Abstract
Suicidal ideation and behaviour (SIB) are among the psychiatric sequela of HIV/AIDS. Few studies have however examined the prevalence and correlates of SIB among persons seeking HIV testing. We set out to document the prevalence and correlates of SIB among people seeking HIV testing in peri-urban areas of Cape Town, South Africa. A cross-sectional research design was used to recruit a sample (n=500) of individuals seeking HIV testing. Self-report measures were used to assess two-week prevalence of SIB as well as lifetime prevalence of suicide attempt. A Structured Clinical Interview was used to assess common mental disorders (CMDs). Regression analysis was used to determine if CMD and socio-demographic variables predicted suicidal ideation. The mean age of the sample was 36 years, 51.6% were female and 46.6% were unemployed. The two-week prevalence of suicidal ideation was 24.27% while the two-week prevalence of suicide attempt and suicide plans was 2.8%. Suicidal ideation was not associated with age, gender, employment status, family income or household food insecurity. CMDs were significantly associated with suicidal ideation; individuals with depressive disorders were approximately 5.5 times more likely to report suicidal ideation, while those with generalised anxiety disorder, trauma related disorders and alcohol use disorder were approximately 7, 4.7 and 2.8 times more likely to report suicidal ideation respectively. Results suggest that persons seeking HIV testing may be a well-delineated group of persons at risk of suicide in this region of SA. Contact with the health care system during HIV testing provides an opportunity for targeted suicide prevention interventions in what appears to be a high risk group.
Early in the HIV epidemic researchers noted that suicidal ideation and behaviour (SIB) were among the psychiatric sequela of HIV/AIDS (Glass, 1988; Hull, Sewell, Wilson, & McFeeley, 1988; Kizer, Green, Perkins, Doebbert, & Hughes, 1988; Marzuk et al., 1988). Studies in high income western countries consistently documented association between HIV and suicide risk (Alfonso et al., 1994; O’Dowd, Biderman, & McKegney, 1993; Pueschel & Heineman, 1995). Much of the literature reports the prevalence and correlates of SIB among people already infected and has failed to take account of how suicide risk fluctuates over time; the implicit assumption is that receipt of an HIV+ diagnosis is a psychological trauma that precipitates SIB. It is within this context that we documented the prevalence and correlates of SIB among persons seeking HIV testing in in peri-urban areas of the Western Cape Province of South Africa (SA). We were also interested in the extent to which socio-demographic characteristics and psychiatric disorders were associated with SIB in this population.
Suicidal behaviour among persons living with HIV
Evidence from high income countries suggests that risk of suicide among persons living with HIV/AIDS (PLWHA) has fallen as treatment outcomes have improved (Basavaraj, Navya & Rashmi, 2010; Keiser et al., 2010; Lönnqvist, 2015). There are however still reports of high rates of SIB among PLWHA. Shirey (2013) reports rates of suicide among PLWHA in Switzerland are three times that of the general population, and, in the USA, one in five HIV+ patients report suicidal ideation (SI) in the previous week. A national registry study in Denmark of 9,900 men who died by suicide and 189,037 controls, found suicide risk was elevated among HIV+ individuals (Jia, Mehlum & Qin, 2012). Similar prevalence rates for SIB are reported from the African continent, which has some of the highest rates of HIV infection in the world.
A study of HIV+ patients attending a HIV clinic in Uganda reported prevalence rates of 13% for current SI (Petrushkin, Boardman & Ovuga, 2005). Twelve-month prevalence rates of attempted suicide among HIV+ adolescents in Uganda are 17.1% (Musisi & Kinyanda, 2009). A SA study reported rates of 16.8% for suicidality among PLWHA (Olley, Zeier, Seedat & Stein, 2005). Researchers from the United Republic of Tanzania reported that the rate of HIV has doubled amongst those who died by suicide compared to the national prevalence for sexually active adults (Ndosi, Mbonde & Lyamuya, 2004). A study of 295 HIV+ individuals attending a West African HIV clinic reported SI prevalence rates of 13.6%.
There are good public health reasons for investigating the prevalence and correlates of SIB among PLWHA; SIB predicts future fatal and non-fatal suicidal behaviour, and is associated with poor quality of life, poor antiretroviral therapy adherence and non-disclosure of HIV status to significant others (Sherr et al., 2008). Ogundipe, Olagunju and Adeyemi (2015) have noted that the paucity of information on the relationship between SIB and HIV impedes evidence-guided interventions. There may be an additional public health prerogative to explore the relationship between SIB and HIV in low and middle income countries, like SA, where rates of HIV infection are high (Zuma et al., 2016) and suicide is considered to be a serious public health problem (authors, 2013).
Correlates of suicidal ideation and behaviour
Evidence suggests that SIB is rare in the absence of mental disorders (Rihmer, 2007). Over 90% of adult suicides are associated with psychiatric disorders (Conwell, Duberstein, Cox & Herman, 1996), notably depressive disorders (Wulsin, Vaillant & Wells, 1999), psychotic illness (Hawton, Sutton, Haw, Sinclair & Harris, 2005), and posttraumatic stress disorder (Krysinska & Lester, 2010).
