Abstract
The rate of suicide among people with HIV (PWH) remains elevated compared to the general population. The aim of the study was to examine the association between a broad range of risk factors, HIV-specific risk factors, and suicide. We conducted a nested case-control study using data from the Veterans Aging Cohort Study (VACS) between 2006 and 2015. The risk of suicide was estimated using conditional logistic regression and models were stratified by HIV status. Most risk factors associated with suicide were similar between PWH and people without HIV; these included affective disorders, use of benzodiazepines, and mental health treatment. Among PWH, HIV-specific risk factors were not associated with suicide. A multiplicative interaction was observed between a diagnosis of HIV and a previous suicide attempt. Among PWH, a high prevalence of psychiatric, substance use disorders and multimorbidity contribute to the risk of suicide.
Resumen
La tasa de suicidio entre las personas con VIH (PWH) sigue siendo elevada en comparación con la población general. El objetivo del estudio fue examinar la asociación entre un amplio rango de factores de riesgo, los riesgos específicos del VIH y el suicidio. Realizamos un estudio anidado de casos y controles usando datos del Veterans Aging Cohort Study (VACS) entre 2006–2015. El riesgo de suicidio fue estimado mediante regresión logística condicional y los modelos se estratificaron por estado serológico. La mayoría de los factores de riesgo asociados con el suicidio fueron similares entre las PWH y las personas sin VIH; estos incluyeron trastornos afectivos, uso de benzodiazepinas y tratamiento de salud mental. Entre las PWH, los factores de riesgo específicos del VIH no se asociaron con el suicidio. Se observó una interacción multiplicativa entre un diagnóstico de VIH y un intento de suicidio previo. Entre las PWH, una alta prevalencia de trastornos psiquiátricos, por consumo de sustancias y multimorbilidad contribuyen al riesgo de suicidio.
Keywords: Veterans Aging Cohort Study, Suicide, Suicide mortality, HIV
Introduction
Persons living with HIV (PWH) are at elevated risk of suicide compared to the general population [1–4]. Advances in treatment and improved tolerability of antiretroviral therapy (ART) transitioned HIV from a terminal illness to a chronic disease. Nevertheless, the rate of suicide among PWH is two to ten times higher than the rate in the general population [1, 3–6]. Furthermore, the rate of suicide remains elevated even compared to individuals with other chronic conditions [1–5, 7]. Recent longitudinal studies have explored risk factors that predisposed or increased vulnerability to suicide among PWH. Many of these factors were similar to those found in the general population, including a previous suicide attempt [1, 7], psychiatric illness [2, 7, 8], alcohol use [7], and drug use [1, 7, 9]. Other studies have examined the role of HIV-specific risk factors (CD4 count, HIV-1 RNA, ART, use of efavirenz (EFV), and adherence to ART), but their associations with suicide remain unclear [1, 5, 7, 9, 10]. Elevated rates of suicide are likely due to several interconnected factors. However, due to relatively few suicides and consequent low statistical power, past studies have largely examined these factors in isolation or in combination with a limited number of risk factors.
Our study aimed to address these limitations, and we examined a broader range of risk factors for suicide among PWH and uninfected comparators from the Veterans Aging Cohort Study (VACS) [11]. The United States Department of Veterans Affairs (VA), the largest provider of HIV-related services in the United States, provides care to over 30,000 PWH. We used a nested case-control design with data from 2006 to 2015 and examined risk factors for suicide among people living with and without HIV (PWoH). We explored associations among demographics, mental health comorbidities, substance use, prescribed medications, service utilization, smoking status, pain, and suicide. We ascertained whether HIV-specific risk factors (e.g., CD4 count, HIV-1 RNA, ART, and exposure to EFV) were associated with suicide. Lastly, we examined whether HIV modified the association between specific risk factors and suicide.
Methods
Data source
We used data from VACS [11], a longitudinal, prospective, observational cohort of Veterans who received services from the VA. VACS includes all Veterans diagnosed with HIV who receive care within the VA. VACS includes EHR data, including inpatient and outpatient medical records, administrative codes, laboratory values, and pharmacy data. Cause of death was ascertained from the National Death Index (NDI). VACS has been used extensively in peer-reviewed studies to examine mental health [12–14], substance use [15–18], and physical health outcomes among Veterans [19, 20].
Study design and sample
For this study, we used a nested case-control design. Cases were defined as individuals who died from suicide from 2006 to 2015, regardless of HIV status. We identified suicide deaths from the NDI using the International Classification of Disease (ICD) 10th revision cause of death codes (X60-84, U03.0). Additional description of ICD-10 cause of death codes is provided in the appendix. Ten controls were selected for each case from a representative sample within VACS using incident density sampling with replacement [21, 22]. Incident density sampling required that controls were alive at the time of the case death. The index date for the matched control was the case date of death. For each case, controls were matched based on HIV status, date of birth ± one year, race, and sex. To ensure the inclusion of individuals who accessed VA services, the sample was limited to individuals who had at least two encounters within the VA system occurring a year or more apart. Baseline characteristics were calculated from each individual’s first encounter with the VA during the study period.
Risk factors
We included the following risk factors in our models: a previous suicide attempt; a diagnosis of affective disorder (major depressive disorder, bipolar disorder, and anxiety disorder); schizophrenia; PTSD; alcohol use disorder; and drug use disorder identified using ICD-9 codes. VA medication fill–refill data was used to identify psychotropic medication (antidepressants, antipsychotics, anticonvulsants, gabapentin, and lithium); benzodiazepines; and opioids prescribed within 90 days prior to the index date. We included the VACS Index 2.0, a prognostic indicator of morbidity and mortality [19]. VACS Index 2.0 included: age; CD4 count; HIV-1 RNA; hemoglobin; Hepatitis-C viral infection (HCV); body mass index (BMI); white blood cell count (WBC); albumin; Fibrosis-4 (Fib-4), an indicator of liver fibrosis [23]; and estimated glomerular filtration rate (eGFR), an indicator of renal function [9]. FIB-4 was calculated from AST, ALT, platelets, and age [23]. The estimated glomerular filtration rate was computed using the Chronic Kidney Disease Epidemiology Collaboration equation using serum creatinine, age, sex, and race [24]. We identified any inpatient psychiatric hospitalization and outpatient mental health visit which occurred in the previous 365 days of the index date. HCV was a composite variable. Individuals were identified as positive if they had one of the following: ICD-9 codes for HCV, positive antibody tests, or a detectable plasma HCV RNA. We also included pain intensity using a numerical rating scale score (NRS; 0–10), and smoking status collected during outpatient visits as part of Health Factors data. HIV-specific risk factors included a current prescription for ART (active prescription at the index date), past exposure to EFV from VA fill–refill data, and most recent CD4 count and HIV-1 RNA from laboratory files.
