Gender, mental health symptoms, loss to clinic, and viral non-suppression 6 months after HIV care initiation among a cohort of people with HIV in Cameroon Authors: Angela M Parcesepe, Lindsey M Filiatreau, Peter Vanes Ebasone, Anastase Dzudie, Marcel Yotebieng, Kathryn Anastos, Eric Pefura-Yone, Annereke Nyenti, Denis Nash Published: 2025 ABSTRACT AbstractBackgroundPeople with HIV (PWH) experience high levels of mental health or substance use disorders (MSDs) which may compromise HIV treatment outcomes. Differences in the relationship between MSDs and HIV treatment outcomes by gender remain understudied.MethodsWe conducted interviews with 426 PWH aged ≥ 21 years initiating HIV care in Cameroon between June 2019 and March 2020. Clinic data and laboratory results were used to ascertain loss to clinic and viral non-suppression six months after HIV care initiation. We used log binomial regression models to assess the relationship between symptoms of depression (PHQ-9 scores > 9), anxiety (GAD-7 scores > 9), post-traumatic stress disorder (PTSD; PCL-5 scores > 30), and harmful alcohol use (AUDIT scores > 15) at HIV care initiation and 6-month loss to clinic and viral non-suppression in the study population overall, and among men and women, separately.ResultsNearly 25% of individuals (N = 100) were lost to clinic at 6 months. Among the 326 individuals retained in care at their original clinic at 6 months, 234 had a measured viral load, of whom 28% were virally non-suppressed (N = 65). Among men, the risk of disengagement was greater among those who reported PTSD symptoms at care initiation compared to those who did not. Viral non-suppression was significantly higher among those reporting symptoms of anxiety at care initiation in the population retained overall and among men.ConclusionsMen initiating HIV care with poor mental health may be particularly vulnerable to unstable care engagement and viral non-suppression. Interventions are needed for men specifically to foster improved mental health.Keywords: HIV, Gender, Cameroon, Africa, Depression, Anxiety Background The HIV epidemic remains a significant public health concern throughout the Global South, including in Cameroon. With approximately 490,000 adults living with HIV in Cameroon, HIV prevalence among adults is approximately 2.5% [1]. While 92% of people with HIV (PWH) in Cameroon know their HIV status, only 82% have suppressed HIV viral loads [1]. HIV disproportionately impacts women in Cameroon, with 320,000 women living with HIV. HIV prevalence among women in Cameroon is 3.3% compared to 1.8% among men [1]. However, women with HIV in Cameroon are more likely to have initiated antiretroviral therapy (ART) compared to men, with 95% of women with HIV on ART compared to 79% of men [1]. PWH experience high levels of mental health or substance use disorders (MSDs) across global settings, including sub-Saharan Africa [2]. MSDs are among the most common comorbidities among PWH. The prevalence of depression, the most common mental disorder among PWH, has been estimated between 14% and 32% among PWH in sub-Saharan Africa [3]. Anxiety has been estimated to impact approximately 22% of PWH [4]. Similarly, the prevalence of post-traumatic stress disorder (PTSD) and alcohol use disorders (AUD) among PWH has been estimated at 28% and 27%, respectively [5, 6]. Untreated MSDs may compromise quality of life and achievement of optimal HIV treatment outcomes [7–10]. MSDs have been associated with suboptimal HIV treatment outcomes throughout the care cascade, including delayed HIV testing and care engagement, suboptimal adherence to ART, unsuppressed HIV viral load, and all-cause mortality [8, 10–13]. Gender differences have been observed in both the prevalence of mental health disorders and achievement of HIV treatment outcomes among PWH in sub-Saharan Africa. Similar to findings among the general population, women with HIV are often more likely to report symptoms of depression, anxiety, and PTSD compared to men with HIV [3, 6]. Conversely, men with HIV are often more likely to report heightened alcohol use or symptoms of AUD compared to women [5]. Among PWH in sub-Saharan Africa, men with HIV experience greater HIV-related morbidity and mortality compared to women and are more likely than women to report suboptimal engagement