of HIV testing (Gage & Ali, Reference Gage and Ali2005; Turan et al., Reference Turan, Bukusi, Onono, Holzemer, Miller and Cohen2011). The influence of one partner over the other could be leverage in promoting HIV testing uptake. One pathway through which knowing one’s partner’s status could increase uptake of HIV testing is through contact tracing, which is a widely adopted strategy in identifying new HIV infections (Hsieh et al., Reference Hsieh, Wang, De Arazoza and Lounes2010). Evidence that offering HIV testing to family members, especially sexual partners of people diagnosed with HIV, doubled the proportion of men in Kenya who took an HIV test lends credence to this assertion (Krakowiak et al., Reference Krakowiak, Kinuthia, Osoti, Asila, Gone and Mark2016). This finding also largely underscores the importance of the influence of couple testing in promoting uptake of HIV testing. Counselling adolescents and youths who test for HIV about the importance of knowing their partners’ HIV status could be an important strategy to promote uptake of HIV testing among this cohort.
Another important finding of this study was that discussion of HIV/STIs with sexual partners increases the odds of uptake of HIV testing. This relationship holds after controlling for other covariates. This finding underscores the importance of open communication between couples in promoting healthy behaviour. Discussion of HIV/STIs could promote safe sex practices among adolescents and youths otherwise characterized as engaging in risky sexual practices. Health promotional messages designed to influence uptake of HIV testing among this cohort could consider including messages promoting the discussion of HIV/STIs with sexual partners.
The study has its limitations. First, its cross-sectional design means that the association between HIV risk perception, discussion of HIV with sexual partners and knowing partners’ status on HIV testing uptake does not infer causation. Also, the study was conducted among adolescents and young adults with higher education levels than those of the general population, so the findings are not generalizable to the overall Nigerian adolescent and young adult population.
In conclusion, the uptake of HIV testing was found to be low in the study setting and fell short of the first 90% UNAIDS target. Encouraging the discussion of HIV/STIs with sexual partners and knowing partners’ HIV status could be important in promoting the uptake of HIV testing among young adults and adolescents. Age-appropriate strategies targeting open communication on HIV/STIs and disclosure of sero-status between sexual partners are required to promote the uptake of HIV testing among young adults and adolescents in Nigerian youth.
Acknowledgments
The authors would like to acknowledge the University of Fort Hare and Govan Mbeki Research Institute for granting AIA a postdoctoral fellowship position during which he completed the empirical aspects of this study. They also express their gratitude to their research assistants, Ojo Oluwayomi and Kafayat Olanike Ismail, for their immense contributions towards the completion of the data collection.
Ethical Approval
The University of Fort Hare and Ondo State Ministry of Health’s Ethical Review Committees approved the study protocol. All study participants gave written consent to indicate their voluntary participation.
Conflicts of Interest
The authors have no conflicts of interest to declare.
Funding
No funding was received towards completion of this work.