Data analysis
All analyses were completed using SAS 9.4 [25], with statistical significance defined as a 2-tailed p-value of less than 0.05. The distribution of risk factors by case-control status was examined using a χ2 statistic for categorical variables and a t-test for continuous variables (Table 1). We assumed missingness as being at random. We performed multiple imputation using PROC MI to address missingness using ten imputations with fully conditional specifications [26, 27] and merged the imputations using PROC MI ANALYZE. We used conditional logistic regression to account for matched data. As we used incident density sampling for controls, the odds ratios (ORs) were an unbiased estimate of the incidence rate ratios (IRR) [28]. For the analysis, we presented the adjusted incidence rate ratio (aIRR) with the associated 95% confidence intervals (Table 2). We stratified by HIV status and constructed identical models to explore differences in the associations between risk factors and suicide [Stratified Model (HIV+) and Stratified Model (HIV-)]. Among PWH, we further included HIV-specific risk factors to examine their association with suicide (Appendix 1). We then examined HIV as a predictor by fitting the model to the entire sample (people with and without HIV). We explored interaction terms between HIV and risk factors that differed between people with and without HIV. We assessed for significance of the interaction terms, including a previous suicide attempt, alcohol, and drug use disorder, benzodiazepine use, opioid use, and mental health service utilization. The included terms were selected a priori based on previous literature identifying their role in the potentiation of adverse outcomes among PWH [16, 17, 29–33]. We assessed for improvement of model fit based on the likelihood ratio test.
Table 1:
Distribution of matched sample of Veterans who died by suicide compared to SIUD
| Suicide | Self-injurious unnatural death (SIUD) | |||||
|---|---|---|---|---|---|---|
| N=1,328 | N=5,036 | |||||
| Parameter | Control | Case | p-value | Control | Case | p-value |
| N=1,206 | N=122 | N=4,575 | N=461 | |||
| Demographics | ||||||
| Male | 1202 (99·6%) | 121 (99·1%) | * | 4463 (97·6%) | 449 (97·4%) | * |
| Age | 52·5 (11·1) | 52·6 (11·4) | * | 53·7 (9·5) | 53·8 (9·7) | * |
| Race | ||||||
| White | 81 (66·3%) | * | 2222 (48·6%) | 224 (48·6%) | * | |
| Black | 222 (18·4%) | 22 (18·0%) | 1731 (37·8%) | 173 (37·5%) | ||
| Other | 183 (15·1%) | 19 (15·5%) | 622 (13·6%) | 64 (13·9%) | ||
| Region | ||||||
| South | 592 (49·0%) | 59 (48·3%) | 0·003 | 2277 (49·8%) | 197 (42·7%) | 0·001 |
| Northeast | 189 (15·6%) | 8 (6·5%) | 725 (15·9%) | 102 (22·1%) | ||
| Midwest | 143 (11·8%) | 11 (9·0%) | 559 (12·2%) | 50 (10·9%) | ||
| West | 282 (23·3%) | 44 (36·0%) | 1014 (22·2%) | 112 (24·3%) | ||
| Mechanism of Death | ||||||
| Firearm | 40 (32·8%) | 42 (9·1%) | ||||
| Poisoning | 34 (27·9%) | 305 (66·2%) | ||||
| Suffocation | 37 (30·3%) | 55 (11·9%) | ||||
| Cut | 5 (4·1%) | 5 (1·1%) | ||||
| Other Method | 6 (4·9%) | 54 (11·7%) | ||||
| Mental Health | ||||||
| Previous Suicide Attempt | 34 (2·8%) | 21 (17·2%) | <·001 | 181 (4·0%) | 72 (15·6%) | <·001 |
| Major Depressive Disorder | 236 (19·5%) | 46 (37·7%) | <·001 | 949 (20·7%) | 183 (39·7%) | <·001 |
| Bipolar Disorder | 118 (9·7%) | 37 (30·3%) | <·001 | 478 (10·5%) | 123 (26·7%) | <·001 |
| Anxiety Disorder | 172 (14·2%) | 36 (29·5%) | <·001 | 725 (15·9%) | 157 (34·1%) | <·001 |
| Schizophrenia | 24 (1·9%) | 8 (6·5%) | 0·002 | 128 (2·8%) | 27 (5·9%) | <·001 |
| PTSD | 112 (9·2%) | 21 (17·2%) | 0·005 | 541 (11·8%) | 107 (23·2%) | <·001 |
| Alcohol Use Disorder | 238 (19·7%) | 48 (39·3%) | <·001 | 1157 (25·3%) | 252 (54·7%) | <·001 |
| Drug Use Disorder | 244 (20·2%) | 46 (37·7%) | <·001 | 1189 (26·0%) | 272 (59·0%) | <·001 |
| Service Utilization Prior 365 Days | ||||||
| Inpatient Mental Health | 25 (2·0%) | 18 (14·7%) | <·001 | 154 (3·4%) | 73 (15·8%) | <·001 |
| Outpatient Mental Health | 281 (23·3%) | 50 (40·9%) | <·001 | 1295 (28·3%) | 214 (46·4%) | <·001 |
| Medication Prior 90 Days | ||||||
| Antidepressants | 253 (20·9%) | 50 (40·9%) | <·001 | 1053 (23·0%) | 174 (37·7%) | <·001 |
| Antipsychotics | 63 (5·2%) | 27 (22·1%) | <·001 | 314 (6·9%) | 84 (18·2%) | <·001 |
| Other Mood Stabilizers | 44 (3·6%) | 8 (6·5%) | 0·114 | 160 (3·5%) | 25 (5·4%) | 0·036 |
| Gabapentin | 68 (5·6%) | 13 (10·6%) | 0·027 | 323 (7·1%) | 62 (13·5%) | <·001 |
| Benzodiazepines | 104 (8·6%) | 33 (27·0%) | <·001 | 380 (8·3%) | 91 (19·7%) | <·001 |
| Opioid | 175 (14·5%) | 33 (27·0%) | <·001 | 819 (17·9%) | 139 (30·2%) | <·001 |
| Other Comorbidities | ||||||
| Hep-C | 202 (16·7%) | 32 (26·2%) | 0·009 | 1035 (22·6%) | 221 (47·9%) | <·001 |
| Physiologic frailty | ||||||
| VACS Index 2.0, mean (SD) | 48·9 (17·3) | 51·0 (16·9) | 0·260 | 50·4 (17·6) | 61·2 (21·9) | <·001 |