Evidence indicates that there are positive associations between non-fatal suicidal behaviour and mental illness (Fliege, Lee, Grimm & Klapp, 2009), most notably depressive disorders and borderline personality disorder (APA, 2013). SI is more than four times more likely among individuals with mental disorders; particularly anxiety disorders (OR=3.4), mood disorders (OR=4.7), impulse control disorders (OR=3.3), and substance use disorders (OR=2.8, 95%) (Nock et al., 2008).
Kinyanda, Hoskins, Nakku, Nawaz and Patel (2012) note that studies in high income countries report that SIB among PLWHA is associated with the following risk factors: socio-demographic factors (female gender, younger age, Black ethnicity); psychiatric conditions (substance abuse, major depressive disorder, antisocial personality disorder, previous attempted suicide); familial factors (family history of a psychiatric disorder and family history of attempted suicide); psychosocial factors (heterosexual orientation, multiple HIV related losses, lack of social support, loss of employment or insurance cover, exhaustion of financial resources, physical and sexual abuse); and clinical/medical features (such as painful and disfiguring physical deterioration, suffering from lipodystrophy-related symptoms, physical and psychological symptoms, and receipt of an AIDS diagnosis). A study of HIV+ individuals in rural USA found that SI was associated with depressive symptoms, less coping self-efficacy worry about transmitting HIV, and anxiety about AIDS-related stigma (Heckman et al., 2002). A national registry case control study in Denmark of men who died of suicide (N=9,900) found that suicide risk was associated with comorbid psychiatric illness, recent HIV+ diagnosis, and more intensive and frequent hospital care (Jia, Mehlum & Qin, 2012). A study in Seoul (N=457) found that independent risk factors for SI among HIV+ adults were: young and middle age, living with someone, history of AIDS-defining opportunistic disease, history of treatment for depression, lower social support, and lower psychological status (Kang et al., 2015). Kang et al. (2015) found that suicide attempts among PLWHA were independently associated with being a beneficiary of National Medical Aid, economic barriers to treatment, history of treatment for depression, and lower psychological status.
Kinyanda et al. (2012) note the relative scarcity of studies from Africa documenting correlates of SIB among PLWHA. Table 1 provides a summary of the results of studies on correlates of SIB among PLWHA in African countries.
Table 1.
Summary of studies conducted in Africa investigating correlates of Sib among PLWHA
| Study setting | Correlates of SIB | references |
|---|---|---|
| HIV+ South Africans | Female gender, psychiatric disorders, partner relational problems, poor social support, fear of disclosure and stigma, socio-economic pressures, and problems with cognitive flexibility, concentration and memory | Schlebusch and Vawda (2010) |
| PLWHA attending HIV clinics in semi-urban Uganda | Female gender, negative life events, a previous psychiatric history, and major depressive disorder | Kinyanda et al. (2012) |
| PLWHA attending HIV clinics in Uganda | Ftate anger, trait anger, depression, hopelessness, anxiety, low social support, inability to provide for others, and stigma | Rukundo, Mishara, & Kinyanda (2016) |
| PLWHA in West Africa | Unmarried, poor medication adherence, and poorer quality of life, unemployment, emotional distress, religion, HIV status non-disclosure and previous suicidal attempt. | Ogundipe et al. (2015). |
The paucity of research on SIB among persons seeking HIV testing has resulted in a lack of information about both the prevalence and correlates of SIB in this population. Our study thus represents an important first step in exploring socio-economic and psychiatric correlates of SIB among persons seeking HIV testing. Many of the studies that investigate psychiatric correlates of SIB among PWLHA assess symptoms of psychopathology using self-report symptom checklists. They thus report associations between SIB and symptoms of psychopathology rather than associations with specific psychiatric conditions. We were thus interested in investigating how SIB might be associated with psychiatric caseness among persons seeking HIV testing, rather than simply investigating symptoms of psychopathology.
Methods
Participants
We recruited a convenience sample of 500 individuals seeking HIV testing at three HIV testing sites in peri-urban areas surrounding Cape Town, SA. These sites were selected because of their proximity to residential areas in the greater Cape Town area and because of their accessibility to members of the public. Test-seekers were invited to participate in the study once they had registered at the reception desk of the testing site. Those who agreed to participate in the study were screened according to inclusion criteria. Eligibility criteria included not being floridly psychotic and being able to understand the interview questions in English. Only 40 potential participants had to be excluded because they were not conversant in English.
Ethics
The study was approved by the Stellenbosch University Health Ethics Committee. All participants gave signed informed consent to participate. Participants found to have clinically significant distress or a mental disorder were referred to a local mental health centre. Participants received a R50 (equivalent to approximately $3.40) grocery voucher.
Measuring instruments
The following self-report data were collected by post-graduate psychology students who had been trained and supervised by registered psychologists:
Socio-demographic information
Participants were asked their age, a description of their race, gender, marital status, level of education, employment status, and annual family income.
Food insecurity
The Household Food Insecurity Scale was used to assess the level of food insecurity during the past 30 days (Coates, Swindale & Bilinsky, 2007).
Socio-economic status
The socio-economic status was assessed using the Socio-demographic Information Questionnaire (Svavarsdottir & Orlygsdottir, 2006).
Common mental disorder
The following modules of the Structured Clinical Interview Schedule (SCID) (First, 1995) adapted for DSM 5 (APA, 2013) were administered: depressive disorders, anxiety disorders, trauma and stress related disorders, and alcohol use disorder. We focused on these disorders because they are the most common forms of mental illness in SA (Joe, Stein, Seedat, Herman, & Williams, 2008).
