Introduction

At present, late diagnosis of the human immunodeficiency virus (HIV) is one of the main public health problems worldwide (^{Becky et al., 2015}; ^{Black et al., 2014}; ^{Fuster, Molero, Nouvilas, Arazo, & Dalmau, 2014}). In recent years, the situation has been directly related to psychosocial factors associated with low rates of HIV diagnostic testing (^{Lopera, Martínez, & Ray, 2011}). A prolonged period of time between the moment of infection and the beginning of antiretroviral treatments affects the rise in new cases of transmission, in the progression of the infection, in difficulties for the immune system to recover, in the decrease of the person's life expectancy, and in the increase of economic costs for society (^{Arrivillaga et al., 2012}; ^{Bolsewicza, Vallelyb, Debattistad, Whittakere, & Fitzgeralda, 2015}; ^{Castel et al., 2012}; ^{Dai et al., 2015}; ^{Kall, Smith, & Delpech, 2012}; ^{Schnall, Rojas, & Travers, 2015}).

In Colombia, according to the Ministry of Health and Social Protection (2015), of the 135,000 people diagnosed with HIV/AIDS in 2014, only 40 % took a diagnostic test in a timely manner and received antiretroviral treatment. The same report indicated that 22 % of those affected were young people between the ages of 15 and 24 years old. According to the National Survey of Demographics and Health (Encuesta Nacional de Demografía y Salud [ENDS], 2015, for its Spanish acronym), nearly 75 % of Colombians between 13 and 24 years old have never been tested for HIV. Even though 65 % know where the test can be taken, only 25 % have taken it at some point in their lives, and 8 % in the last 12 months. Altogether, only 20 % of this population group knows their serological status, creating a circumstance that impedes identification of the reach of the epidemic and hinders early pharmacological and non-pharmacological treatments.

Low risk perception, lack of knowledge about the disease and excessive trust in sexual partners are the principal risk factors that have been associated with the contraction of HIV in this population, according to different studies (^{Belza et al., 2014}; ^{Fuster et al., 2014}; ^{Hoyos et al., 2019}; ^{Morales, Espada, & Orgilés, 2016}; ^{Peralta, Deeds, Hipszer, & Ghalib, 2007}). Moreover, the social stigma attached to HIV has been identified with an increased probability that young people with a high-risk perception develop a larger number of avoidance behaviors towards taking the test (^{Decker et al., 2015}; Espada, Escribano, Morales, & Orgilés, 2017; Fuster et al., 2014). To that effect, ^{García, Reyes and Caylà (2012)} found that high-risk perception and offering the test during pregnancy are motivations for women taking the test more frequently than men.

The reasons why young people report a low risk perception are related to an excess in confidence, a lack of knowledge about the test, stereotyped conceptions about the disease, and a low number of sexual relationships without protection. On the other hand, the reason for high-risk perception is associated with risky sexual relationships with people suspected of having HIV (^{Fuster et al., 2014}). Other studies demonstrate that although high-risk perception reduces the behaviors that expose young people to the disease, it is not related to a greater frequency of taking the test (^{Bradley, Tsui, Kidanu, & Gillespie, 2011}; ^{Brooks, Lee, Stover, & Barkley, 2011}). Likewise, it has been found that a positive attitude toward the test does not imply higher testing rates (^{Espada, Escribano, Orgilés, Morales, & Guillén, 2015}; Morales et al., 2016).