| Smoking Status | ||||||
| Never Smoked | 336 (29·8%) | 28 (23·3%) | 0·263 | 1,201 (26·3%) | 70 (15·2%) | <·001 |
| Current Smoker | 603 (53·5%) | 73 (60·8%) | 2,477 (54·1%) | 335 (72·7%) | ||
| Former Smoker | 187 (16·6%) | 19 (15·8%) | 620 (13·5%) | 45 (9·8%) | ||
| Most Recent Pain Score | ||||||
| 0 No Pain | 645 (67·3%) | 67 (62·6%) | 0·344 | 2,361 (63·8%) | 207 (52·3%) | <·001 |
| 1–3 Low | 93 (9·7%) | 13 (12·2%) | 372 (10·1%) | 44 (11·1%) | ||
| 4–6 Moderate | 97 (10·2%) | 8 (7·5%) | 442 (11·9%) | 43 (10·9%) | ||
| 7–10 High | 123 (12·8%) | 19 (17·7%) | 527 (14·2%) | 102 (25·7%) | ||
| HIV Specific | ||||||
| Undetectable Viral Load† | 477 (54·7%) | 55 (48·3%) | <·001 | 2,365 (51·7%) | 183 (39·7%) | <·001 |
| CD4 Count | 507·7 (294·5) | 557·6 (345·4) | 0·122 | 518·4 (306·7) | 468·0 (313·8) | 0·002 |
| Current ART | 699 (57·9%) | 79 (64·7%) | 0·147 | 2,674 (58·4%) | 275 (59·6%) | 0·617 |
| Ever on EFV | 386 (32·0%) | 37 (30·3%) | 0·704 | 1,501 (32·8%) | 158 (34·3%) | 0·524 |
Undetectable HIV-1 RNA at 50copies/mL or lower, ART = Antiretroviral therapy, EFV = Efavirenz,
matched variables
Table 2:
Characteristics of PWH who died by suicide, unintentional or undetermined causes·
| Parameter | Suicide | Unintentional | Undetermined | |
|---|---|---|---|---|
| N= 122 | N= 306 | N=33 | P-value | |
| Demographics | ||||
| Male | 121 (99·18%) | 298 (97·39%) | 30 (90·91%) | 0·03 |
| Age | 52·55 (11·40) | 54·45 (9·08) | 52·27 (7·34) | 0·12 |
| Race | ||||
| White | 81 (66·39%) | 127 (41·50%) | 16 (48·48%) | <·001 |
| Black | 22 (18·03%) | 137 (44·77%) | 14 (42·42%) | |
| Other | 19 (15·57%) | 42 (13·73%) | 3 (9·09%) | |
| Region | ||||
| South | 59 (48·36%) | 123 (40·20%) | 15 (45·45%) | <·001 |
| Northeast | 8 (6·56%) | 87 (28·43%) | 7 (21·21%) | |
| Midwest | 11 (9·02%) | 36 (11·76%) | 3 (9·09%) | |
| West | 44 (36·07%) | 60 (19·61%) | 8 (24·24%) | |
| Mechanism of Death | ||||
| Firearm | 40 (32·8%) | 2 (0·65%) | 0 (0·0%) | <·001 |
| Poisoning | 34 (27·9%) | 250 (81·7%) | 21 (63·6%) | |
| Suffocation | 37 (30·3%) | 18 (5·9%) | 0 (0·0%) | |
| Cut | 5 (4·1%) | 0 (0·0%) | 0 (0·0%) | |
| Other Method | 6 (4·9%) | 36 (11·8%) | 12 (36·4%) | |
| Mental Health | ||||
| Previous Suicide Attempt | 21 (17·2%) | 45 (14·7%) | 6 (18·2%) | 0·743 |
| Major Depressive Disorder | 46 (37·7%) | 119 (38·9%) | 18 (54·6%) | 0·19 |
| Bipolar Disorder | 37 (30·3%) | 76 (24·8%) | 10 (30·3%) | 0·453 |
| Anxiety Disorder | 36 (29·5%) | 109 (35·6%) | 12 (36·4%) | 0·464 |
| Schizophrenia | 8 (6·6%) | 15 (4·9%) | 4 (12·1%) | 0·227 |
| PTSD | 21 (17·2%) | 77 (25·2%) | 9 (27·3%) | 0·181 |
| Alcohol Use Disorder | 48 (39·3%) | 177 (57·8%) | 27 (81·8%) | <·001 |
| Drug Use Disorder | 46 (37·7%) | 200 (65·4%) | 26 (78·8%) | <·001 |
| Service Utilization Prior 365 Days | ||||
| Inpatient Mental Health | 18 (14·8%) | 49 (16·0%) | 6 (18·2%) | 0·882 |
| Outpatient Mental Health | 50 (41·0%) | 146 (47·7%) | 18 (54·6%) | 0·282 |
| Medication Prior 90 Days | ||||
| Antidepressants | 50 (41·0%) | 113 (36·9%) | 11 (33·3%) | 0·636 |
| Antipsychotics | 27 (22·1%) | 51 (16·7%) | 6 (18·2%) | 0·191 |
| Other Mood Stabilizers | 8 (6·6%) | 15 (4·9%) | 2 (6·1%) | 0·781 |
| Gabapentin | 13 (10·7%) | 42 (13·7%) | 7 (21·2%) | 0·28 |
| Benzodiazepine | 33 (27·1%) | 54 (17·7%) | 4 (12·1%) | 0·046 |
| Opioid | 33 (27·1%) | 93 (31·0%) | 13 (39·4%) | 0·386 |
| Other Comorbidities | ||||
| Hep-C | 32 (26·2%) | 169 (55·2%) | 20 (60·6%) | <·001 |
| Physiologic frailty | ||||
| VACS Index 2.0, mean (SD) | 51·0 (16·9) | 65·2 (22·6) | 66·1 (19·9) | <·001 |
| Smoking Status | ||||
| Never Smoked | 28 (22·9%) | 42 (13·7%) | 1 (3·0%) | <·001 |
| Current Smoker | 75 (61·5%) | 237 (77·5%) | 32 (97·0%) | |
| Former Smoker | 19 (15·6%) | 27 (8·8%) | 0 (0·0%) | |
| Most Recent Pain Score | ||||
| 0 No Pain | 67 (62·6%) | 127 (48·7%) | 13 (46·4%) | 0·204 |
| 1–3 Low | 13 (12·1%) | 28 (10·7%) | 3 (10·7%) | |
| 4–6 Moderate | 8 (7·5%) | 31 (11·9%) | 4 (14·3%) | |
| 7–10 High | 19 (17·8%) | 75 (28·7%) | 8 (28·6%) | |
| HIV Specific | ||||
| Undetectable Viral Load† | 58 (47·5%) | 118 (38·6%) | 9 (27·3%) | 0·068 |
| CD4 Count | 539 (±322·8) | 466·4 (±288·7) | 453·1 (±298·3) | 0·064 |
| Current ART | 79 (64·8%) | 117 (57·8%) | 19 (57·6%) | 0·417 |
| Ever on EFV | 37 (30·3%) | 110 (36·0%) | 11 (33·3%) | 0·539 |
† Undetectable HIV-1 RNA at 50copies/mL or lower, ART = Antiretroviral therapy, EFV = Efavirenz,
Results
Cases and controls
Of the 4,383 individuals included in the study, 400 died by suicide. Of these, 122 were PWH. Individuals were followed for an average of 6.9 (± 6.4) years. The sample was predominantly male (98.9%) with an average age of 54.2 (± 10.7) years at the time of death or matched date. Detailed characteristics of the study population are further described in Table 1.