Suicidal ideation
Information on the two-week prevalence of suicidal ideation was obtained from item nine on the Beck Depression Inventory which asks individuals if they currently have no thoughts of killing themselves (absence of suicidal behaviour), thoughts of killing themselves without any intention to follow through (passive suicidal ideation), a desire to be dead (suicidal desire), or if they would kill themselves given the opportunity (suicidal intention) (Beck, Steer, Garbin, 1996). The BDI has demonstrated excellent reliability and validity in a variety of studies internationally and in SA (see for example, Foa, Riggs, Dancu, & Rothbaum, 1993; Martin & Kagee, 2010).
Suicidal behaviour
Lifetime and two-week prevalence of suicide attempts and plans were assessed using the major depressive disorder modules of the SCID (First, 1995).
Data collection
Data were captured on a Lenovo tablet via a web-based platform to facilitate data storage and manipulation. Data were collected prior to participants undergoing HIV testing.
Data analysis
Data were entered into Statistical Package for the Social Sciences (SPSS v. 19) (Norusis, 1990) and analysed using descriptive and multivariate statistics. Logistic regression analysis was used to determine whether suicidal ideation was associated with any of the socio-demographic variables, SES or food insecurity. Odds ratios were calculated to investigate the association between common mental disorders (CMDs) and SI. Hierarchical regression analysis was used to determine the relationship between SI and past suicide attempt, depressive disorders (i.e. major depressive disorder and persistent depressive disorder, generalised anxiety disorder (GAD), trauma and stress related disorders (i.e., acute stress disorder and posttraumatic stress disorder), and alcohol use disorder (AUD).
Results
Sample characteristics
A total of 500 participants took part in the study of which 51.6% were females. The mean age of the sample was 36 years. The demographic features and prevalence of CMDS are illustrated in Table 2. Most participants identified as Coloured1 (72.6%) and reported that they were Afrikaans speaking (69.0%). Only 2% of participants tested HIV+. Depressive disorders were the most common form of psychopathology (22.41%) followed by AUD (20.33%).
Table 2.
Demographic and clinical characteristics of the sample
| Number of respondents (n=500) |
% of total sample | |
|---|---|---|
| Gender | ||
| Male | 242 | 48.4 |
| Female | 258 | 51.6 |
| Age (years) | ||
| 18 – 19 | 27 | 5.6 |
| 20 – 29 | 147 | 30.2 |
| 30 – 39 | 136 | 28.0 |
| 40 – 49 | 99 | 20.4 |
| 50 – 71 | 77 | 15.8 |
| Race | ||
| Black | 131 | 26.2 |
| Coloured | 363 | 72.6 |
| White | 4 | 0.8 |
| Other | 2 | 0.4 |
| Marital status | ||
| Single | 265 | 53.6 |
| Widowed | 13 | 2.6 |
| Separated | 28 | 5.6 |
| Divorced | 35 | 7.0 |
| Married/living together | 159 | 31.8 |
| First language | ||
| Afrikaans | 345 | 69.0 |
| English | 30 | 6.0 |
| Xhosa | 98 | 19.6 |
| Other | 27 | 5.4 |
| Educational level | ||
| No formal education | 23 | 4.6 |
| Completed primary school | 24 | 4.8 |
| Attended secondary school | 289 | 57.8 |
| Completed secondary school | 125 | 25.0 |
| Attended tertiary institution | 22 | 4.4 |
| Graduated from tertiary institution | 17 | 3.4 |
| Current work situation | ||
| Employed fulltime | 97 | 19.4 |
| Employed part-time | 106 | 21.2 |
| Unemployed | 233 | 46.6 |
| Homemaker | 11 | 2.2 |
| Student | 29 | 5.8 |
| Receiving a disability grant | 7 | 1.4 |
| Retired | 17 | 3.4 |
| Annual family income | ||
| Less than *ZAR10 000 | 203 | 40.6 |
| ZAR 10 001 – ZAR40 000 | 205 | 41.0 |
| ZAR 40 001 – ZAR 80 000 | 57 | 11.4 |
| ZAR 80 001 – ZAR 110 000 | 20 | 4.0 |
| ZAR 110 001 – ZAR 170 000 | 8 | 1.6 |
| ZAR 170 001 – ZAR 240 000 | 4 | 0.8 |
| ZAR 240 000 and above | 3 | 0.6 |
| Prevalence of common mental disorders | ||
| Major Depressive Disorder | 72 | 14.94% |
| Persistent Depressive Disorder | 36 | 7.47% |
| Generalized Anxiety Disorder | 17 | 3.53% |
| Acute stress disorder | 2 | 0.41% |
| Posttraumatic stress disorder | 25 | 5.19% |
| Alcohol Use Disorder | 98 | 20.33% |
ZAR 14.76 = 1USD at June 2016 prices
Prevalence of suicidal ideation and behaviour
Twenty-four percent of the sample reported SI in the previous two weeks (; i.e., having suicidal thoughts (18.05%); having a desire to kill themselves (2.07%), and stating that they would kill themselves given the opportunity (4.15%). No significant gender differences were observed with respect to SI (x2=0.362, p=.547).