in HIV care, adherence to ART, and viral suppression [14–16]. However, little is known about whether gender modifies the relationship between mental health and HIV care outcomes. Greater understanding of the relationship between gender, mental health, and HIV treatment outcomes can inform the development and implementation of mental health and HIV support services for PWH. The objectives of this research are to examine the relationships among symptoms of depression, anxiety, PTSD, and harmful alcohol use and two HIV treatment outcomes: loss to clinic and HIV viral non-suppression six months after HIV care initiation and the extent to which these relationships differ by gender. Methods This study has been previously described in detail [17]. Briefly, a total of 426 PWH aged 21 years or older who were initiating HIV care at three public HIV treatment centers in Cameroon between June 2019 and March 2020 participated in structured interviews. Interviews captured participants’ sociodemographic characteristics and symptoms of mental health disorders, among other factors. Medical record data, including clinic visit dates, viral load measurements, documented transfers out, and participant vital status were extracted for all participants between the time of enrollment into care and January 1, 2022. The present analysis is focused on the first six months following HIV care initiation. Ethical approval for this study was received by the University of North Carolina at Chapel Hill and the National Ethical Committee of Research for Human Health in Cameroon. Written informed consent was received from all study participants. Measures HIV care outcomes Clinic visit data was used to ascertain loss to clinic and HIV viral non-suppression. Loss to clinic is often defined as being 90 or more days late for a scheduled clinic visit [18, 19]. In keeping with this definition, we allowed for a 90-day window around our measurement point of interest (i.e., month 6 following enrollment into HIV care). That is, if an individual had evidence of an HIV clinic visit 3–9 months after HIV care initiation, they were categorized as being in care or having had an HIV clinic visit at month 6. HIV viral non-suppression was defined as HIV viral load > 40 copies/mL 3–9 months following HIV care initiation. For participants who had multiple viral load measurements within this window, the measurement closest to the individual’s 6-month anniversary of care initiation was used. Mental and substance use disorder symptoms The Patient Health Questionnaire-9 (PHQ-9) was used to assess depressive symptoms [20]. Scores of 10 or greater were categorized as moderate or severe depressive symptoms [20]. The General Anxiety Disorder-7 (GAD-7) was used to assess anxiety symptoms [21]. Scores of 10 or greater were categorized as moderate or severe anxiety symptoms [21]. The PTSD Checklist for DSM-5 (PCL-5) was used to screen for PTSD symptoms [22]. Scores of 31 or greater were categorized as symptoms of probable PTSD. The Alcohol Use Disorders Identification Test (AUDIT) was used to assess alcohol use [23]. Scores of 16 or greater were categorized as symptoms of harmful drinking. Statistical analyses Univariate analyses were conducted to describe loss to clinic and viral non-suppression 6 months following HIV care initiation. Bivariate analyses between each MSD exposure and HIV care outcome were conducted using Pearson chi-squared tests. Separate log binomial regression models were used to quantify the relationship between symptoms of depression, anxiety, PTSD, and harmful alcohol use and each outcome of interest—6-month loss to clinic and viral non-suppression— in the population overall and stratified by gender. Adjusted analyses controlled a priori for identified confounders selected using directed acyclic graphs. Only individuals retained in care at their original clinic 6 months following care initiation were included in regression models estimating the effect of mental health exposures on viral non-suppression. For individuals who were retained in care at their original clinic at 6 months following care initiation but were missing a viral load measurement between 3 and 9 months following care initiation, two individuals (AMP and LMF) independently reviewed the patient’s health records and assigned the most