Suicide decedents
Among suicide decedents, PWH died at a mean age of 52.6 (± 11.4) years compared to 54.9 (± 10.4) years among PWoH (p = 0.04). The cause of death among PWH included other mechanisms (e.g., strangulation, cutting) (39.3%), firearms (32.8%), and poisoning (27.9%). Among PwoH, a greater percentage died by firearms (53.6%), followed by other mechanisms (27.3%) then poisoning (19.1%) (p < 0.001).
Decedents with HIV compared to decedents without HIV, were more likely to have had a prior suicide attempt (17.2% vs. 9.7%, respectively, p = 0.03) and a diagnosis of HCV (26.2% vs. 14.3%, respectively, p = 0.005). A previous suicide attempt was less frequent among individuals who died by firearms compared to those who died by poisoning or other mechanisms of injury (5.8% vs. 17.2% vs. 17.7%, p = 0.002). Decedents with HIV, compared to those without, also reported a lower frequency of moderate levels of pain (5–7 on the NSC) (6.5% vs. 25.8%, p < 0.001) using a 0–10 pain intensity NRS. No difference was observed for severe pain (8–10) (15.0% vs. 13.4%, p = 0.705).
Among decedents, most other characteristics did not differ by HIV status (Table 1). These include the number of individuals diagnosed with affective disorders, schizophrenia, PTSD, drug use disorder, and alcohol use disorder. A prescription for psychotropic medications, benzodiazepines, or opioids in the previous 90 days, as well as outpatient and inpatient psychiatric services, were also comparable.
HIV specific factors
Among Veterans with HIV, the distribution of HIV-specific risk factors (recent HIV-1 RNA and CD4 count, current ART, and previous prescription of Efavirenz) did not differ between cases and controls. Additionally, these factors were not significantly associated with suicide in bivariate or multivariable analysis (appendix).
Multivariable analysis
Among PWH, a previous suicide attempt (aIRR = 2.82, 95% CI: 1.36–5.83 ), diagnosis of an affective disorder (aIRR = 1.76, 95% CI: 1.06–2.91), prescription for benzodiazepines in the previous 90 days (aIRR = 3.05, 95% CI: 1.78–5.22), and inpatient psychiatric hospitalization in the previous year (aIRR = 3.58, 95% CI: 1.59–8.04) were associated with suicide (Table 2).
Among PWoH (Table 2), risk factors for suicide also included a diagnosis of an affective disorder (aIRR = 2.66, 95% CI: 1.94–3.63), a prescription for benzodiazepines in the previous 90 days (aIRR = 1.82, 95% CI: 1.28–2.61 ), an outpatient mental health services in the previous year (aIRR = 1.77, 95% CI: 1.3–2.42), or current smoking status compared to those who never smoked (aIRR = 1.53, 95% CI: 1.07–2.17).
Combined sample
In the combined model that included those with and without HIV, risk factors for suicide included a previous diagnosis of an affective disorder (aIRR = 2.36, 95% CI: 1.81–3.06), alcohol use disorder (aIRR = 1.42, 95% CI: 1.06–1.92), prescription for benzodiazepines in the previous 90 days (aIRR = 2.12, 95% CI: 1.58–2.85), outpatient (aIRR = 1.54, 95% CI: 1.18–1.99), inpatient mental health services in the previous year (aIRR = 1.86, 95% CI: 1.20–2.89), and current smoking status (aIRR = 1.38, 95% CI: 1.04–1.85). Additionally, a significant multiplicative interaction (aIRR = 2.31, 95% CI: 1.04–5.1) was identified between HIV (aIRR = 1.08, 95% CI: 0.07–15.82) and previous suicide attempts (aIRR = 1.43, 95% CI: 0.87–2.37) with the presence of HIV increasing the risk of suicide above that observed by a previous suicide independently. No significant interaction was observed between HIV and the other tested interaction terms: alcohol and drug use disorder, benzodiazepine or opioid use in the past 90 days, and inpatient or outpatient hospitalization in the previous 365 days.
Discussion
Most of the factors associated with suicide were similar among PWH and PWoH; these included affective disorders, recent use of benzodiazepines, and mental health treatment. A previous suicide attempt was associated with suicide among PWH. Among PWoH, an increased risk of suicide was also observed; however, the association did not reach significance.
The lethality of the method used in a suicide attempt could explain the interaction between HIV status and a previous suicide attempt. As estimated by Connors et al. (2019), firearms were considerably more lethal (Case Fatality Rate (CFR), 89.6%) than other methods, such as poisoning (CFR,1.9%) or cutting (CFR, 0.7%) [34]. Furthermore, upwards of 79% of individuals die after a first attempt [35], with firearms the predominant method [35]. In our study, we found that PWoH were far more likely to die from firearm-related self-harm compared to PWH, who were more likely to die from poisoning or other mechanisms. We also found that a previous suicide attempt was less frequent among individuals who died by firearms than those who died by poisoning or other mechanisms of injury. Thus, a previous suicide attempt was likely confounded by the lethality of the method selected.
