The Effects of HIV-Related Stigma on HIV Counselling and Testing in Nigeria: A Mediation Analysis

Abstract

Human immunodeficiency viruses (HIV) counselling and testing (HCT) plays a major role in the continuum of HIV programmes in Nigeria. However, HIV-related stigma (STGM) poses a serious threat to its success. Consequently, a cross-sectional study was carried out to examine whether STGM mediates the relationships between the explanatory variables (HIV-related knowledge, HIV transmission misconception (MSHIV), and perceived seriousness of HIV/acquired immune deficiency syndrome (AIDS)), and the outcome variable (HCT service utilization). The mediation analysis was undertaken using data from 768 individuals collected through convenience sampling in the Local Government Areas of Alimosho, Ikorodu, and Surulere of Lagos state, and deploying partial least squares–structural equation modelling. The results show that STGM played a mediating role in the relationship between MSHIV, perceived seriousness of HIV/AIDS, and HCT utilization. These findings offer wide ramifications for the intensification and enforcement of Nigeria’s HIV/AIDS Anti-Discrimination Act 2014 to eradicate stigma, which is important to enhance uptake of HCT to achieve the United Nations’ 90-90-90 HIV targets by 2020.

Keywords

Human immunodeficiency viruses (HIV)-related stigma HIV counselling and testing HIV/Acquired Immune Deficiency partial least squares–structural equation modelling Lagos state Nigeria

Introduction

With a population of about 182 million (National Population Commission, 2017), Nigeria ranked third among countries burdened with human immunodeficiency viruses (HIV)/acquired immune deficiency syndrome (AIDS) in the world (United Nations Programme on HIV and AIDS, 2018). According to the National Agency for the Control of AIDS, (2019), the latest national HIV prevalence rate was 1.4% for the age group 15–49 years old, while 1.9 million people are living with HIV. HIV counselling and testing (HCT) is among the several programmes established to reduce the spread of HIV in Nigeria. HCT plays a critical role in the continuum of HIV programmes, which range from prevention to care (Bwambale et al., 2008; Ostermann et al., 2015; United Nations Programme on HIV and AIDS, 2015).

Studies have shown that HCT contributes to reducing risky sexual behaviour which may ultimately lead to a reduction in the probability of HIV transmission (Denison et al., 2008; Fonner et al., 2012). HCT affords individuals the opportunity to know their serostatus and spur their motivation for self-care (Okawa et al., 2017). Also, it plays a significant role in attaining the much touted United Nations’ 90-90-90 HIV targets by 2020 and Sustainable Development Goals (SDGs) by 2030 (United Nations Programme on HIV and AIDS, 2014). However, the utilization of this essential HIV service in Nigeria, like other sub-Saharan Africa (SSA) countries, is still low, which is a major policy concern. According to Staveteig et al. (2013) about 50% of people living with HIV/AIDS (PLWHA) in SSA do not know their HIV status. According to the Nigeria Demographic and Health Survey report, only 21.5% adult men and 29.5% adult women were tested for HIV in Nigeria in 2013 ( Nigeria Demographic and Health Survey, 2014).

One of the major problems confronting HCT utilization is the stigma associated with HIV infection (National Agency for the Control of AIDS, 2015). HIV-related stigma (STGM) is among the non-financial factors that pose significant barriers to the successful implementation of HCT in high HIV burdened SSA countries. Stigma is regarded as a negative judgement cast by the society which is often generated by fear, ignorance and false belief (Word Health Organization, 2015). Fear of knowing one’s HIV status owing to the stigmatization of HIV-positive individuals has been documented in the literature as the primary barrier to HCT utilization among youths and students (Buldeo and Gilbert, 2015; Ndabarora and Mchunu, 2014; Nothling and Kagee, 2013); pregnant women attending antenatal health care facilities (Sarker et al., 2007; Workagegn et al., 2015); PLWHA (Earnshaw et al., 2014, 2016), and the population in general (Bova et al., 2016; Bwambale et al., 2008; De Jesus et al., 2015; Ma et al., 2007). Although Studies that demonstrate a bivariate relationship between HIV serostatus with knowledge about HIV and stigma have been reported (e.g. Yang et al., 2006), little is known about the mediation effects of STGM on the relationship between HIV-related knowledge (KNHA), HIV transmission misconception (MSHIV), perceived seriousness of HIV/AIDS (PHSEV) and HCT service utilization (HCTU). Therefore, the purpose of this study is to examine the mediating role of STGM on HCT utilization.

In his contribution to stigma theory, Goffman (1963) defined stigma as “an attribute that is deeply discrediting”. STGM is specifically defined, according to Steward et al. (2008), as “socially shared knowledge about the devalued status of people living with HIV”. Also, Nyblade and MacQuarrie (2006) describe HIV stigma as the “beliefs and attitudes that deeply discredit a person or group because of an association with HIV”. STGM often attracts discrimination, prejudice, and discrediting toward PLWHA and individuals (such as friends, family members, and caregivers), groups, associations, and communities that are associated with them (Bos et al., 2013; Steward et al., 2008). STGM also manifests in different settings, such as educational bodies, workplaces, health care centres, and religious institutions (Feyissa et al., 2012; Parker and Aggleton, 2003; Sekoni et al., 2012; Ulasi et al., 2009).

This STGM has been recognized as an essential driver of epidemics, which poses a serious threat to the effectiveness of HIV/AIDS prevention, treatment, and care programmes (Herek, 1999; Niang et al., 2003; Parker and Aggleton, 2003; Piot, 2006). For example, Yang et al. (2004) reported that STGM was associated with a decrease in individual’s willingness to participate in HIV and sexually transmitted disease prevention intervention activities. Derlega et al. (2002) found that stigma prevented people living with HIV from disclosing their serostatus, and discouraged them from assessing available HCT and treatment services. The low level of HIV/AIDS awareness is associated with discriminating and stigmatizing attitudes towards PLWHA (Babalola, 2007; Vorasane et al., 2017). Therefore, we hypothesized stigma as a mediator in the relationship between KNHA, HIV/AIDS transmission misconception, PHSEV, and HCTU. The analytical framework and hypotheses for the mediating effect of STGM is presented in Figure 1.