Previous research, guided by the health belief model proposed by ^{Janz and Becker (1984)}, has identified that the main barriers for young people taking the test are: (a) at an interpersonal level, the unknowledge about HIV, rejection of a positive diagnosis, stigmatized perceptions of the disease, fear about the disclosure of the results, low risk perception, and lack of familiarization with the test and healthcare services (^{Bilardi et al., 2013}; ^{Deblonde et al., 2010}; ^{Fuster et al., 2014}; ^{Kaai, Bullock, Burchell & Major, 2012}); (b) at an intrapersonal level, the negative perception of patients about the quality of services, resistance of the healthcare providers to talk about HIV and offer the test, and fear of the impact of the diagnosis on their employment situation (^{Becky et al., 2015}; ^{Hampanda, Ybarra & Bull, 2014}; ^{Hoyos et al., 2012}, 2013; ^{Hyden, Allegrante & Cohall, 2014}); and (c) at an extrapersonal level, the sociodemographic characteristics, public policy, difficulties accessing the test in the healthcare system, scarcity of funds, and lack of human resource formation in healthcare professionals (Bilardi et al., 2013; Deblonde et al., 2010; Kaii et al., 2012; ^{Kurth, Lally, Choko, Inwani & Fortenberry, 2015}; ^{Newman et al., 2013}).

In particular, the intrapersonal barriers, like skills and reasons, are connected to the extrapersonal, like social, cultural, economic, and political elements (Beglery, McLaws, Ross & Gold, 2008). Structural limitations hinder access to services and generate a negative impact on young people's reasons to get tested (^{Deblonde et al., 2010}; ^{Newman et al., 2013}). The perceptions of each country's public policy about the disease may perpetuate the stigma associated with HIV, a situation that directly affects attitudes in the community. A fear of interaction with patients, a lack of understanding of the epidemiology of HIV, and a unknowledge of the test's protocols on the part of healthcare professionals may all become barriers (Kaii et al., 2012; Deblonde et al., 2010).

Lacking knowledge about one's serological status makes it impossible to begin timely antiretroviral treatments and gets in the way of young people developing protective behaviors regarding their health (^{Becky et al., 2015}; ^{Cooney, Hiransuthikil, & Lertmaharit, 2015}; ^{Deblonde et al., 2010}; ^{Flowers, Knussen, Li, & McDaid, 2013}; ^{Hoyos et al., 2013}; ^{Kall et al., 2012}; ^{Schnall et al., 2015}). Taking the HIV diagnostic test is related to, according to the health belief model (^{Janz & Becker, 1984}), people's knowledge, attitudes, and self-efficacy, such as their perception of barriers and facilitators associated with this process. In this context, identifying the multiple psychosocial factors associated with the low rates of testing in Colombia may contribute to the design of campaigns geared toward prevention and timely treatment of the disease, and aimed at reducing mortality and morbility rates (^{Fuster et al., 2014}; Hoyos et al., 2013; ^{Newman et al., 2013}). Moreover, this could help to strengthen the sexual and reproductive healthcare services that correspond to this population group.

For this reason, this research had two main objectives: (a) to relate sexual risk, knowledge, and attitudes towards HIV testing in Colombian university students; and (b) to describe the perceived barriers in those who had not been tested, analyzing the differences according to sex.

Method

Results

This section presents the psychosocial factors associated with HIV testing in Colombian university students.

Sociodemographic data

In total, 1057 students participated, 651 (63 %) women and 384 (37 %) men, whose ages were between 18 and 22 years old (M = 19, SD = 1.27), and were in their first (23.1 %), second (29 %), third (28.8 %), fourth (14.6 %), and fifth (4.5 %) year of study at a university in Cali, Colombia. Of the total of participants, 214 (20.2 %) informed having taken the HIV test.

Table 1 presents the sociodemographic data of the participants, differentiated by those who have and have not taken the test.

Table 1 Socio-demographic data, differentiated by those participants who have taken and those who have not taken the HIV diagnostic test 

Note. * p < .05, ** p < .01, *** p < .001. MMS = Minimum Monthly Salary.

No significant differences were found between the participants who reported having taken the HIV diagnostic test in the variables sex (X ² = 0.011, p = .91), socioeconomic status (X ² = 0.002, p = .96), family income (X ² = 0.059, p = .80), and the people with whom they live (X ² = 3.293, p = .07). Nevertheless, significant associations of a low magnitude were identified according to sexual orientation and if the participant was religious, with a greater presence of homosexual or bisexual participants (X ² = 16.482, p < .001, φ = 0.12) and lesser presence of practitioners of any given religion (X ² = 7.311, p < .01, φ = 0.08) in the group that had taken the diagnostic test.