A previous psychiatric hospitalization was associated with suicide among PWH. Any mental health treatment—either inpatient or outpatient—was associated with suicide in the combined sample. As psychiatric hospitalization is a well-established risk factor for suicide [36–39], low power may have contributed to a lack of association in the sample stratified to PWoH. A recent meta-analysis by Chung et al. (2017) estimated the overall rate of suicide within three and twelve months post-discharge to be 1,132 and 654 per 100,000 person-years, respectively [40]. Higher-risk subpopulations, such as those with psychiatric admissions for suicidal thoughts or behaviors, were at even greater risk [40]. It is plausible that PWH represent an additional population at increased risk of post-discharge suicide. Future studies should explore post-discharge suicide risk among PWH, as more targeted interventions may be necessary.
In accordance with previous literature, alcohol use disorder was positively associated with an increased risk of suicide in the combined sample [7, 41]. Alcohol use can affect the risk of suicide through several pathways. These include dysfunction in neurobiology and neurocognitive impairment [42, 43], exacerbation of psychiatric disorders [44], and familial, social and environmental consequences [44]. Among PWH, the adverse effects of acute and chronic alcohol use occur at lower consumption levels than in individuals without HIV [17]. Specifically, among PWH, alcohol use has been associated with accelerated disease progression and worsening physical and psychiatric symptoms [17, 18] relative to individuals without HIV.
The use of benzodiazepines in the previous 90 days was strongly associated with suicide in both the stratified and combined sample. However, the relationship and directionality between benzodiazepines and suicide is challenging to untangle. Many of the indicated uses of benzodiazepines (e.g., acute anxiety, alcohol withdrawal, insomnia, and used in psychiatric emergencies) are also related to suicide [45]. Additionally, characteristics such as age, psychiatric comorbidity, substance use, and chronic versus acute benzodiazepine use further alter their association with suicide [29]. As with alcohol, PWH appear vulnerable to the adverse effects of benzodiazepines, particularly neurocognitive impairment [30, 46]. In our combined sample, the interaction between HIV and benzodiazepine use in the previous 90 days was not significant; however, we may have been underpowered. Given the potential vulnerability of PWH to the harmful effects of benzodiazepines, the relationship between benzodiazepines and suicide among PWH merits further investigation.
HIV specific factors
In our study, HIV-specific risk factors (CD4 cell count, HIV-1 RNA, current ART use, and exposure to EFV) were not associated with suicide. These findings were consistent with a recent study using the French cohort, Dat’AIDS, which found no association between suicide and HIV-specific risk factors [7]. Recent studies include samples from more current ART eras, and these individuals demonstrate consistently high CD4 counts and low HIV-1 RNA [1, 7]. This lack of variation in recent study populations makes it difficult to assess the impact of CD4 count and HIV-1 RNA. Exposure to EFV was also not associated with suicide, consistent with the current evidence [47, 48]. However, knowledge of adverse psychiatric side effects of EFV plausibly limited the provision of EFV to individuals with known psychiatric illness, thus diminishing our ability to assess an association.
Despite a lack of significance observed between HIV-specific risk factors and suicide, PWH remain at higher risk of suicide than uninfected comparators [1–4, 7]. Higher rates of medical comorbidities and substance use, as well as environmental stressors, social sigma and bias experienced by PWH are likely potent drivers of suicide risk.
Clinical implications and future research
Consistent with recent literature, we found that risk factors for suicide among PWH in the current treatment era were similar to those in the general population. However, despite similarities in risk factors between groups, clinicians should remain attuned to suicide risk among this high-risk group. Vulnerability to suicide is influenced by a constellation of risk factors, and the intersection of these risk factors with periods of heightened stress can elevate the risk of suicide. Additionally, PWH have a higher prevalence of psychiatric illness, substance use, and multimorbidity than PWoH. They are also disproportionately affected by discrimination, trauma and economic hardships, further potentiating risk [49, 50]. These factors point to the heightened risk for suicide in this population, and providers need to be cognizant of the ongoing risk even though HIV has become a chronic – rather than a lethal—diagnosis. Preventative treatments will need to address both the psychological needs as well as the social and economic disparities and complexities confronting this community. PWH will continue to benefit from the integration of mental health services allowing for a comprehensive approach to managing their complexity of care. Additional research is needed to further understand the risk factors that drive the elevated risk of suicide among PWH and to continue tailoring suicide prevention programs for this community [1, 3, 4].
Clinicians should also exercise caution when prescribing benzodiazepines and regularly reassess both their risks and benefits. Clinicians should also be aware that PWH may be more vulnerable to the adverse effects of benzodiazepines compared to PWoH due to multimorbidity, polypharmacy, altered medication metabolism, and physiologic frailty [30, 46]. In addition, the indicated uses of benzodiazepines are often associated with suicide (e.g. acute anxiety, alcohol withdrawal, insomnia, psychiatric emergencies) [45], and thus complicate the determination of the directionality of associations [45]. Thus, future research is needed to further understand this association and provide clearer guidance on best practices.
Strengths and limitations
The study had several strengths. First, we used a large longitudinal sample of PWH who received care from the largest provider of HIV services in the United States. Longitudinal records included inpatient and outpatient care for medical and psychiatric conditions, including substance use disorders. In past studies, high-risk subgroups – such as PWH—were frequently incorporated into larger at-risk populations due to sample size constraints. Consequently, risk factors associated with suicide among smaller at-risk subgroups were often underestimated or not considered in those studies. While sample size remained a limitation in our stratified analyses, we were still able to adjust for a large number of risk factors. Secondly, we could compare our sample of PWH to a sample of matched, uninfected individuals, all of whom accessed service at the VA.
Our study had some limitations. First, our sample was predominantly male, with only five female suicide decedents. The low number of females was reflective of using data from the VA. Yet, we know that women with HIV confront countless forms of victimization, violence, and discrimination [51, 52]. While women are generally at lower risk for suicide, this may not hold among women with HIV [54]. A second limitation was that we were unable to include important risk factors such as sexual orientation and measures of discrimination and stigma [49, 50]. Third, we likely underestimated the number of suicide deaths in our study. Classification of death as a suicide requires knowledge of intentionality, yet determination of intent to self-harm is challenging and at times impossible to ascertain. This is particularly true among ambiguous and less violent methods, such as poisoning [54, 55].