Figure 1.

Conceptual/analytical framework of human immunodeficiency viruses (HIV)-related stigma mediator.

The following three hypotheses were tested in the mediation model:

Hypothesis 1: STGM significantly mediates the relationship between KNHA and HCTU.

Hypothesis 2: STGM significantly mediates the relationship between MSHIV and HCTU.

Hypothesis 3: STGM significantly mediates the relationship between PHSEV and HCTU.

Materials and methods

Study location and participants

The study was conducted in Lagos state, Nigeria. The three Local Government Areas (LGAs)—Alimosho, Ikorodu, and Surulere covered in this study had an estimated 2.3 million people according to Nigeria’s 2006 population census. Lagos state has a metropolitan population of about 17 million. It is among the states with the highest HIV/AIDS prevalence rate (1.4%) in Nigeria (National Agency for the Control of AIDS, 2015). Lagos state is in the southwestern part of Nigeria and has 20 LGAs and 38 Local Council Development Areas. The state is politically divided into three districts: North, Central and South Senatorial Districts. The participants were adult men and women aged 20 years and above chosen from the households in the selected areas.

Research design, sampling procedure, and sample size

The study deployed a cross-sectional survey using a nonprobability sampling method from the target population. A nonprobability sampling method was adopted because of the sensitivity of the topic involved, as well as a lack of data on both users and non-users of HCT in the study area. To improve the representativeness of the respondents for a reasonable interpretation and generalization of the results, this study combined both quota and convenience sampling methods in which one LGA was selected from each of the senatorial districts of Lagos state (Lagos East senatorial district – Ikorodu LGA, Lagos West senatorial district – Alimosho LGA, and Lagos Central senatorial district – Surulere LGA) based on their HIV prevalence rate. Trained research enumerators went to the selected locations to recruit eligible individuals from the households for the survey. The enumerators obtained informed consent from each participant before pilot testing and revising self-administered questionnaires. The completion time for the questionnaire was approximately 40–60 minutes. A sample of 768 adults were recruited from the three LGAs: Alimosho; Ikorodu; and Surulere.

Data collection, instrument and measurements

Data for this study were collected using the pre-tested structured questionnaire administered face to face by twelve enumerators from the three LGAs of Lagos state lasting from May 2015 to July 2015. The questionnaires were administered to the individuals who were available in the households at the point of data collection. The questionnaires were presented in English with close-ended questions. The survey instrument was structured into sections: sociodemographic information; knowledge about HIV/AIDS and HCT; STGM questions; and MSHIV. The questions mainly covered the three latent predictor constructs: KNHA; MSHIV; and PHSEV; and the mediator variable, STGM. The latent constructs were measured by multi-items, which were sourced from previous studies with some modifications to suit this study.

STGM

According to Baron and Kenny (1986), a “variable may be said to function as a mediator to the extent that it accounts for the relation between the predictor and the criterion”. The mediator, STGM latent construct was measured using 9-items, which were adapted from past studies (Berger et al., 2001; Genberg et al., 2008; Nigeria Demographic and Health Survey, 2014; Nyblade and MacQuarrie, 2006; Nyblade et al., 2005; Van Brakel, 2006). The items include: “Can you play with a person who has HIV?”; “Can you eat/share food with a person who has HIV”; and “Can you buy things from a person who has HIV?”. These measurement items or indicators were measured on a 3-point categorical nominal scale in which 1 stands for “Agree”, 2 stands for “Disagree”, and 3 stands for “Don’t Know/Not Sure/Depends”.

KNHA

Three items were used to measure this construct: “Have you heard about HIV and AIDS”; “Do you know someone infected with HIV virus”; and “Do you know someone who died as a result of AIDS”. These indicators were measured on a 2-point nominal scale: yes and no (Nigeria Demographic and Health Survey, 2014; Yang et al., 2007).

MSHIV

This latent predictor was operationalized using seven items, such as, “Can HIV can be transmitted through mosquito bites?”; “Can HIV be transmitted by sharing foods or drinks with an HIV infected person?”; and “Can HIV be transmitted by shaking hands or hugging an HIV infected person?” These items were measured on a 2-point nominal scale: yes and no.

PHSEV

A 9-item questionnaire assessed the perceived severity of HIV/AIDS. The measurement items include, “AIDS is a hopeless disease”, “AIDS would be a serious problem for me”, “I believe contracting HIV/AIDS would interfere with my role in the family” (Janz and Champion, 2002; Rosenstock, 2005). Responses were measured on a 5-point Likert-type scale, ranging from 1 (strongly disagree) to 5 (strongly agree).

HCTU

The criterion variable, HCTU, was operationalized by knowledge about HCT and HCT testing history. Two main items were used to measure the latent construct, HCTU: “Do you know where to get an HIV test?”; and “Have you been tested for HIV?”

Analytic techniques

We deployed two statistical software packages – SPSS v22.0 and WarpPLS 5.0 – to analyse the primary data collected. SPSS v22 was used to examine the descriptive statistics, while WarpPLS version 5.0 was deployed in the structural equation modelling (SEM) analysis of mediating effects of STGM. SEM is preferred over regression analysis when two conditions are clear. First, SEM tends to produce less biased results when using cross-sectional datasets. Secondly, it is preferred when there is more than one measurement item for each of the constructs (Frazier et al., 2004; Gunzler et al., 2013; Holmbeck, 2002; Jaccard and Wan, 1996; Peyrot, 1996). In light of the above, we deployed partial least squares–structural equation modelling (PLS-SEM) to analyse the mediating effect of STGM. WarpPLS is a robust PLS-based SEM software developed by Ned Kock. PLS is a non-parametric variance-based SEM approach suitable for SEM analysis where the sample is not too large and data are not normally distributed (Kock, 2015).