probable outcome based on prior treatment-seeking and laboratory results. Assigned outcomes were compared and discrepancies (N = 13/92; 14%) were discussed until consensus was reached. These results are referred to as the “adjudicated” results below. Sensitivity analyses were also conducted to explore alternative options for handling missing viral suppression data. In the first sensitivity analysis, the thirteen discrepant outcomes from the investigator adjudication process were reverse coded. In the second, multiple imputation was used under the assumption that data were missing at random. Pearson’s chi-squared tests were used to identify measured covariates associated with viral non-suppression in the complete case analysis and measured covariates associated with missingness. One hundred imputations were conducted using the monotone logistic method for each exposure: depression, anxiety, PTSD, and harmful alcohol use, separately. Ultimately, each imputation model included the exposure of interest, covariates in the final analytic model (clinic, hunger, and gender), and three predictors of viral non-suppression (age, time away from home, and employment status). In the final sensitivity analysis, all individuals with missing measurements were considered virally unsuppressed. Results from each sensitivity analysis are presented in the Appendix (Appendix Tables 7, 8, and 9). Results Most participants were women (58.7%), aged 21–39 years (58.5%), in a relationship (58.5%) and working for pay (64.6%) (Table 1). Overall, 76.5% of individuals were retained in HIV care at their original clinic six months after care initiation (72.7% of men and 79.2% of women). Among the 326 individuals retained in HIV care at their original clinic six months after care initiation, 71.8% (n = 234) had a viral load measurement between 3 and 9 months after HIV care initiation. Overall, 27.8% of those retained in care at their original clinic with a viral load measurement were virally non-suppressed (22.4% of women and 36.3% of men). Among individuals with missing viral load data (n = 92), 61% of missing viral loads were adjudicated as non-suppressed, yielding an estimated 37% prevalence of viral non-suppression among those retained in HIV care six months after care initiation. Table 1Sociodemographic characteristics at study enrollment among 426 PWH entering HIV care in Cameroon between June 2019-March 2020, overall and stratified by HIV treatment outcomesFull sample(N = 426)N (%)Lost to clinic(N = 100)N (%)Retained at clinic(N = 326)N (%)Non-missing suppression dataa (N = 234)N (%)Suppresseda(N = 169)N (%)Non-suppresseda(N = 65)N (%)Female250 (58.7)52 (52.0)198 (60.7)143 (61.1)111 (65.7)32 (49.2)Age 21–39249 (58.5)66 (66.0)183 (56.1)124 (53.0)87 (51.5)37 (56.9) 40+177 (41.5)34 (34.0)143 (43.9)110 (47.0)82 (48.5)28 (43.1)Education None31 (7.3)12 (12.0)19 (5.8)13 (5.6)11 (6.5)2 (3.1) Primary218 (51.2)49 (49.0)169 (51.8)122 (52.1)91 (53.8)31 (47.7) ≥Secondary177 (41.5)39 (39.0)138 (42.3)99 (42.3)67 (39.6)32 (49.2)Not in a relationship177 (41.5)49 (49.0)128 (39.3)92 (39.3)66 (39.1)26 (40.0)Childlessb79 (18.6)22 (22.0)57 (17.5)44 (18.8)34 (20.1)10 (15.4)Working for pay275 (64.6)57 (57.0)218 (66.9)161 (68.8)108 (63.9)53 (81.5)Away from home > 1 month in past year164 (38.5)44 (44.0)120 (36.8)84 (35.9)51 (30.2)33 (50.8)Moderate to severe household hungerb119 (28.1)25 (25.0)94 (29.1)62 (26.7)38 (22.6)24 (37.5)aAmong those retained in care at 6-months (N = 326)bMissing: number of children N = 2; household hunger N = 3 Sociodemographic characteristics at study enrollment among 426 PWH entering HIV care in Cameroon between June 2019-March 2020, overall and stratified by HIV treatment outcomes aAmong those retained in care at 6-months (N = 326) bMissing: number of children N = 2; household hunger N = 3 Among the full population, symptoms of depression, anxiety, PTSD, and harmful alcohol use were reported by 20.4%, 19.6%, 15.7%, and 13.4% of respondents, respectively (Table 2). In bivariate analyses, loss to clinic was more common among those reporting symptoms of depression, anxiety, and PTSD. Among those retained in care at six months, viral non-suppression was more common among those who reported symptoms of depression, anxiety, and harmful alcohol use at enrollment into care. Table 2Symptoms of