Conclusion
We found that most of the factors associated with suicide were similar among PWH and PWoH; which included a diagnosis of affective disorders, alcohol use disorder, recent use of benzodiazepines, and inpatient or outpatient mental health treatment. Despite a larger sample size, we did not observe an association with CD4 cell count or HIV-1 RN. We did find a significant interaction between HIV and a previous suicide attempt, controlling for other factors. Whether these associations represent modifiable factors remains to be seen.
Supplementary Material
Table 3:
Multivariable analysis of risk factors associated with dying by suicide compared to dying by self-injurious unnatural death
| Suicide | Self-injurious unnatural death (SIUD) | |||
|---|---|---|---|---|
| Controls = 1,206 | Controls = 4,575 | |||
| Cases = 122 | Cases = 461 | |||
| Parameter | aIRR (95% CI) | p-value | aIRR (95% CI) | p-value |
| Demographics | ||||
| Age | 1·2 (0·71–2·02) | * | 1·11 (0·84–1·45) | * |
| Region | ||||
| Ref· South | ||||
| Northeast | 0·32 (0·14–0·71) | 0·005 | 1·22 (0·91–1·62) | 0·177 |
| Midwest | 0·79 (0·38–1·63) | 0·523 | 0·95 (0·67–1·36) | 0·8 |
| West | 1·6 (1·01–2·55) | 0·049 | 1·46 (1·11–1·93) | 0·007 |
| Mental Health | ||||
| Previous suicide attempt | 2·8 (1·35–5·8) | 0·006 | 1·55 (1·08–2·23) | 0·019 |
| Affective disorders | 1·76 (1·06–2·92) | 0·028 | 1·51 (1·18–1·94) | 0·001 |
| Schizophrenia | 1·61 (0·58–4·47) | 0·363 | 0·65 (0·39–1·07) | 0·087 |
| PTSD | 1·1 (0·57–2·11) | 0·783 | 1·02 (0·77–1·36) | 0·876 |
| Alcohol use disorder | 1·53 (0·85–2·75) | 0·158 | 1·46 (1·1–1·93) | 0·008 |
| Drug use disorder | 1·03 (0·56–1·9) | 0·912 | 1·73 (1·29–2·33) | <0·001 |
| Medication Prior 90 Days | ||||
| Psychotropics | 1·08 (0·65–1·77) | 0·774 | 1·22 (0·95–1·56) | 0·124 |
| Benzodiazepine | 3·07 (1·79–5·28) | <0·001 | 2·09 (1·54–2·85) | <0·001 |
| Opioid | 1·31 (0·77–2·22) | 0·319 | 1·17 (0·89–1·52) | 0·255 |
| Service Utilization Prior 365 Days | ||||
| Inpatient mental health | 3·58 (1·59–8·06) | 0·002 | 2·06 (1·41–2·99) | <0·001 |
| Outpatient mental health | 0·98 (0·59–1·62) | 0·933 | 1·12 (0·88–1·43) | 0·365 |
| Other Comorbidities | ||||
| Hep-C | 1·1 (0·63–1·92) | 0·748 | 1·52 (1·17–1·97) | 0·002 |
| Physiologic frailty | ||||
| VACS Index 2·0 per 5 points | 1·07 (0·99–1·16) | 0·074 | 1·16 (1·12–1·2) | <0·001 |
| Smoking Status | ||||
| Ref· Nonsmoker | ||||
| Current smoker | 1·09 (0·65–1·83) | 0·739 | 1·42 (1·05–1·92) | 0·023 |
| Former smoker | 1·2 (0·62–2·3) | 0·586 | 1·14 (0·75–1·72) | 0·54 |
| Most Recent Pain Score | ||||
| Ref· No Pain 0 | ||||
| 1–3 low | 0·86 (0·42–1·79) | 0·696 | 1·04 (0·71–1·51) | 0·843 |
| 4–6 moderate | 0·64 (0·27–1·51) | 0·306 | 0·83 (0·56–1·22) | 0·34 |
| 7–10 high | 1·03 (0·54–1·97) | 0·931 | 1·45 (1·07–1·98) | 0·018 |
matched variables
Acknowledgements
This work was supported by the Robert Wood Johnson, Future of Nursing Scholars program; Yale Center for Clinical Investigation Multidisciplinary Pre-Doctoral Training Program in Translational Research; and NIH National Institute on Alcohol Abuse and Alcoholism U24 AA020794, U01 AA020790, U24 AA022001, U10 AA013566.
Funding
Robert Wood Johnson, Future of Nursing Scholars; Yale Center for Clinical Investigation Multidisciplinary Pre-Doctoral Training Program in Translational Research; NIH National Institute on Alcohol Abuse and Alcoholism U24 AA020794, U01 AA020790, U24 AA022001, U10 AA013566-completed and in kind by the US Department of Veterans Affairs.
Footnotes
Declarations
Ethics Approval This study was performed in line with the principles of the Declaration of Helsinki. IRB approval was granted from the institutional review boards of the coordinating center at Yale University, New Haven, CT, the Veterans Affairs (VA) Connecticut Healthcare System, West Haven, CT, and from the participating clinical sites. (Yale HIC 0309025943; VA IRB AJ0001; IRBNet 1583210)
Disclaimer The views and opinions expressed in this manuscript are those of the authors and do not necessarily represent those of the Department of Veterans Affairs or the United States government.
Supplementary Information The online version contains supplementary material available at https://doi.org/10.1007/s10461–023-04164–3.
Conflict of interest No conflicts of interest.
Data Availability (data transparency)
While data are not publicly available, reasonable requests will be considered by the corresponding author.
Code Availability (software application or custom code)
All programing was conducted in SAS 9.4 and will be made available by the corresponding author if requested.