A total of 5.20% individuals reported a life-time prevalence of suicide attempt while 2.8% reported a two-week prevalence of suicide attempt and suicide plans. No significant gender differences were observed with respect to life-time prevalence of suicide attempt (x2=0.408, p=.523) or two-week prevalence of suicide plans and attempt (x2=2.268, p=.132).
Socio-demographic predictors
Logistic regression analysis was performed to determine the association between two-week prevalence of SI and demographic variables. The results of this analysis are presented in Tables 4 and 5. No significant association was found between SI and demographic characteristics. Similarly, logistic regression analysis indicated that there was no statistically significant relationship between food insecurity and SI.
Table 4.
Parameters for logistic regression analysis to investigate the relationship between demographic factors and food insecurity as predictive variables and suicidal ideations as the criterion variable
| Model | Unstandardized Coefficients | Standardized Coefficients | t | Sig. | ||
|---|---|---|---|---|---|---|
|
| ||||||
| B | Std. Error | Beta | ||||
|
| ||||||
| 1 | (Constant) | .215 | .223 | .961 | .337 | |
| Age | −.002 | .002 | −.043 | −.785 | .433 | |
| Gender | .015 | .040 | .017 | .365 | .715 | |
| Race | −.011 | .073 | −.012 | −.149 | .882 | |
| Marital status | −.022 | .013 | −.094 | −1.770 | .077 | |
| Living situation | .018 | .019 | .044 | .938 | .349 | |
| Educational level | .014 | .023 | .030 | .614 | .540 | |
| Employment status | .019 | .010 | .090 | 1.884 | .060 | |
| Annual family income | −.014 | .020 | −.036 | −.730 | .466 | |
| Place of Birth | .044 | .029 | .070 | 1.515 | .130 | |
| First Language | −.045 | .037 | −.100 | −1.230 | .219 | |
|
| ||||||
| 2 | (Constant) | .061 | .482 | .126 | .900 | |
| Food insecurity | .021 | .023 | .137 | .938 | .353 | |
Model 1: Age, gender, race, marital status, living situation, education level, employment status, annual family income, place of birth, first language
Model 2: Model: Household food insecurity
Table 5.
Summary statistics for the regression analysis in which the relationship was investigated between demographic factors and food insecurity as predictive variables and suicidal ideations as the criterion variable
| Model | R | R Square | Adjusted R Square | Std. Error of the Estimate | Change Statistics | Durbin-Watson | ||||
|---|---|---|---|---|---|---|---|---|---|---|
| R Square Change | F Change | df1 | df2 | Sig. F Change | ||||||
| 1 | .194 | .038 | .017 | .42615 | .038 | 1.784 | 10 | 457 | .061 | 1.938 |
| 2 | .137 | .019 | −.003 | .79997 | .019 | 0.879 | 1 | 46 | .353 | 2.473 |
Model 1: Age, gender, race, marital status, living situation, education level, employment status, annual family income, place of birth, first language
Model 2: Household food insecurity
Mental disorder
Table 6 shows the number of persons reporting SI for each group of CMDs. Odds ratios were calculated to determine the level of association between each group of disorders and two-week prevalence of SI. Having a CMD was significantly associated with two-week prevalence of SI; individuals with depressive disorders were approximately 5.5 times more likely to report SI, while those with GAD, trauma related disorders and AUD were approximately 7, 4.7 and 2.8 times more likely to report SI respectively.
Table 6.
Association between common mental disorders and suicidal ideation among persons seeking HIV testing (N=500)
| Suicidal ideation | X2 | p | OR | 95% CI | |||
|---|---|---|---|---|---|---|---|
| Yes | No | ||||||
| Depressive disorders (n=482) | Yes | 46 | 41 | 47.24 | 0.00 | 5.1 | 3.038–9.641 |
| No | 71 | 324 | |||||
| Generalised Anxiety Disorder (n=459) | Yes | 10 | 5 | 15.77 | 0.00 | 6.9 | 2.129–24.035 |
| No | 99 | 345 | |||||
| Trauma and stress related disorders (n=482) | Yes | 15 | 11 | 16.697, | 0.00 | 4.73, | 1.977–11.432 |
| No | 102 | 354 | |||||
| Alcohol Use Disorders (n=472) | Yes | 35 | 57 | 17.046, | 0.00 | 2.797 | 1.644 – 4.755 |
| No | 82 | 308 | |||||
Hierarchical logistic regression analysis was performed to determine the association between SI and past suicide attempt, depressive disorders, GAD, trauma and related disorders and AUD. Table 7 displays the predictor variables as they were entered into this model, together with their standardised regression coefficients and significance levels. Table 8 shows the summary statistics for the regression analysis. The first model consisting only of past suicide attempt was significant and explained 2.3% of the variance in SI. When depressive disorders were added in the second model there was a significant increase in explained variance with depressive disorders accounting for an additional 8.3% of the variance. When GAD was added to the model there was also a significant increase in explained variance, as was the case when trauma and related disorders, and AUD, were added. GAD accounted for an additional 2.1% of the variance, while trauma and related disorders, and AUD respectively accounted for an additional 1.0% and 1.6% of the variance. In this final model, past suicide attempt and CMDs accounted for 14.6% of the variance in SI.
Table 7.