Mediation analysis

In the analysis of the mediation effects of STGM, we deployed the alternative approach to the analysis of mediating effect as suggested by Preacher and Hayes (2008) and (Kock, 2014). We first examined the significance of the direct effect of the exogeneous variables on the criterion variable and subsequently assessed the significance of the indirect effect from exogenous variables via the mediator to criterion variable. Both the simple models – which show the bivariate relationship between the constructs and the complex model that includes the mediator – were simultaneously estimated using WarpPLS (Kock, 2015). The PLS-SEM mediation analysis followed the Anderson and Gerbing (1988) recommendation of a two-step approach, which takes care of both measurement and structural model analyses. We used the Stable-3 resampling method in WarpPLS for the analysis (Kock, 2017).

Stage 1: Assessment of measurement model

The measurement models were first analysed to establish the validity and reliability of the measured items of the three predictors and mediator constructs in the model.

Reliability and validity of measurement items

We first established the reliability of the measurement items of each construct by examining their individual loadings to confirm whether they satisfied the recommended minimum loading of 0.5 (Bagozzi and Yi, 1988). In addition, we assessed the reliability of each construct in the model by using the widely used construct reliability measure in survey research, the Cronbach’s alpha (recommended minimum loadings = 0.70) and composite reliability (recommended minimum loadings = 0.60) (Chin, 1998; Fornell and Larcker, 1981; Wetzels et al., 2009). The two validity tests recommended in PLS-SEM analysis are convergent and discriminant tests (Chin, 2010; Hair et al., 2012). The criteria we considered for the establishment of convergent validity were item factor loadings on parent factor (which must be above 0.50 and significant at p < 0.05) and average variance extracted (AVE) (above 0.50). The discriminant validity was examined by comparing the correlations among the latent constructs in the model with the square roots of AVE for each construct (Chin, 2010). According to Fornell and Larcker (1981), to establish discriminant validity all values in the diagonal must be higher than other elements that are off-diagonal. Items that failed to meet the minimum criteria were removed from the model.

Multicollinearity test

We checked for the presence of multicollinearity in our measurement models by assessing the variance inflation factor (VIF). To establish absence of multicollinearity in the SEM model, Kock (2015) recommended a VIF of 3.3.

Stage 2: Assessing structural model of STGM mediator

Having satisfied the reliability and validity of the measurement model in stage 1 of the analysis, we estimated the structural model of the mediating relationship of STGM on HCT utilization. Before we examined the mediating effect in the path analysis, we first considered the coefficient of determination (R²) and effect sizes. To establish the mediation effect of a mediating variable in the model, the extant literature suggests three outcomes of mediating effect: full or complete; partial; and no mediation effect (Baron and Kenny, 1986; Hayes and Preacher, 2014; James and Brett, 1984; Kock, 2014). We deployed the mediation analytical approach of Kock (2014), which requires testing for the significance of both the direct and indirect effects to establish the presence of mediation in the model. Full mediation is achieved when the path running from independent variables to the dependent variable (direct effect) is not significant while the path through the mediator variable (indirect effect) is significant. If both the direct and indirect effects are significant then there is partial mediation. Lastly, if the path running from the mediator to dependent variable is not significant, there is no mediation effect.

Ethical considerations

This study protocol was approved by the Health Research and Ethics Committee of Lagos State University Teaching Hospital, (LASUTH) Ikeja, Lagos State, Nigeria (Approval No: LREC/10/06/497).

Results

Respondents’ sociodemographic characteristics

Table 1 presents the summary of participants’ sociodemographic characteristics. Of the 768 participants, 54% (413) were males, 57% were married, 88% were below 50 years of age, 98% have heard about HIV/AIDS, and only 18% know someone living with HIV/AIDS.

Table 1.

Sociodemographic characteristics of the participants.

Variable		Percentage
Gender
Male	413	54
Female	355	46
Age
20–29	270	35
30–39	288	38
40–49	117	15
50–59	55	7
60–69	28	4
70+	10	1
Marital status
Single	305	40
Married	435	57
Divorced	9	1
Separated	7	1
Widowed	12	1
Educational qualification
Primary	20	3
Secondary	133	17
Certificate	93	12
Diploma	154	20
Bachelor	254	33
Master	88	12
PhD	11	1
No formal education	15	2
Ethnic origin/tribe
Yoruba	476	62
Igbo	197	26
Hausa/Fulani	22	2
Other	73	10
Heard about human immunodificiecy viruses (HIV)/acquired immune deficiency syndrome (AIDS)
Yes	753	98
No	15	2
Heard about HIV counselling and testing
Yes	747	97
No	21	3
Know someone living with HIV/AIDS
Yes	138	18
No	530	82
Know someone who died of AIDS
Yes	221	29
No	447	71

Source: authors’ survey.

Mediation analysis (PLS-SEM)

Assessment of the measurement model

The PLS-SEM analysis results show that the measurement indicators passed the items reliability test with loadings more than the recommended value (Table 2). Therefore, all the constructs measurement items show a satisfactory level of reliability. Composite reliability for all predictor constructs and mediator were higher than the recommended 0.70 (Hair et al., 2011; Urbach and Ahlemann, 2010). Also, the Cronbach’s alpha statistics for the constructs were satisfied (>0.60) (Lyberg et al., 1997), which indicates good construct reliability.

Table 2.

Reliability and convergent validity of the measurement model.