mental health disorders among 426 PWH entering HIV care in Cameroon, overall and stratified by HIV treatment outcomesFull sample(N = 426)N (%)Lost to clinic(N = 100)N (%)Retained at clinic(N = 326)N (%)Non-missing suppression dataa(N = 234)N (%)Suppresseda(N = 169)N (%)Non-suppresseda(N = 65)N (%)Depression87 (20.4)24 (24.0)63 (19.3)37 (15.8)23 (13.6)14 (21.5)Anxietyb83 (19.6)24 (24.2)59 (18.2)39 (16.7)23 (13.7)16 (24.6)Post-traumatic stress disorder67 (15.7)19 (19.0)48 (14.7)34 (14.5)27 (16.0)7 (10.8)Harmful alcohol use57 (13.4)9 (9.0)48 (14.7)34 (14.5)21 (12.4)13 (20.0)aAmong those retained in care at 6-months (N = 326)bMissing anxiety n = 2 Symptoms of mental health disorders among 426 PWH entering HIV care in Cameroon, overall and stratified by HIV treatment outcomes aAmong those retained in care at 6-months (N = 326) bMissing anxiety n = 2 In multivariable models, in the study population overall, no relationships were observed between symptoms of depression, anxiety, PTSD, or harmful alcohol use and 6-month loss to clinic (Table 3). Among men, the risk of loss to clinic was greater among those who reported PTSD symptoms at care initiation compared to those who did not (aRR 2.0 [95% CI 1.1, 3.6]). Symptoms of depression, anxiety, and PTSD were not associated with 6-month loss to clinic among women. Women who reported harmful alcohol use had 0.5 (95% CI 0.1, 1.9) times the risk of loss to clinic at 6 months, in adjusted models, compared to women who did not report harmful drinking. Table 3Log-binomial regression estimates characterizing the relationship between symptoms of depression, anxiety, PTSD, and harmful alcohol use and loss to clinic within 6-months following HIV care initiationAmong total sample population (n = 426)Among men (n = 176)Among women (n = 250)Unadjusted Risk Ratio (95% CI)Adjusted Risk Ratioa (95% CI)Unadjusted Risk Ratio (95% CI)Adjusted Risk Ratiob (95% CI)Unadjusted Risk Ratio (95% CI)Adjusted Risk Ratiob (95% CI)Depression1.23 (0.83, 1.82)1.05 (0.69, 1.59)1.45 (0.84, 2.50)1.54 (0.83, 2.85)1.13 (0.65, 1.96)0.79 (0.45, 1.40)Anxiety1.31 (0.89, 1.95)1.19 (0.80, 1.77)1.06 (0.56, 2.01)1.05 (0.54, 2.03)1.60 (0.96, 2.67)1.25 (0.74, 2.12)Post-traumatic stress disorder1.26 (0.82, 1.92)1.08 (0.70, 1.69)1.75 (1.02, 3.01)1.95 (1.05, 3.63)0.98 (0.52, 1.86)0.76 (0.39, 1.45)Harmful alcohol use0.64 (0.34, 1.20)0.61 (0.33, 1.13)0.62 (0.30, 1.27)0.62 (0.30, 1.28)0.49 (0.13, 1.85)0.51 (0.14, 1.87)aAdjusted for gender, hunger, employment, clinicbAdjusted for hunger, employment, clinic Log-binomial regression estimates characterizing the relationship between symptoms of depression, anxiety, PTSD, and harmful alcohol use and loss to clinic within 6-months following HIV care initiation aAdjusted for gender, hunger, employment, clinic bAdjusted for hunger, employment, clinic Among those retained in care at 6 months, viral non-suppression was significantly higher among those reporting symptoms of anxiety at care initiation in both the population retained overall (aRR 1.4 [95% CI 1.1, 2.0]), and among retained men (aRR 1.9 [95% CI 1.1, 2.9]) (Tables 4 and 5). Among those retained in care at 6 months, the risk of viral non-suppression was 1.3 (95% CI 1.0, 1.9) among those with vs. without depressive symptoms among the population overall and 1.5 (95% CI 0.9, 2.5) among retained men. Neither symptoms of PTSD nor harmful alcohol use were associated with viral suppression at 6 months, in the population overall or when stratified by gender. Similarly, MSD symptoms were not associated with viral suppression among women with HIV retained in care at 6 months (Table 6). Table 4Log-binomial regression estimates characterizing the relationship between depression, anxiety, PTSD, and harmful alcohol use and unsuppressed viral load (viral load³40 copies/mL) 6 months following entry into care, among overall sample population (N=326)Complete Case Analysis Unadjusted Risk Ratio (95% CI)Adjusted Risk Ratioa(95% CI)AdjudicationUnadjusted Risk Ratio (95% CI)Adjusted Risk Ratioa(95% CI)Depression1.46 (0.91, 2.35)1.68 (0.98, 2.88)1.38 (1.01, 1.87)1.35 (0.97, 1.88)Anxiety1.62 (1.04, 2.54)1.70 (1.06, 2.71)1.49 (1.10, 2.01)1.44 (1.06, 1.96)Post-traumatic stress disorder0.71 (0.35, 1.42)0.71 (0.35, 1.47)0.82 (0.52, 1.28)0.77 (0.49, 1.22)Harmful alcohol use1.47 (0.90, 2.39)1.23 (0.74, 