Parameters for all the regression analysis to establish the association between past suicide attempt, depressive disorders, generalised anxiety disorder, trauma and related disorders, and substance use disorders, with suicidal ideation
| Model | Unstandardized Coefficients | Standardized Coefficients | T | Sig. | ||
|---|---|---|---|---|---|---|
|
| ||||||
| B | Std. Error | Beta | ||||
|
| ||||||
| 1 | (Constant) | .221 | .020 | 10.945 | .000 | |
| Past suicide attempt | .075 | .022 | .159 | 3.451 | .001 | |
|
| ||||||
| 2 | (Constant) | .169 | .021 | 8.077 | .000 | |
| Past suicide attempt | .052 | .021 | .110 | 2.458 | .014 | |
| Depressive disorders | .327 | .050 | .292 | 6.507 | .000 | |
|
| ||||||
| 3 | (Constant) | .162 | .021 | 7.755 | .000 | |
| Past suicide attempt | .053 | .021 | .112 | 2.532 | .012 | |
| Depressive disorders | .304 | .050 | .271 | 6.041 | .000 | |
| Generalised anxiety disorder | .349 | .106 | .146 | 3.301 | .001 | |
|
| ||||||
| 4 | (Constant) | .156 | .021 | 7.472 | .000 | |
| Past suicide attempt | .051 | .021 | .108 | 2.448 | .015 | |
| Depressive disorders | .285 | .051 | .254 | 5.610 | .000 | |
| Generalised anxiety disorder | .313 | .106 | .131 | 2.937 | .003 | |
| Trauma and stress related disorders | .193 | .084 | .103 | 2.290 | .023 | |
|
| ||||||
| 5 | (Constant) | .132 | .022 | 5.961 | .000 | |
| Past suicide attempt | .051 | .021 | .108 | 2.462 | .014 | |
| Depressive disorders | .267 | .051 | .238 | 5.258 | .000 | |
| Generalised anxiety disorder | .310 | .106 | .129 | 2.933 | .004 | |
| Trauma and stress related disorders | .187 | .084 | .100 | 2.242 | .025 | |
| Alcohol use disorder | .138 | .047 | .129 | 2.959 | .003 | |
Model 1: Past suicide attempts as predictors of suicidal ideation
Model 2: Depressive disorders as predictors of suicidal ideation
Model 3: Depressive disorders and anxiety disorders as predictors of suicidal ideation
Model 4: Depressive, anxiety and trauma-related disorders as predictors of suicidal ideation
Model 5: Depressive, anxiety, trauma-related and substance use disorders as predictors of suicidal ideation
Table 8.
Summary statistics for the regression analysis to establish the association between past suicide attempt, depressive disorders, generalised anxiety disorder, trauma and related disorders, and substance use disorders, with suicidal ideation
| Model | R | R Square | Adjusted R Square | Std. Error of the Estimate | Change Statistics
|
Durbin-Watson | ||||
|---|---|---|---|---|---|---|---|---|---|---|
| R Square Change | F Change | df1 | df2 | Sig. F Change | ||||||
|
| ||||||||||
| 1 | .159a | .025 | .023 | .42102 | .025 | 11.907 | 1 | 457 | .001 | |
| 2 | .329b | .108 | .104 | .40318 | .083 | 42.335 | 1 | 456 | .000 | |
| 3 | .359c | .129 | .123 | .39887 | .021 | 10.895 | 1 | 455 | .001 | |
| 4 | .373d | .139 | .131 | .39703 | .010 | 5.242 | 1 | 454 | .023 | |
| 5 | .394e | .155 | .146 | .39368 | .016 | 8.757 | 1 | 453 | .003 | 1.952 |
Model 1: Past suicide attempts as predictors of suicidal ideation
Model 2: Depressive disorders as predictors of suicidal ideation
Model 3: Depressive disorders and anxiety disorders as predictors of suicidal ideation
Model 4: Depressive, anxiety and trauma-related disorders as predictors of suicidal ideation
Model 5: Depressive, anxiety, trauma-related and substance use disorders as predictors of suicidal ideation
Discussion
The two-week prevalence of SI among HIV test seekers in our sample was 24.27%. The lack of comparable data on the characteristics of persons seeking HIV testing makes it difficult to draw meaningful comparisons. However, the 24.27% prevalence rate for SI in our sample is almost three times higher than the 9.1% life-time prevalence of SI among a nationally representative sample of the general population of SA (Joe, Stein, Seedat, Herman, & Williams, 2008). Similarly, the life-time prevalence of suicide attempt among persons seeking HIV testing was 5.2%, compared to the 2.9% reported by Joe et al. (2008) for SA’s general population. This data suggest that HIV test seekers in this region may constitute a clearly delineated group of people at higher risk of suicide than the general population. These individuals have come into contact with the health care system, by virtue of seeking an HIV test, which may provide an opportunity for targeted suicide prevention interventions.
The prevalence rates for SI in our sample were also higher than the prevalence rates of 16.8 % for suicidality found among HIV+ patients in SA (Olley et al., 2005). This raises questions about the extent to which SIB may predate receipt of an HIV+ diagnosis and the assumption that being diagnosed as HIV+ is psychologically distressing and precipitates suicidality. Longitudinal studies which trace the progression of SIB over time are needed to further illuminate the nature of the relationship between HIV and SIB.