Latent construct	Construct items	Items loadings	Standard error	p-value	Composite reliability	Cronbach’s alpha	Average variance extracted	Variance inflation factor
Human immunodeficiency viruses (HIV)-related stigma (STGM)	STGM1	0.801	0.036	<0.001	0.941	0.929	0.639	1.059
	STGM2	0.779	0.022	<0.001
	STGM3	0.797	0.033	<0.001
	STGM4	0.804	0.033	<0.001
	STGM5	0.799	0.020	<0.001
	STGM6	0.850	0.035	<0.001
	STGM7	0.842	0.046	<0.001
	STGM8	0.765	0.063	<0.001
	STGM9	0.754	0.049	<0.001
Knowledge of HIV/acquired immunodeficiency syndrome (AIDS) (KNHA)	KNHA1	0.839	0.025	<0.001	0.826	0.578	0.703	1.001
	KNHA2	0.839	0.046	<0.001	0.826	0.578	0.703	1.001
The misconception of HIV/AIDS transmission (MSHIV)	MSHIV1	0.656	0.056	<0.001	0.822	0.708	0.540	1.048
	MSHIV2	0.822	0.089	<0.001
	MSHIV3	0.831	0.099	<0.001
	MSHIV4	0.603	0.090	<0.001
Perceived seriousness of HIV/AIDS (PHSEV)	PHSEV1	0.637	0.032	<0.001	0.872	0.826	0.500	1.021
	PHSEV2	0.680	0.030	<0.001
	PHSEV3	0.660	0.031	<0.001
	PHSEV4	0.666	0.029	<0.001
	PHSEV5	0.635	0.033	<0.001
	PHSEV6	0.898	0.026	<0.001
	PHSEV7	0.719	0.038	<0.001
HIV counselling and testing service utilization (HCTU)	HCTU1	0.750	0.178	<0.001	0.720	0.522	0.562
HIV counselling and testing service utilization (HCTU)	HCTU2	0.750	0.279	0.004	0.720	0.522	0.562

Source: authors’ survey.

The results of the measurement model as presented in Table 2 reveal that all the measurement item loadings were more than the recommended minimum value of 0.50 and significantly loaded on the parent factor p < 0.05 (Gefen et al., 2000; Kock, 2015). AVE of all the constructs were well above the accepted threshold of 0.50 (Hair et al., 2013). Thus, all the latent predictor and mediator constructs satisfied the recommended minimum value and were considered valid. The discriminant validity results as shown in Table 3 reveal that all the values in the diagonal are larger than correlation values in both column and row. Therefore, the discriminant validity criterion was satisfied. In addition, as revealed in Table 2, the VIFs were lower than the Kock (2015) recommendation of 3.3. Thus, there was no multicollinearity among the predictor variables in the model.

Table 3.

Discriminant validity of the measurement model.

	STGM	KNHA	MSHIV	PHSEV	HCTU
STGM	(0.800)	−0.047	−0.246	−0.124	0.106
KNHA	−0.047	(0.839)	0.018	−0.085	0.119
MSHIV	−0.246	0.018	(0.735)	0.087	−0.092
PHSEV	−0.124	−0.085	0.087	(0.704)	−0.270
HCTU	0.106	0.119	−0.092	−0.270	(0.750)

Note: The boldface values are the square roots of variance shared between the constructs and their measures (AVE), off-diagonal elements are the correlations among constructs. STGM, human immunodeficiency viruses (HIV)-related stigma; KNHA, HIV-related knowledge; MSHIV, HIV transmission misconception; PHSEV, perceived seriousness of HIV/acquired immune deficiency syndrome; and HCTU, HIV counselling and testing service utilization.

Source: author’s survey.

Assessment of structural model

The results of the structural model reveal that the R² was 0.13 as shown in Figure 2, which signifies that all the four exogenous variables accounted for 13% variation in HCT. The R² is low. The effect sizes range from 0.001 to 0.062 as shown in Table 3, which shows small effect sizes (Wong, 2016).

Figure 2.

The estimated mediating model of human immunodeficiency viruses (HIV)-related stigma.

The mediation analysis and results

In this subsection, we present the mediation effect of STGM on the relationship between KNHA, MSHIV, PHSEV and HCTU as specified in the three hypotheses from the model.

Hypothesis 1: The results as revealed in Table 4, show that the direct path from KNHA to HCTU is significant (β = 0.119; standard error (SE) = 0.039; p = 0.001), while the indirect path from KNHA to HCTU through STGM was not significant (β = −0.005; SE = 0.005; p = 0.181). Thus, we can conclude that STGM did not mediate the relationship between KNHA and HCTU.

Hypothesis 2: The direct path from MSHIV to HCTU was not significant, (β = −0.121; SE = 0.084; p = 0.074). However, the indirect path was significant (β = −0.018; SE = 0.011; p = 0.054). Thus, there is evidence of full mediation effect. In other words, STGM served as a mediator in the relationship between the misconception of HIV transmission and HCTU.

Hypothesis 3: The direct path from PHSEV to HCTU was significant (β = −0.285; SE = 0.109; p = 0.005), while the indirect effect was also significant (β = −0.011; SE = 0.007; p = 0.059). The results indicate that stigma partially mediates the relationship between PHSEV and HCTU. The effect sizes of the mediation relationship fell within a small category (Chin, 1998).

Table 4.

Summary of direct, indirect, and total effect of the structural mediating model.

	Paths	Path coefficient	Standard error	p-value	Effect sizes	Decision
Direct effects
A	KNHA→HCTU	0.119	0.039	0.001	0.015	Significant
B	MSHIV→HCTU	−0.121	0.084	0.074	0.019	Not significant
C	PHSEV→HCTU	−0.285	0.109	0.005	0.088	Significant
Indirect effects
a₁	KNHA→STGM→HCTU	−0.005	0.005	0.181	0.001	Not significant
b₁	MSHIV→STGM→HCTU	−0.018	0.011	0.054	0.003	Significant
c₁	PHSEV→STGM→HCTU	−0.011	0.007	0.059	0.003	Significant
Total effects
	a + a₁	0.114	0.040	0.002	0.001	Significant
	b + b₁	−0.140	0.084	0.049	0.003	Significant
	c + c₁	−0.296	0.108	0.003	0.003	Significant

Note: KNHA, human immunodeficiency viruses (HIV)-related knowledge; HCTU, HIV counselling and testing service utilization; MSHIV, HIV transmission misconception; PHSEV, perceived seriousness of HIV/acquired immune deficiency syndrome; and STGM, HIV-related stigma.