Sexual practices and the HIV Test

In order to relate sexual practices with the test, the two groups were compared by the age of sexual initiation, the types of sexual relationships, the number of sexual partners, and the use of condoms in their first and last sexual intercourse, limiting the sample to 83.2 % of those who affirmed having had some type of sexual intercourse, be it vaginal, anal, or oral-genital. Table 2 shows that the percentage of participants who reported having had vaginal intercourse was significantly greater (X ² = 29.38,p < .001, φ = 0.16) in the participants who took the HIV test (93.7 %) than in those that did not (77 %). However, the magnitude of this association was low.

Table 2 Sexual practices differentiated by participants who have taken and those who have not taken the HIV test 

Note. * p < .05, ** p < .01, *** p < .001.

Moreover, statistically significant differences were identified associated to the age of the first sexual intercourse (t = -3.01, p < .01) and to the number of sexual partners in the last six months (t = 1.99, p = .04), when the age of sexual initiation was slightly less (mean difference = -0.45, d = 0.23) and the number of sexual partners was slightly more (mean difference = 0.27, d = 0.2) in participants who took the HIV test. The effect size between both mean differences was small.

As for the use of condoms, significant associations of low magnitude were found in the first (X ² = 5.01, p = .02, φ = 0.07) and the last (X ² = 3.95, p = .04, φ = 0.06) sexual intercourse, with a greater percentage of use in the first (78.5 % vs. 71.4 %) and the last sexual encounter (51.7 % vs. 43.6 %) in participants who did not take the test.

Knowledge associated with the HIV Test

For the total sample, the main sources of information about the HIV test were healthcare professionals (37.1 %), the Internet (29.5 %), and educational institutions (26.3 %). Meanwhile, TV or radio programs (19.3 %), news articles (19.1 %), and advertising (5.2 %) were the least employed sources.

To measure the knowledge associated with the HIV test, 14 items were applied composed of correct and incorrect affirmations. The participants who took the test scored (range from 0 to 14) higher (M = 9.21, SD = 1.98) compared to those who had not taken it (M = 7.87, SD = 2.19), and these were significant differences (t = 7.79, p < .001, d = 0.65), with a medium effect size. Moreover, it stands out that when comparing by sex, significant differences were not found (t = 1.69, p = .09) in the knowledge reported by men (M = 8.30, SD = 2.46) and women (M = 8.03, SD = 2.06).

With respect to incorrect knowledge, it was found that the participants who had reported taking the test presented higher indices of correct answers in the affirmations "if the results of my partner's test are negative, it is not necessary for me to take the test" (90.1 % vs. 80.1 %) and "if the results of my test are negative, then I don't have anything to worry about" (66 % vs. 51 %), these being significant differences (p < .05). Meanwhile, considering the correct knowledge, the highest indices of correct answers were reported in the affirmations "the results of the HIV test are confidential" (86.9 % vs. 77 %) and "healthcareproviders cannot administer the test without my consent" (84.1 % vs. 76.2 %), evidence, again, significant differences (p < .02).

Tables 3 and 4 present the index of correct knowledge about the HIV test, differentiated by the participants who have taken it and those who have not.

Table 3 Correctly identified incorrect knowledge about the HIV test, differentiated by participants who have taken it, and those who have not 

Note. *p < .05, ** p < .01, *** p < .001.

Table 4 Correctly identified correct knowledge about the HIV test, differentiated by participants who have taken the test and those who have not 

Note. * p < .05, ** p < .01, *** p < .001.

A low index of correctly identified incorrect affirmations is notable in participants from both groups in statements like "I can only know if I have HIV through a blood test" (13.6 %), "Rapid HIV detection tests are not reliable"" (25.9 %), and "Onlypeople of legal adult age can take the test" (57.8 %), likewise in correct affirmations like, "I can only be tested with pre and post-test counseling" (12.6 %), "The HIV test is free" (14.5 %), and "I can request testing in my healthcare provider's clinic" (62.5 %).