In our sample, SI was not associated with age, gender, unemployment, family income or food insecurity. This is inconsistent with much of the literature on SI in community samples and among PLWHA. Our findings may be in part a result of the relative homogeneity of our sample in terms of socio-economic status. Future studies which draw from a broader cross-section of individuals may help to identify the socio-economic and contextual factors that contribute to SIB in this population.
In our sample, SI was significantly associated with past suicide attempt and depressive disorders, GAD, trauma and stress related disorders, and AUD. This finding is consistent with previous studies which found strong associations between psychiatric symptoms and SI among PLWHA. In our sample, however, CMDs and past suicide attempt only accounted for 14.6% of the variance in SI, which raises questions about what other factors might account for the elevated rates of SI. This is an area in need of further research, with particular focus on non-psychiatric contributors to SIB. Given the shortage of mental health care professionals in SA and the significant mental health treatment gap (Jack et al., 2014), it would be important to identify suicide prevention interventions that move beyond focusing only on the mental health determinants of suicidality in this population.
Our data draw attention to the high rates of SIB among persons seeking HIV testing in in the Western Cape province of SA. Further research is needed to establish how generalizable these results are to other HIV testing sites in SA. Rates of HIV infection are lower in the Western Cape than in other parts of the country (Simbayi et al. 2014); it would thus be important to replicate this study in regions where the prevalence of HIV (and therefore the likelihood of testing HIV positive) is higher.
Limitations
This is a cross sectional study which employs convenient sampling. This calls into question the extent to which our findings can be generalised for all HIV test seekers. Nonetheless this work represents an important first step in arguing that persons seeking HIV testing are a well delineated group who come into contact with the health care system and may have particular characteristics and psychological needs.
Our study focused on the association between SI and a diagnosis for a CMD. We did not consider the issue of symptom count and did not include any assessment of personality function, psychiatric comorbidity or use of substance use disorders other than AUD. It may be useful for future studies in this area to explore these issues, particularly given that SIB have been associated with psychiatric co-morbidity, personality disorders, severity of psychiatric symptoms and the use of substances other than alcohol (Hawton & Van Heeringen, 2000). Future studies might also focus on cluster B personality disorders in this population and their interaction with risk taking, impulsivity, increased risk of HIV and SIB.
Conclusion
High rates of SIB among persons seeking HIV testing has implications for suicide prevention. Rates of fatal and non-fatal suicidal behaviour in SA are high and suicide is considered a serious public health problem. Our data suggest that persons seeking HIV testing in peri-urban areas of Cape Town might constitute a well delineated population who is at particular risk for suicide. Contact with the health care system while seeking HIV testing provides an opportunity for targeted suicide prevention interventions. Hitherto this opportunity for intervention has been missed, in part because HIV testing has been uncoupled from the provision of other medical and psychosocial support services in SA. In an effort to provide greater access to HIV testing in SA, a large number of non-medical HIV testing sites have been established. In the context of our data, it would appear that there is a need for the integration of mental health services, especially screening for mental disorders and referral of those persons who are in need of services, as part of a national suicide prevention strategy.
Table 3.
Prevalence of suicidal ideation and behaviours among persons seeking HIV testing (N=500)
| Male | Female | Total | |||||||
|---|---|---|---|---|---|---|---|---|---|
| n | % | n | % | n | % | X2 | p | ||
| Suicidal ideation (n=482, missing data = 18) | No suicidal ideation | 177 | 76.96% | 188 | 0.746 | 365 | 75.73% | 0.362, | .547 |
| Thoughts of killing myself | 39 | 16.96% | 48 | 19.05% | 87 | 18.05% | |||
| Desire to kill myself | 3 | 1.30% | 7 | 2.78% | 10 | 2.07% | |||
| Suicidal intent given the opportunity | 11 | 4.78% | 9 | 3.57% | 20 | 4.15% | |||
| Suicidal behaviour (n=500) | 2-week prevalence of suicide plan | 0 | 0.00% | 6 | 2.33% | 6 | 1.20% | 2.268 | 0.132 |
| 2-week prevalence of suicide attempt | 4 | 1.65% | 4 | 1.55% | 8 | 1.60% | |||
| Life-time prevalence of suicide attempt | 11 | 4.55% | 15 | 5.81% | 26 | 5.20% | 0.408 | 0.523 | |
Acknowledgments
Funding
This work was supported by a South African Medical research Council grant awarded to XXXXXX
Footnotes
In Southern Africa, the term Coloured is an official ethnic label for people of mixed ethnic origin who possess ancestry from Europe, Asia, and various Khoisan and Bantu ethnic groups.
Disclosure statement
No potential conflict of interest was reported by the authors.