Source: authors’ survey.

Discussion

This research examined the mediating effect of STGM on the relationship between individuals’ misconception about HIV transmission, KNHA, PHSEV, and HCT. The stigma associated with being HIV-positive served as a barrier to HCTU in Nigeria (Nyblade and MacQuarrie, 2006). Findings from this study reveal that two of the three hypotheses formulated were satisfied.

However, the results did not confirm our first hypothesis that STGM would significantly mediate the relationship between KNHA and HCTU, which contrasts with the findings of Yang et al. (2006). Moreover, the results show that KNHA has significant direct effects on HCTU and the relationship between them is positive, which suggests that an increase in awareness about HIV will increase the uptake of HCT. This finding concurs with AIDS-risk reduction behaviour with three components (Fisher and Fisher, 1992): information; motivation; and behavioural skills. Information about HIV can guide individuals to participate in activities that will reduce his or her risk of contracting HIV. Also, it supports Fisher et al. (1996), who found that information about HIV/AIDS was germane to AIDS risk reduction behaviour among college students in the United States.

Subsequently, we hypothesized that STGM would significantly mediate the relationship between MSHIV and HCTU. Our hypothesis was confirmed as the relationship between individuals’ misconception of HIV transmission and HCTU was mediated by STGM. Stigma associated with HIV/AIDS is significant in determining how individuals willing to do HIV tests will respond to HCTU. Information about HIV/AIDS transmission and specific methods of prevention are necessary requirements of action aimed at reducing HIV infections (Fisher and Fisher, 2000). According to Adeokun et al. (2006) and Qian et al. (2007), misconception about the HIV transmission mechanisms could impact on STGM and attract discriminating attitudes. An increase in stigmatization and discriminating attitudes towards individuals infected with HIV or living with AIDS could hinder HCTU. These findings agree with a qualitative study by Meiberg et al. (2008), in which AIDS stigma was found to be a solid barrier towards HIV voluntary counselling and testing among university students in South Africa. Low willingness to participate in the intervention programmes designed to protect against HIV infection and sexually transmitted diseases among rural–urban migrants in China was reported by Yang et al. (2004), which suggests that STGM plays a negative role in HCTU.

Finally, we hypothesized that STGM mediates the relationship between PHSEV and HCTU. Our hypothesis was confirmed as stigma associated with HIV/AIDS mediated the relationship between perception of individuals concerning the potential threats of HIV and HCTU, which shows that individuals may perceive HIV/AIDS to be a serious problem but the presence of stigma may hinder participation in HIV testing. In other words, people may be aware of the threat HIV/AIDS poses, and yet may not participate in HIV testing because of the fear of being stigmatized by society. Individuals’ perceptions about the seriousness of HIV and AIDS could lead to stigmatization of PLWHA, which in turn could adversely affect HCTU. The results concur with the findings of Meiberg et al. (2008) and Odimegwu et al. (2013) where fear and stigma associated with HIV were found to be strong barriers to HIV voluntary counselling and testing in Nigeria.

Also, Bova et al. (2016) found that the low testing rates and poor attitudes toward HIV testing were due to stigma attached to AIDS among African born men in the United States. Denial, fear of disclosure, fear of knowing one’s HIV status, social exclusion and isolation, negative reactions from partners and other people, were all tied up with HIV stigma, which invariably dissuade people from participating in HCT activities (De Jesus et al., 2015; Lindkvist et al., 2015; Ma et al., 2007; Sarker et al., 2007). Our result also partially concur with the findings of Yang et al. (2006), which showed that STGM mediated the relationship between the HIV knowledge and willingness to disclose HIV serostatus among rural–urban migrants in China. If stigmatization and negative attitudes rampantly manifest in ostracization, abuses, rejection, mistreatment, and denials, people will be deterred from seeking HIV testing. Therefore, the focus should not only be on promoting the benefits of having HIV tests, but also on eliminating HIV stigma in all its forms from society, which should enhance HCT uptake in Nigeria. Furthermore, consistent with the findings of Chin et al. (2003), the small size effect in the mediation models does not signify that the impact of stigma is inconsequential.

Conclusion

This study examined the effect of STGM on HCT. The findings show that stigma impacted negatively on the relationship between PHSEV, MSHIV, and HCT. The significant findings offer implications for HCT uptake in Nigeria specifically and the world generally. For HCT to contribute effectively towards the achievement of the first 90 of the United Nations’ HIV 90-90-90 targets by 2020 and SDGs by 2030 in Nigeria, the problem of stigmatization of PLWHA must be addressed. The Nigerian government must strive to achieve zero stigmatization and discrimination towards PLWHA in society. We recommend that the government and all stakeholders collaborate towards ensuring that STGM is eradicated or substantially reduced to improve HCTU, which inter alia, can be achieved by continuously educating people to have an attitudinal change towards PLWHA. Vigorous publicity and enforcement of the Nigeria HIV/AIDS Anti-Discrimination Act 2014 should be carried out to discourage people from stigmatizing individuals with HIV.

Is spite of its significance, the study has some limitations. Firstly, the study was conducted in an urban setting in Lagos state, which invariably has characteristics different from rural areas, and as such, the results obtained might not perfectly fit into other contexts. Secondly, the study deployed a cross-sectional survey, which does not allow the establishment of causality. Lastly, the HIV-stigma data obtained were based on the perception of the general population rather than the individual living with HIV and facing stigmatization and discrimination in society. Therefore, there is a possibility of biases, which might have led to overestimating or underestimating of the level of STGM.