Testing barriers

The primary reasons for men and women not to have taken the HIV diagnostic test were: (a) they were sure their sexual partners did not have HIV (29.8 %), (b) they trusted the sexual partners they had had (29.2 %), (c) the test had never been offered to them (25.2 %), and (d) they had never had sexual intercourse without protection (23.8 %). Table 5 presents the barriers the participants associated with the HIV diagnostic test.

Table 5 Barriers associated with the HIV diagnostic test 

Note. * p < .05, ** p < .01, *** p < .001.

Regarding the differences by sex, it was found that women do not take the test in a higher percentage than men because they have not had sexual intercourse (20.7 % vs. 11.3 %) and because they do not want their parents to know that they have had sexual intercourse (3.7 % vs. 2.7 %), being these significant differences (p < .015). Meanwhile, men did not get tested in a higher percentage than women because they were sure their partner did not have the virus (34.8 % vs. 26.9 %), because they had always used a condom during sexual intercourse (28.5 % vs. 21 %), and for fear of getting an HIV positive result (12 % vs. 9.65 %), also showing significant differences (p < .01).

Discussion

The two primary objectives of this research were: (a) to relate sexual risk, knowledge, and attitudes towards HIV testing in Colombian university students; and (b) to describe the perceived barriers in those who had not been tested, analyzing the differences according to sex.

The results identify that even though the majority of participants were already sexually active (83.2 %), nearly half of them did not use a condom in their last sexual intercourse (49.8 %). The literature has connected this situation to factors like lack of access to condoms at the moment of sexual intercourse, low risk perception of contracting HIV, and embarrassment towards demanding its use from the partner during the encounters (^{Belza et al., 2014}; ^{Fuster et al., 2014}; ^{Hoyos et al., 2019}; ^{Morales et al., 2016}). Despite high participant exposure to sexual risk and that the majority knew that the HIV diagnostic test existed, only 20 % had taken it in order to know their serological status, a percentage that approximates to 22 % that were reported in the study by Hoyos et al. (2012) in Spain, to 29.5 % in the research by ^{Peralta et al. (2007)} in the United States, and to 25 % from the ENDS report (2015) in Colombia.

However, even though the scores on knowledge about the test obtained by the participants who had taken it were superior to those of the participants who had not, and the majority of the information was acquired from sources of information like healthcare professionals, Internet, and educational institutions, it was observed that a significant number do not know that in Colombia, by law, the test is free, that they can get it taken whenever required at the Empresa Prestadora de Servicios de Salud (EPS) [Healthcare Providing Service] as part of the Plan Obligatorio de Salud (POS) [Obligatory Health Plan], that it is freely accessible to people of all ages, and that in order to be tested they must receive pre and post-test counseling through which they will be given, respectively, basic information about the function of the test, and help for them to understand the result (Profamilia, 2017).

Likewise, some of the study's participants did not know that healthcare professionals cannot reveal information about the diagnosis, results, and evolution of the disease (World Health Organization [WHO], 2012). A large proportion was unaware of the existence of tests that were different from those that use blood screening (ELISA and rapid detection tests) and confirmative tests (Western Blot), like saliva and urine testing as a method to diagnose the disease (Ministerio de Salud y Protección Social [UNFPA], 2011). Similarly, they did not know that it is recommendable to be tested every year, and that even when the results are negative, they should continue tracking their serological status (Infosida, 2016).

Another factor reported in previous research is the relationship between taking the test and the participants' attitude toward HIV (^{Espada et al., 2015}; ^{Morales et al., 2016}). Different from the results of previous studies (^{Belza et al., 2014}; ^{Bilardi et al., 2013}; ^{Bradley et al., 2011}; ^{Brooks et al., 2011}; ^{Decker et al., 2015}; ^{Deblonde et al., 2010}; Espada et al., 2017; ^{Fuster et al., 2014}; ^{Kaai et al., 2012}), this research found that the exposure to risky sexual behavior and knowledge about the HIV test were related to a greater tendency to taking the test, with a medium effect on knowledge and a small effect on sexual risk. Nevertheless, in line with the discoveries from the cited studies, a relation between favorable attitudes towards the test and taking it was not identified.