References
- Alfonso CA, Cohen MAA, Aladiem AD, Morrison F, Powell DR, Winters RA, Orlowski BK. HIV seropositivity as a major risk factor for suicide in the general hospital. Psychosomatics. 1994;35(4):368–373. doi: 10.1016/s0033-3182(94)71758-8. [DOI] [PubMed] [Google Scholar]
- American Psychiatric Association. Diagnostic and statistical manual of mental disorders (DSM-5®) American Psychiatric Pub; 2013. [Google Scholar]
- Bantjes J, Kagee A. Epidemiology of suicide in South Africa: setting an agenda for future research. South African Journal of Psychology. 2013;43(2):238–251. [Google Scholar]
- Basavaraj KH, Navya MA, Rashmi R. Quality of life in HIV/AIDS. Indian Journal of Sexually Transmitted Diseases and AIDS. 2010;31(2):75. doi: 10.4103/2589-0557.74971. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Beck AT, Steer RA, Brown GK. Beck depression inventory-II. San Antonio, TX: 1996. 78204-2498. [Google Scholar]
- Coates J, Swindale A, Bilinsky P. Household Food Insecurity Access Scale (HFIAS) for measurement of food access: indicator guide. Washington, DC: Food and Nutrition Technical Assistance Project, Academy for Educational Development; 2007. [Google Scholar]
- Conwell Y, Duberstein PR, Cox C, Herrmann JH. Relationship of age and axis I diagnoses in victims of completed suicide: A psychological autopsy study. American Journal of Psychiatry. 1996;153(8):1001–1008. doi: 10.1176/ajp.153.8.1001. [DOI] [PubMed] [Google Scholar]
- First MB. Structured clinical interview for the DSM (SCID) John Wiley & Sons, Inc.; 1995. [Google Scholar]
- Fliege H, Lee JR, Grimm A, Klapp BF. Risk factors and correlates of deliberate self-harm behavior: A systematic review. Journal of Psychosomatic Research. 2009;66(6):477–493. doi: 10.1016/j.jpsychores.2008.10.013. [DOI] [PubMed] [Google Scholar]
- Foa EB, Riggs DS, Dancu CV, Rothbaum BO. Reliability and validity of a brief instrument for assessing post‐traumatic stress disorder. Journal of traumatic stress. 1993;6(4):459–473. [Google Scholar]
- Glass RM. AIDS and suicide. JAMA. 1988;259(9):1369–1370. [PubMed] [Google Scholar]
- Hawton K, Sutton L, Haw C, Sinclair J, Harris L. Suicide and attempted suicide in bipolar disorder: A systematic review of risk factors. Journal of Clinical Psychiatry. 2005;66:693–704. doi: 10.4088/jcp.v66n0604. [DOI] [PubMed] [Google Scholar]
- Heckman TG, Miller J, Kochman A, Kalichman SC, Carlson B, Silverthorn M. Thoughts of suicide among HIV-infected rural persons enrolled in a telephone-delivered mental health intervention. Annals of Behavioral Medicine. 2002;24(2):141–148. doi: 10.1207/S15324796ABM2402_11. [DOI] [PubMed] [Google Scholar]
- Hawton K, Van Heeringen K, editors. The international handbook of suicide and attempted suicide. John Wiley & Sons; 2000. [Google Scholar]
- Hull HF, Sewell CM, Wilson J, McFeeley P. The risk of suicide in persons with AIDS. JAMA. 1988;260(1):29–30. [PubMed] [Google Scholar]
- Jack H, Wagner RG, Petersen I, Thom R, Newton CR, Stein A, Hofman KJ. Closing the mental health treatment gap in South Africa: a review of costs and cost-effectiveness. Global Health Action. 2014;7:137–147. doi: 10.3402/gha.v7.23431. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Jia CX, Mehlum L, Qin P. AIDS/HIV infection, comorbid psychiatric illness, and risk for subsequent suicide: a nationwide register linkage study. Journal of Clinical Psychiatry. 2012;73(10):1315–1321. doi: 10.4088/JCP.12m07814. [DOI] [PubMed] [Google Scholar]
- Joe S, Stein DJ, Seedat S, Herman A, Williams DR. Non-fatal suicidal behavior among South Africans. Social Psychiatry and Psychiatric Epidemiology. 2008;43(6):454–461. doi: 10.1007/s00127-008-0348-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kagee A, Martin L. Symptoms of depression and anxiety among a sample of South African patients living with HIV. AIDS care. 2010;22(2):159–165. doi: 10.1080/09540120903111445. [DOI] [PubMed] [Google Scholar]
- Kang CR, Bang JH, Cho SI, Kim KN, Lee HJ, Ryu BY, Lee JK. Suicidal ideation and suicide attempts among human immunodeficiency virus-infected adults: differences in risk factors and their implications. AIDS Care. 2015:1–8. doi: 10.1080/09540121.2015.1093593. [DOI] [PubMed] [Google Scholar]
- Knueppel D, Demment M, Kaiser L. Validation of the household food insecurity access scale in rural Tanzania. Public health nutrition. 2010;13(03):360–367. doi: 10.1017/S1368980009991121. [DOI] [PubMed] [Google Scholar]
- Keiser O, Spoerri A, Brinkhof MW, Hasse B, Gayet-Ageron A, Tissot F, Egger M. Suicide in HIV-infected individuals and the general population in Switzerland, 1988–2008. American Journal of Psychiatry. 2010;167:143–150. doi: 10.1176/appi.ajp.2009.09050651. [DOI] [PubMed] [Google Scholar]