Footnotes

Author’s note

Felix Oluyemi Adekunjo is also affiliated with the Department of Economics, Faculty of Social Sciences, Lagos State University, Lagos, Nigeria.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Felix Oluyemi Adekunjo

Author biographies

Felix Oluyemi Adekunjo holds doctorate degree in Health Economics from University of Malaya, Kuala Lumpur Malaysia. He currently teaches in the Department of Economics, Faculty of Social Sciences, Lagos State University, Lagos, Nigeria. His research interests focus on health economics, development economics, environmental economics.

Rajah Rasiah is a distinguished professor of economics at the Asia Europe Institute, Universiti Malaya. He is the 2015 recipient of the Celso Furtado prize from the World Academy of Sciences for advancing the frontiers of social science (Economics) thought.

Maznah Dahlui is a public health physician with interest in health economics. She is a professor at University of Malaya and a known researcher and expert in economic evaluation of health interventions. She has conducted economic evaluation analysis with high impact to policy on the various Public Health programs in Malaysia.

Chiu Wan Ng is a public health physician. She is a professor at the Faulty of Medicine, University of Malaysia. Her main research interests are in health systems especially achievement of universal health coverage in low and medium income countries.

References

Adeokun

Okonkwo

Ladipo

(2006) The stigmatization of people living with HIV/AIDS. In: Adeyi

Kanki

Odutolu

, et al. (eds) AIDS in Nigeria: A Nation on the Threshold. New York: Harvard University Press, 213–233.

Anderson

Gerbing

(1988) Structural equation modeling in practice: A review and recommended two-step approach. Psychological Bulletin 103(3): 411–423.

Babalola

(2007) Readiness for HIV testing among young people in northern Nigeria: The roles of social norm and perceived stigma. AIDS and Behavior 11(5): 759–769.

Bagozzi

(1988) On the evaluation of structural equation models. Journal of the Academy of Marketing Science 16: 74–94.

Baron

Kenny

(1986) The moderator–mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology 51(6): 1173.

Berger

Ferrans

Lashley

(2001) Measuring stigma in people with HIV: Psychometric assessment of the HIV stigma scale. Research in Nursing & Health 24(6): 518–529.

Bos

AER

Pryor

Reeder

, et al. (2013) Stigma: Advances in theory and research. Basic and Applied Social Psychology 35(1): 1–9. DOI: 10.1080/01973533.2012.746147.

Bova

Nnaji

Woyah

, et al. (2016) HIV stigma, testing attitudes and health care access among African-born men living in the United States. Journal of Immigrant and Minority Health 18(1): 187–193.

Buldeo

Gilbert

(2015) Exploring the health belief model and first-year students’ responses to HIV/AIDS and VCT at a South African university. African Journal of AIDS Research 14(3): 209–218.

10.

Bwambale

Ssali

Byaruhanga

, et al. (2008) Voluntary HIV counselling and testing among men in rural western Uganda: Implications for HIV prevention. BMC Public Health 8(1): 263.

11.

Chin

(1998) The partial least squares approach to structural equation modeling. In: Marcoulides

(ed.) Modern Methods for Business Research. New Jersey: Lawrence Erlbaum Associate, 295–336.

12.

Chin

(2010) How to write up and report PLS analyses. In: Vinzi

Chin

Henseler

, et al. (eds) Handbook of Partial Least Squares: Concepts, Methods and Applications. Berlin and Heidelberg: Springer, 655–690.

13.

Chin

Marcolin

Newsted

(2003) A partial least squares latent variable modeling approach for measuring interaction effects: Results from a Monte Carlo simulation study and an electronic-mail emotion/adoption study. Information Systems Research 14(2): 189–217.

14.

De Jesus

Carrete

Maine

, et al. (2015) Attitudes, perceptions and behaviours towards HIV testing among African-American and East African immigrant women in Washington, DC: Implications for targeted HIV testing promotion and communication strategies. Sexually Transmission Infections 91(8): 569–575.

15.

Denison

McCauley

Dunnett-Dagg

, et al. (2008) The HIV testing experiences of adolescents in Ndola, Zambia: Do families and friends matter? AIDS Care-Psychological and Socio-Medical Aspects of AIDS/HIV 20(1): 101–105.

16.

Derlega

Winstead

Greene

, et al. (2002) Perceived HIV-related stigma and HIV disclosure to relationship partners after finding out about the seropositive diagnosis. Journal of Health Psychology 7(4): 415–432.

17.

Earnshaw

Jin

Wickersham

, et al. (2014) Exploring intentions to discriminate against patients living with HIV/AIDS among future healthcare providers in Malaysia. Tropical Medicine & International Health 19(6): 672–679.

18.

Earnshaw

Jin

Wickersham

, et al. (2016) Stigma toward men who have sex with men among future healthcare providers in Malaysia: Would more interpersonal contact reduce prejudice? AIDS and Behavior 20(1): 98–106.

19.

Feyissa

Abebe

Girma

, et al. (2012) Stigma and discrimination against people living with HIV by healthcare providers, Southwest Ethiopia. BMC Public Health 12: 1.

20.

Fisher

(1992) Changing AIDS-risk behavior. Psychological Bulletin 111(3): 455–474.

21.

Fisher

(2000) Theoretical approaches to individual-level change in HIV risk behavior. In: Peterson

DiClemente

(eds) Handbook of HIV Prevention. New York: Springer, 3–55.

22.

Fisher

Misovich

, et al. (1996) Changing AIDS risk behavior: Effects of an intervention emphasizing AIDS risk reduction information, motivation, and behavioral skills in a college student population. Health Psychology 15(2): 114–123.

23.

Fonner

Denison

Kennedy

, et al. (2012) Voluntary counseling and testing (VCT) for changing HIV-related risk behavior in developing countries. The Cochrane Database of Systematic Reviews 9: CD001224.

24.