Similar to what has been reported in other studies, these results demonstrate that the main personal barriers associated with low rates of HIV testing are excess trust in sexual partners, low perception of risk of exposure to the disease, and the certainty of having had sexual intercourse with protection (^{Belza et al., 2014}; ^{Fuster et al., 2014}; ^{Morales et al., 2016}; ^{Peralta et al., 2007}). At this point, it is noteworthy that trust in sexual partners, certainty of having always used a condom during intercourse, and the fear of receiving positive test results are reasons that are significantly more present in men than in women. Conversely, not wanting their parents to know they have had sexual intercourse was a reason more evident in the women of this research.

Other related factors were the lack of test offer by healthcare institutions and the fact that the participants still had not become sexually active (^{Morales et al., 2016}; ^{Peralta et al., 2007}). In this research, fear of getting a positive diagnosis, stigmatized perceptions of the disease, and fear of disclosure of the results were not personal barriers that interfered with taking the test (^{Bilardi et al., 2013}; ^{Deblonde et al., 2010}; ^{Kaai et al., 2012}).

Currently, the intervention strategies have centered on identifying the perceptions of risk concerning HIV, recognizing the strategies to better access to the tests, increase the number of specialized healthcare professionals, and reduce the stigma associated with the disease (^{Deblonde et al., 2010}; ^{Kaai et al., 2012}; ^{Kall et al., 2012}; ^{Newman et al., 2013}). Authors like ^{Becky et al. (2015)} propose facilitators that increase the probability that people will take the test, which include: (a)politics that are associated with the elimination of the written consent requirement, rationalization of pretest and post-test counseling, and obligation of doctors to perform routine HIV testing; (b) community, referring to decreasing the stigma and developing campaigns geared toward the acceptance of patients; (c)practices which imply counseling about HIV, educating healthcare professionals, and including the test as a routine practice in clinical work and the use of oral HIV tests; and (d) personal, which consist of educating people about the risk of contracting HIV, increasing acceptance of the advantages of being tested, timely attending the requests of patients, and addressing sexuality in an comprehensive manner.

Faced with this panorama, it becomes a priority to strengthen the conceptual training and the empowerment of individuals, families, communities, and healthcare institutions regarding beliefs and knowledge associated with sexuality and sexual and reproductive rights, from a perspective of self-care, where scientific information is the most important factor, establishing a differential focus adjusted to people's life course and contributing to the reduction of stigma and discrimination. Moreover, strengthening educational sexual health and reproduction programs from early childhood will contribute to the well- being and quality of life of these populations and decrease the cases of teenage pregnancy, HIV, and other STDs. Notwithstanding, these interventions should be coordinated within the public policy frameworks of each country (^{Hoyos et al., 2019}; ^{Morales et al., 2016}; ^{Sevilla, Sanabria, Orcasita, & Palma, 2016}), and should favor inclusion, comprehensive care, and easy access to sexual and reproductive health programs.

Therefore, the findings of this study are presented as a tool for the design of programs to promote sexual and reproductive health, as well as the prevention of disease and teenage pregnancy, specifically related to HIV, condom use, and HIV testing. The primary limitation of this study was that the non-randomized sample employed does not allow for a generalization of the results in this population. The validation of the instruments Barriers to HIV Testing (Barreras hacia la Prueba del VIH;^{Peralta et al., 2007}) and the Scale of Attitudes Toward HIV/AIDS (Escala de Actitudes hacia el VIH-Sida;^{Espada et al., 2013}) in Colombian populations presents an opportunity for researchers interested in this topic to further study it. Future work should continue to explore, through the design of different methodologies, the psychosocial factors associated with the low rate of HIV diagnostic testing.