- Kinyanda E, Hoskins S, Nakku J, Nawaz S, Patel V. The prevalence and characteristics of suicidality in HIV/AIDS as seen in an African population in Entebbe district, Uganda. BMC Psychiatry. 2012;12(1):1. doi: 10.1186/1471-244X-12-63. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kizer KW, Green M, Perkins CI, Doebbert G, Hughes MJ. AIDS and suicide in California. Jama. 1988;260(13):1881. [PubMed] [Google Scholar]
- Krysinska K, Lester D. Post-traumatic stress disorder and suicide risk: a systematic review. Archives of Suicide Research. 2010;14(1):1–23. doi: 10.1080/13811110903478997. [DOI] [PubMed] [Google Scholar]
- Lönnqvist J. Physical illnesses and suicide. Suicide: An unnecessary death. 2015;125 [Google Scholar]
- Marzuk PM, Tierney H, Tardiff K, Gross EM, Morgan EB, Hsu MA, Mann JJ. Increased risk of suicide in persons with AIDS. Jama. 1988;259(9):1333–1337. [PubMed] [Google Scholar]
- Norusis MJ. Statistical package for the social sciences. Professional statistics release. 1990:7. [Google Scholar]
- Musisi S, Kinyanda E. Emotional and behavioural disorders in HIV seropositive adolescents in urban Uganda. East African Medical Journal. 2009;86(1):16–24. doi: 10.4314/eamj.v86i1.46923. [DOI] [PubMed] [Google Scholar]
- Ndosi N, Mbonde M, Lyamuya E. Profile of suicide in Dar es Salaam. East African Medical Journal. 2004;81(4):207–211. doi: 10.4314/eamj.v81i4.9157. [DOI] [PubMed] [Google Scholar]
- Nock MK, Borges G, Bromet EJ, Alonso J, Angermeyer M, Beautrais A, Williams D. Cross-national prevalence and risk factors for suicidal ideation, plans and attempts. The British Journal of Psychiatry. 2008;192(2):98–105. doi: 10.1192/bjp.bp.107.040113. [DOI] [PMC free article] [PubMed] [Google Scholar]
- O’Dowd MA, Biderman DJ, McKegney FP. Incidence of suicidality in AIDS and HIV-positive patients attending a psychiatry outpatient program. Psychosomatics. 1993;34(1):33–40. doi: 10.1016/s0033-3182(93)71925-8. [DOI] [PubMed] [Google Scholar]
- Ogundipe OA, Olagunju AT, Adeyemi JD. Suicidal ideation among attendees of a West African HIV clinic. Archives of Suicide Research. 2015;19(1):103–116. doi: 10.1080/13811118.2014.915776. [DOI] [PubMed] [Google Scholar]
- Olley BO, Zeier MD, Seedat S, Stein DJ. Post-traumatic stress disorder among recently diagnosed patients with HIV/AIDS in South Africa. AIDS Care. 2005;17(5):550–557. doi: 10.1080/09540120412331319741. [DOI] [PubMed] [Google Scholar]
- Petrushkin A, Boardman J, Ovuga E. Psychiatric disorders in HIV-positive individuals in urban Uganda. Psychiatric Bulletin. 2005;29:455–458. [Google Scholar]
- Pueschel K, Heineman A. HIV and suicide in Hamburg. AIDS Care. 1995;7(5):129–134. doi: 10.1080/09540129550126119. [DOI] [PubMed] [Google Scholar]
- Rihmer Z. Suicide risk in mood disorders. Current Opinion in Psychiatry. 2007;20:17–22. doi: 10.1097/YCO.0b013e3280106868. [DOI] [PubMed] [Google Scholar]
- Rukundo GZ, Mishara BL, Kinyanda E. Burden of suicidal ideation and attempt among Persons Living with HIV and AIDS in semiurban Uganda. AIDS Research and Treatment. 2016 doi: 10.1155/2016/3015468. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Schlebusch L, Vawda N. HIV-infection as a self-reported risk factor for attempted suicide in South Africa. African Journal of Psychiatry. 2010;13(4):280–283. doi: 10.4314/ajpsy.v13i4.61877. [DOI] [PubMed] [Google Scholar]
- Sherr L, Lampe F, Fisher M, Arthur G, Anderson J, Zetler S, Harding R. Suicidal ideation in UK HIV clinic attenders. AIDS. 2008;22:1651–1658. doi: 10.1097/QAD.0b013e32830c4804. [DOI] [PubMed] [Google Scholar]
- Shirey KG. Mental health practitioner’s guide to HIV/AIDS. New York: Springer; 2013. Suicide and HIV; pp. 405–407. [Google Scholar]
- Simbayi L, Shisana O, Rehle T, Onoya D, Jooste S, Zungu N, Zuma K. South African national HIV prevalence, incidence and behaviour survey, 2012. Pretoria: Human Sciences Research Council; 2014. [DOI] [PubMed] [Google Scholar]
- Svavarsdottir EK, Orlygsdottir B. Sociodemographic information questionnaire. Reykjavik, Iceland: University of Iceland, Faculty of Nursing; 2006. [Google Scholar]
- Wulsin LR, Vaillant GE, Wells VE. A systematic review of the mortality of depression. Psychosomatic Medicine. 1999;61(1):6–17. doi: 10.1097/00006842-199901000-00003. [DOI] [PubMed] [Google Scholar]
- Zuma K, Shisana O, Rehle TM, Simbayi LC, Jooste S, Zungu N, Labadarios D, Onoya D, Evans M, Moyo S, Abdullah F. New insights into HIV epidemic in South Africa: key findings from the National HIV Prevalence, Incidence and Behaviour Survey, 2012. African Journal of AIDS Research. 2016;15(1):67–75. doi: 10.2989/16085906.2016.1153491. [DOI] [PubMed] [Google Scholar]