Fornell

Larcker

(1981) Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research 18(1): 39–50.

25.

Frazier

Tix

Barron

(2004) Testing moderator and mediator effects in counseling psychology research. Journal of Counseling Psychology 51(1): 115–134.

26.

Gefen

Straub

Boudreau

M-C

(2000) Structural equation modeling and regression: Guidelines for research practice. Communications of the Association for Information Systems 4(1): 35–37.

27.

Genberg

Kawichai

Chingono

, et al. (2008) Assessing HIV/AIDS stigma and discrimination in developing countries. AIDS and Behavior 12(5): 772–780.

28.

Goffman

(1963) Stigma: Notes on the Management of Identity. London: Penguin.

29.

Gunzler

Chen

, et al. (2013) Introduction to mediation analysis with structural equation modeling. Shanghai Archives of Psychiatry 25(6): 390–394.

30.

Hair

Ringle

Sarstedt

(2011) PLS-SEM: Indeed a silver bullet. Journal of Marketing Theory and Practice 19(2): 139–152.

31.

Hair

Ringle

Sarstedt

(2013) Partial least squares structural equation modeling: Rigorous applications, better results and higher acceptance. Long Range Planning 46(1–2): 1–12.

32.

Hair

Sarstedt

Ringle

, et al. (2012) An assessment of the use of partial least squares structural equation modeling in marketing research. Journal of the Academy of Marketing Science 40: 414–433.

33.

Hayes

Preacher

(2014) Statistical mediation analysis with a multicategorical independent variable. British Journal of Mathematical and Statistical Psychology 67(3): 451–470.

34.

Herek

(1999) AIDS and stigma. American Behavioral Scientist 42(7): 1106–1116.

35.

Holmbeck

(2002) Post-hoc probing of significant moderational and mediational effects in studies of pediatric populations. Journal of Pediatric Psychology 27(1): 87–96.

36.

Jaccard

Wan

(1996) LISREL Approaches to Interaction Effects in Multiple Regression. Thousand Oaks, CA: SAGE.

37.

James

Brett

(1984) Mediators, moderators, and tests for mediation. Journal of Applied Psychology 69(2): 307–321.

38.

Janz

Champion

Strecher

(2002) The health belief model. In: Glanz

Rimer

Lewis

(eds) Health Behavior and Health Education: Theory, Research, and Practice. San Francisco: Jossey-Bass.

39.

Kock

(2014) Advanced mediating effects tests, multi-group analyses, and measurement model assessments in PLS-based SEM. International Journal of e-Collaboration (IJeC) 10(1): 1–13.

40.

Kock

(2015) PLS-based SEM algorithms: The good neighbor assumption, collinearity, and nonlinearity. Information Management and Business Review 7(2): 113–130.

41.

Kock

(2017) WarpPLS User Manual: Version 6.0. Available at: http://warppls.com/ (accessed 1 February 2018).

42.

Lindkvist

Johansson

Hylander

(2015) Fogging the issue of HIV – Barriers for HIV testing in a migrated population from Ethiopia and Eritrea. BMC Public Health 15: 82.

43.

Lyberg

Biemer

Collins

, et al. (eds) (1997) Survey Measurement and Process Quality. New York: John Wiley & Sons.

44.

Detels

Feng

, et al. (2007) Acceptance of and barriers to voluntary HIV counselling and testing among adults in Guizhou province, China. AIDS (London, England) 21(Suppl 8): S129–S135.

45.

Meiberg

Bos

AER

Onya

, et al. (2008) Fear of stigmatization as barrier to voluntary HIV counselling and testing in South Africa. East African Journal of Public Health 5(2): 49–54.

46.

National Agency for the Control of AIDS (2015) Global AIDS Response: Country Progress Report, Nigeria GARPR 2015. Abuja, Nigeria: National Agency for the Control of AIDS. Available at: http://www.unaids.org/sites/default/files/country/documents/NGA_narrative_report_2015.pdf (accessed 20 July 2017).

47.

National Agency for the Control of AIDS (2019) Revised National HIV and AIDS Strategic Framework 2019–2021: Future Directions for the HIV/AIDS Response in Nigeria. Abuja, Nigeria: National Agency for the Control of AIDS. Available at: https://naca.gov.ng/wp-content/uploads/2019/03/NATIONAL-HIV-AND-AIDS-STRATEGIC-FRAMEWORK-1.pdf (accessed 22 June 2019).

48.

National Population Commission (2017) Nigeria Population. Abuja, Nigeria: National Population Commission.

49.

Ndabarora

Mchunu

(2014) Factors that influence utilisation of HIV/AIDS prevention methods among university students residing at a selected university campus. Sahara J-Journal of Social Aspects of HIV-AIDS 11(1): 202–210.

50.

Niang

Tapsoba

Weiss

, et al. (2003) ‘It’s raining stones’: Stigma, violence and HIV vulnerability among men who have sex with men in Dakar, Senegal. Culture, Health & Sexuality 5: 499–512.

51.

Nigeria Demographic and Health Survey (2014) Nigeria Demographic and Health Survey 2013. Abuja, Nigeria: Nigeria Demographic and Health Survey. Available at: https://dhsprogram.com/pubs/pdf/FR293/FR293.pdf (accessed 1 February 2018).

52.

Nothling

Kagee

(2013) Acceptability of routine HIV counselling and testing among a sample of South African students: Testing the health belief model. African Journal of AIDS Research 12(3): 141–150.

53.

Nyblade

MacQuarrie

(2006) Can We Measure HIV/AIDS-Related Stigma and Discrimination? Current Knowledge About Quantifying Stigma in Developing Countries. Washington, DC: USAID Publications. Available at: http://www.policyproject.com/pubs/generalreport/Measure%20HIV%20Stigma.pdf (accessed 30 August 2018).

54.