References
- 1.Ruffieux Y, Lemsalu L, Aebi-Popp K, Calmy A, Cavassini M, Fux CA, et al. Mortality from suicide among people living with HIV and the general swiss population: 1988–2017. J Int AIDS Soc. 2019;22(8):e25339. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Jia CX, Mehlum L, Qin P. AIDS/HIV infection, comorbid psychiatric illness, and risk for subsequent suicide: a nationwide register linkage study. J Clin Psychiatry. 2012;73(10):1315–21. [DOI] [PubMed] [Google Scholar]
- 3.Aldaz P, Moreno-Iribas C, Egues N, Irisarri F, Floristan Y, Sola-Boneta J, et al. Mortality by causes in HIV-infected adults: comparison with the general population. BMC Public Health. 2011;11:300. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Croxford S, Kitching A, Desai S, Kall M, Edelstein M, Skingsley A, et al. Mortality and causes of death in people diagnsoed with HIV in the era of highly active antiretroviral therapy compared with the general population: an analysis of a national observational cohort. The Lancet Public Health. 2017;2(1):e35–e47. [DOI] [PubMed] [Google Scholar]
- 5.Keiser O, Spoerri A, Brinkhof MW, Hasse B, Gayet-Ageron A, Tissot F, et al. Suicide in HIV-infected individuals and the general population in Switzerland, 1988–2008. Am J Psychiatry. 2010;167(2):143–50. [DOI] [PubMed] [Google Scholar]
- 6.Nishijima T, Inaba Y, Kawasaki Y, Tsukada K, Teruya K, Kikuchi Y, et al. Mortality and causes of death in people living with HIV in the era of combination antiretroviral therapy compared with the general population in Japan. AIDS. 2020;34(6):913–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Hentzien M, Cabie A, Pugliese P, Billaud E, Poizot-Martin I, Duvivier C, et al. Factors associated with deaths from suicide in a french nationwide HIV-infected cohort. HIV Med. 2018;19:551–8. [DOI] [PubMed] [Google Scholar]
- 8.McManus H, Petoumenos K, Franic T, Kelly MD, Watson J, O’Connor CC, et al. Determinants of suicide and accidental or violent death in the australian HIV Observational Database. PLoS ONE [Electronic Resource]. 2014;9(2):e89089. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Gurm J, Samji H, Nophal A, Ding E, Strehlau V, Zhu J, et al. Suicide mortality among people accessing highly active antiretroviral therapy for HIV/AIDS in British Columbia: a retrospective analysis. CMAJ open. 2015;3(2):E140–E8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Antiretroviral Therapy Cohort Collaboration. Causes of death in HIV-1-infected patients treated with antiretroviral therapy, 1996–2006: collaborative analysis of 13 HIV cohort studies. Clin Infect Dis. 2010;50(10):1387–96. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Justice AC, Dombrowski E, Conigliaro J, Fultz SL, Gibson D, Madenwald T, et al. Veterans Aging Cohort Study (VACS): overview and description. Med Care. 2006;44(8 Suppl 2):13–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Justice AC, McGinnis KA, Atkinson JH, Heaton RK, Young C, Sadek J, et al. Psychiatric and neurocognitive disorders among HIV-positive and negative veterans in care: Veterans Aging Cohort five-site study. AIDS. 2004;18(Suppl 1):49–59. [PubMed] [Google Scholar]
- 13.So-Armah K, Gupta SK, Kundu S, Stewart JC, Goulet JL, Butt AA, et al. Depression and all-cause mortality risk in HIV-infected and HIV-uninfected US veterans: a cohort study. HIV Med. 2019;20(5):317–29. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Hooshyar D, Goulet J, Chwastiak L, Crystal S, Gibert C, Mattocks K, et al. Time to depression treatment in primary care among HIV-infected and uninfected veterans. J Gen Intern Med. 2010;25(7):656–62. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Womack JA, Murphy TE, Rentsch CT, Tate JP, Bathulapalli H, Smith AC, et al. Polypharmacy, Hazardous Alcohol and Illicit Substance Use, and Serious Falls among PLWH and Uninfected Comparators. J Acquir Immune Defic Syndr. 2019;82(3):305–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Becker WC, Gordon KS, Edelman EJ, Goulet JL, Kerns RD, Marshall BDL et al. Are we missing opioid-related deaths among people with HIV? Drug Alcohol Depend. 2020;212:Epub 2020 Apr 25. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Justice AC, McGinnis KA, Tate JP, Braithwaite RS, Bryant KJ, Cook RL, et al. Risk of mortality and physiologic injury evident with lower alcohol exposure among HIV infected compared with uninfected men. Drug Alcohol Depend. 2016;161:95–103. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Sullivan LE, Goulet JL, Justice AC, Fiellin DA. Alcohol consumption and depressive symptoms over time: a longitudinal study of patients with and without HIV infection. Drug Alcohol Depend. 2011;117(2–3):158–63. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Tate JP, Sterne JAC, Justice AC, Albumin. white blood cell count, and body mass index improve discrimination of mortality in HIV-positive individuals. AIDS. 2019;33(5):903–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Womack JA, Murphy TE, Bathulapalli H, Smith A, Bates J, Jarad S, et al. Serious Falls in Middle-Aged veterans: development and validation of a predictive risk model. J Am Geriatr Soc. 2020;68(12):2847–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Lubin JH, Gail MH. Biased selection of controls for case-control analyses of cohort studies. Biometrics. 1984;40(1):63–75. [PubMed] [Google Scholar]
- 22.Beamont JJ, Steenland K, Minton A, Mayer S. A computer program for incidence density sampling of controls in case-control studies nested within occupational cohort studies. Am J Epidemiol. 1989;129(1):212–9. [DOI] [PubMed] [Google Scholar]
- 23.Sterling RK, Lissen E, Clumeck N, Sola R, Correa MC, Montaner J, et al. Development of a simple noninvasive index to predict significant fibrosis in patients with HIV/HCV coinfection. Hepatology. 2006;43(6):1317–25. [DOI] [PubMed] [Google Scholar]
- 24.Levey AS, Stevens LA, Schmid CH, Zhang YL, Castro AF 3rd, Feldman HI, et al. A new equation to estimate glomerular filtration rate. Ann Intern Med. 2009;150(9):604–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.SAS Institute Inc., SAS 9.4. Cary, NC, USA. [Google Scholar]
- 26.White IR, Royston P, Wood AM. Multiple imputation using chained equations: issues and guidance for practice. Stat Med. 2011;30(4):377–99. [DOI] [PubMed] [Google Scholar]
- 27.Rubin DB, Schenker N. Multiple imputation in health-care databases: an overview and some applications. Stat Med. 1991;10(4):585–98. [DOI] [PubMed] [Google Scholar]