Nyblade

MacQuarrie

Phillip

, et al. (2005) Working Report Measuring HIV Stigma: Results of a Field Test in Tanzania. Washington, DC: USAID Publications. Available at: https://www.icrw.org/wp-content/uploads/2016/10/Working-Report-Measuring-HIV-Stigma-Results-of-a-Field-Test-in-Tanzania.pdf (accessed 30 August 2018).

55.

Odimegwu

Adedini

Ononokpono

(2013) HIV/AIDS stigma and utilization of voluntary counselling and testing in Nigeria. BMC Public Health 13: 1–14.

56.

Okawa

Mwanza-Kabaghe

Mwiya

, et al. (2017) Adolescents’ experiences and their suggestions for HIV serostatus disclosure in Zambia: A mixed-methods study. Frontiers in Public Health 5: 326.

57.

Ostermann

Brown

Mühlbacher

, et al. (2015) Would you test for 5000 shillings? HIV risk and willingness to accept HIV testing in Tanzania. Health Economics Review 5: 24.

58.

Parker

Aggleton

(2003) HIV and AIDS-related stigma and discrimination: A conceptual framework and implications for action. Social Science & Medicine 57(1): 13–24.

59.

Peyrot

(1996) Causal analysis: Theory and application. Journal of Pediatric Psychology 21(1): 3–24.

60.

Piot

(2006) AIDS: From crisis management to sustained strategic response. The Lancet 368(9534): 526–530.

61.

Preacher

Hayes

(2008) Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behavior Research Methods 40: 879–891.

62.

Qian

Wang

Dong

, et al. (2007) Association of misconceptions about HIV transmission and discriminatory attitudes in rural China. AIDS Care-Psychological and Socio-Medical Aspects of AIDS/HIV 19(10): 1283–1287.

63.

Rosenstock

(2005) Why people use health services. Milbank Quarterly 83(4): Online-only.

64.

Sarker

Sanou

Snow

, et al. (2007) Determinants of HIV counselling and testing participation in a prevention of mother-to-child transmission programme in rural Burkina Faso. Tropical Medicine & International Health 12(2): 1475–1483.

65.

Sekoni

Obidike

Balogun

(2012) Stigma, medication adherence and coping mechanism among people living with HIV attending General Hospital, Lagos Island, Nigeria. African Journal of Primary Health Care & Family Medicine, 4(1): 1–10.

66.

Staveteig

Wang

Head

, et al. (2013) Demographic Patterns of HIV Testing Uptake in Sub-Saharan Africa. DHS Comparative Reports. Report no. 30. Calverton: ICF International.

67.

Steward

Herek

Ramakrishna

, et al. (2008) HIV-related stigma: Adapting a theoretical framework for use in India. Social Science & Medicine 67(8): 1225–1235.

68.

Ulasi

Preko

Baidoo

, et al. (2009) HIV/AIDS-related stigma in Kumasi, Ghana. Health Place 15(1): 255–262.

69.

United Nations Programme on HIV and AIDS (2014) 90-90-90: An Ambitious Treatment Target to Help End the AIDS Epidemic. Geneva: United Nations Programme on HIV and AIDS. Available at: https://www.unaids.org/sites/default/files/media_asset/90-90-90_en.pdf (accessed 1 February 2018).

70.

United Nations Programme on HIV and AIDS (2015) On the Fast-Track to End AIDS by 2030: Focus on Location and Population. Geneva: United Nations Programme on HIV and AIDS. Available at: https://www.unaids.org/sites/default/files/media_asset/WAD2015_report_en_part01.pdf (accessed 7 December 2017).

71.

United Nations Programme on HIV and AIDS (2018) UNAIDS Data 2018. Geneva: United Nations Programme on HIV/AIDS. Available at: https://www.unaids.org/sites/default/files/media_asset/unaids-data-2018_en.pdf (accessed 22 June 2019).

72.

Urbach

Ahlemann

(2010) Structural equation modeling in information systems research using partial least squares. Journal of Information Technology Theory and Application 11(2): 5.

73.

Van Brakel

(2006) Measuring health-related stigma—A literature review. Psychology, Health & Medicine 11(3): 307–334.

74.

Vorasane

Jimba

Kikuchi

, et al. (2017) An investigation of stigmatizing attitudes towards people living with HIV/AIDS by doctors and nurses in Vientiane, Lao PDR. BMC Health Services Research 17: 125.

75.

Wetzels

Odekerken-Schröder

Van Oppen

(2009) Using PLS path modeling for assessing hierarchical construct models: Guidelines and empirical illustration. MIS Quarterly 33(1): 177–195.

76.

Wong

KK-K

(2016) Mediation analysis, categorical moderation analysis, and higher-order constructs modeling in partial least squares structural equation modeling (PLS-SEM): A B2B example using SmartPLS. Marketing Bulletin 26: 1–22.

77.

World Health Organization (2015) Consolidated Guidelines on HIV Testing Services: 5Cs: Consent, Confidentiality, Counselling, Correct Results and Connection. Geneva: World Health Organization.

78.

Workagegn

Kiros

Abebe

(2015) Predictors of HIV-test utilization in PMTCT among antenatal care attendees in government health centers: Institution-based cross-sectional study using health belief model in Addis Ababa, Ethiopia, 2013. HIV AIDS-Research and Palliative Care 7: 215–222.

79.

Yang

Stanton

, et al. (2004) Willingness to participate in HIV/STD prevention activities among Chinese rural–to–urban migrants. AIDS Education and Prevention: Official Publication of the International Society for AIDS Education 16(6): 557–570.

80.

Yang

Stanton

, et al. (2006) HIV-related knowledge, stigma, and willingness to disclose: A mediation analysis. AIDS Care-Psychological and Socio-Medical Aspects of AIDS/HIV 18(7): 717–724.

81.

Yang

Kleinman

Link

, et al. (2007) Culture and stigma: Adding moral experience to stigma theory. Social Science & Medicine 64(7): 1524–1535.