- 28.Labrecque JA, Hunink MMG, Ikram MA, Ikram MK. Do Case-Control Studies Always Estimate Odds Ratios? Am J Epidemiol. 2021;190(2):318–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Dodds TJ. Prescribed Benzodiazepines and suicide risk: a review of the literature. Prim Care Companion CNS Disord. 2017;19(2). [DOI] [PubMed] [Google Scholar]
- 30.Saloner R, Grelotti DJ, Tyree G, Sundermann EE, Ma Q, Letendre S, et al. Benzodiazepine Use is Associated with an increased risk of neurocognitive impairment in people living with HIV. J Acquir Immune Defic Syndr. 2019;82(5):475–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Bohnert ASB, Ilgen MA. Understanding links among opioid use, overdose, and suicide. N Engl J Med. 2019;380(1):71–9. [DOI] [PubMed] [Google Scholar]
- 32.Bryant KJ, Nelson S, Braithwaite RS, Roach D. Integrating HIV/AIDS and alcohol research. Alcohol Res Health. 2010;33(3):167–78. [PMC free article] [PubMed] [Google Scholar]
- 33.Hartzler B, Dombrowski JC, Crane HM, Eron JJ, Geng EH, Christopher Mathews W, et al. Prevalence and predictors of Substance Use Disorders among HIV Care Enrollees in the United States. AIDS Behav. 2017;21(4):1138–48. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Conner A, Azrael D, Miller M. Suicide case-fatality rates in the United States, 2007 to 2014: a Nationwide Population-Based study. Ann Intern Med. 2019;171(12):885–95. [DOI] [PubMed] [Google Scholar]
- 35.Jordan JT, McNiel DE. Characteristics of persons who die on their first suicide attempt: results from the National Violent Death Reporting System. Psychol Med. 2019;50(8):1390–7. [DOI] [PubMed] [Google Scholar]
- 36.Katz IR, Peltzman T, Jedele JM, McCarthy JF. Critical periods for increased mortality after Discharge from Inpatient Mental Health Units: Opportunities for Prevention. Psychiatr Serv. 2019;70(6):450–6. [DOI] [PubMed] [Google Scholar]
- 37.Franklin JC, Ribeiro JD, Fox KR, Bentley KH, Kleiman EM, Huang X, et al. Risk factors for suicidal thoughts and behaviors: a meta-analysis of 50 years of research. Psychol Bull. 2017;143(2):187–232. [DOI] [PubMed] [Google Scholar]
- 38.O’Connell PH, Durns T, Kious BM. Risk of suicide after discharge from inpatient psychiatric care: a systematic review. Int J Psychiatry Clin Pract. 2021;25(4):356–66. [DOI] [PubMed] [Google Scholar]
- 39.Valenstein M, Kim HM, Ganoczy D, McCarthy JF, Zivin K, Austin KL, et al. Higher-risk periods for suicide among VA patients receiving depression treatment: prioritizing suicide prevention efforts. J Affect Disord. 2009;112(1–3):50–8. Epub Oct 22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Chung DT, Ryan CJ, Hadzi-Pavlovic D, Singh SP, Stanton C, Large MM. Suicide Rates after Discharge from Psychiatric Facilities: a systematic review and Meta-analysis. JAMA Psychiatry. 2017;74(7):694–702. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Isaacs JY, Smith MM, Sherry SB, Seno M, Moore ML, Stewart SH. Alcohol use and death by suicide: a meta-analysis of 33 studies. Suicide Life Threat Behav. 2022;18(10):12846. [DOI] [PubMed] [Google Scholar]
- 42.Oscar-Berman M, Marinković K. Alcohol: effects on neurobehavioral functions and the brain. Neuropsychol Rev. 2007;17(3):239–57. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Pompili M, Serafini G, Innamorati M, Dominici G, Ferracuti S, Kotzalidis GD, et al. Suicidal behavior and alcohol abuse. Int J Environ Res Public Health. 2010;7(4):1392–431. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Crum RM, Mojtabai R, Lazareck S, Bolton JM, Robinson J, Sareen J, et al. A prospective assessment of reports of drinking to self-medicate mood symptoms with the incidence and persistence of alcohol dependence. JAMA Psychiatry. 2013;70(7):718–26. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Cato V, Holländare F, Nordenskjöld A, Sellin T. Association between benzodiazepines and suicide risk: a matched case-control study. BMC Psychiatry. 2019;19(1):317. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Sundermann EE, Saloner R, Rubtsova A, Nguyen AL, Letendre S, Moore RC, et al. The association between benzodiazepine use and greater risk of neurocognitive impairment is moderated by medical burden in people with HIV. J Neurovirol. 2022;7(10):022–1076. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Nkhoma ET, Coumbis J, Farr AM, Johnston SS, Chu BC, Rosenblatt LC, et al. No evidence of an Association between Efavirenz exposure and suicidality among HIV Patients initiating antiretroviral therapy in a Retrospective Cohort Study of Real World Data. Med (Baltim). 2016;95(3):e2480. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Smith C, Ryom L, Monforte A, Reiss P, Mocroft A, El-Sadr W, et al. Lack of association between use of efavirenz and death from suicide: evidence from the D:A:D study. J Int AIDS Soc. 2014;17(4 Suppl 3):19512. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Arias-Colmenero T, Pérez-Morente M, Ramos-Morcillo AJ, Capilla-Díaz C, Ruzafa-Martínez M, Hueso-Montoro C. Experiences and attitudes of people with HIV/AIDS: a systematic review of qualitative studies. Int J Environ Res Public Health. 2020;17(2). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.Chambers LA, Rueda S, Baker DN, Wilson MG, Deutsch R, Raeifar E, et al. Stigma, HIV and health: a qualitative synthesis. BMC Public Health. 2015;15(1):848. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51.Meyer JP, Springer SA, Altice FL. Substance abuse, violence, and HIV in women: a literature review of the syndemic. J Womens Health (Larchmt). 2011;20(7):991–1006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52.Pellowski JA, Kalichman SC, Matthews KA, Adler N. A pandemic of the poor: social disadvantage and the U.S. HIV epidemic. Am Psychol. 2013;68(4):197–209. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53.Smith A, Breazeale S, Goulet JL, Vlahov D, Justice AC, Womack JA. A systematic review of risk factors for suicide among persons living with HIV (1996–2020). AIDS Behav. 2022;26(8):2559–73. [DOI] [PubMed] [Google Scholar]
- 54.Bohnert AS, McCarthy JF, Ignacio RV, Ilgen MA, Eisenberg A, Blow FC. Misclassification of suicide deaths: examining the psychiatric history of overdose decedents. Inj Prev. 2013;19(5):326–30. [DOI] [PubMed] [Google Scholar]
- 55.Rockett IRH, Caine ED, Banerjee A, Ali B, Miller T, Connery HS et al. Fatal self-injury in the United States, 1999–2018: unmasking a national mental health crisis. EClinicalMedicine. 2021;32. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
While data are not publicly available, reasonable requests will be considered by the corresponding author.
All programing was conducted in SAS 9.4 and will be made available by the corresponding author if requested.