2.05)1.29 (0.91, 1.82)1.12 (0.79, 1.60)aAdjusted for gender, hunger, clinic Log-binomial regression estimates characterizing the relationship between depression, anxiety, PTSD, and harmful alcohol use and unsuppressed viral load (viral load³40 copies/mL) 6 months following entry into care, among overall sample population (N=326) aAdjusted for gender, hunger, clinic Table 5Log-binomial regression estimates characterizing the relationship between depression, anxiety, PTSD, and harmful alcohol use and unsuppressed viral load (viral load≥40 copies/mL) 6 months following entry into care among men (N = 128) Complete Case Analysis Unadjusted Risk Ratio(95% CI) Adjusted Risk Ratioa (95% CI)Adjudication Unadjusted Risk Ratio (95% CI)Adjusted Risk Ratioa (95% CI)Depression1.80 (1.00, 3.26)2.22 (1.02, 4.86)1.50 (1.00, 2.25)1.54 (0.94, 2.53)Anxiety2.06 (1.24, 3.44)2.51 (1.32, 4.76)1.72 (1.19, 2.48)1.86 (1.21, 2.86)Post-traumatic stress disorder0.59 (0.17, 2.06)0.50 (0.14, 1.79)0.66 (0.28, 1.52)0.57 (0.24, 1.35)Harmful alcohol use1.14 (0.61, 2.12)1.71 (0.78, 3.72)1.19 (0.79, 1.8)1.13 (0.74, 1.74)aAdjusted for hunger, clinic Log-binomial regression estimates characterizing the relationship between depression, anxiety, PTSD, and harmful alcohol use and unsuppressed viral load (viral load≥40 copies/mL) 6 months following entry into care among men (N = 128) aAdjusted for hunger, clinic Table 6Log-binomial regression estimates characterizing the relationship between depression, anxiety, PTSD, and harmful alcohol use and unsuppressed viral load (viral load ≥ 40 copies/mL) 6 months following entry into care among women (N=198)Complete Case Analysis Unadjusted Risk Ratio (95% CI) Adjusted Risk Ratioa (95% CI)Adjudication Unadjusted Risk Ratio (95% CI)Adjusted Risk Ratioa (95% CI)Depression1.43 (0.72, 2.83)1.27 (0.61, 2.65)1.40 (0.91, 2.16)1.25 (0.79, 1.99)Anxiety1.31 (0.64, 2.69)1.17 (0.53, 2.59)1.38 (0.88, 2.15)1.25 (0.79, 2.00)Post-traumatic stress disorder0.87 (0.37, 2.05)0.87 (0.37, 2.07)0.99 (0.58, 1.69)0.90 (0.52, 1.56)Harmful alcohol use1.71 (0.78, 3.72)1.46 (0.90, 2.36)1.12 (0.57, 2.21)1.07 (0.56, 2.06)aAdjusted for hunger, clinic Log-binomial regression estimates characterizing the relationship between depression, anxiety, PTSD, and harmful alcohol use and unsuppressed viral load (viral load ≥ 40 copies/mL) 6 months following entry into care among women (N=198) aAdjusted for hunger, clinic Discussion Among this cohort of PWH initiating HIV care in Cameroon suboptimal HIV care outcomes were common, with almost one-quarter (23.5%) of participants having disengaged from HIV care at their original clinic six months after care initiation. Among those retained in HIV care at their original clinic at six months, over one-quarter (27.8%) were virally non-suppressed six months after HIV care initiation. Symptoms of PTSD were associated with greater risk of loss to clinic 6 months following care initiation specifically among men. Among those retained in care at 6-months, symptoms of anxiety were associated with greater risk of viral non-suppression among the study population overall, and among men specifically. Similarly, among those retained in care at 6-months, symptoms of depression were associated with greater risk of viral non-suppression among the study population overall, and among men specifically, though these results did not achieve statistical significance. Loss to care at an individual’s original clinic was common six months after HIV care initiation among this cohort of PWH initiating HIV care in Cameroon. This is similar to previous research that has shown that disengagement from HIV care and documented and undocumented clinic transfers are common, particularly early in one’s HIV care journey [24–27]. A study with PWH initiating ART in Kenya found that 33% of individuals were lost to care during the first two years. Notably, half of those who disengaged from HIV care within the first two years disengaged within 6 months of ART initiation [28]. Efforts to retain PWH with MSD symptoms early in their care journeys may be particularly beneficial as PWH newly initiating care may have heightened vulnerability to disengagement. Among men, symptoms of PTSD were associated with loss to clinic in the first six months. Little is known about the relationship between PTSD and HIV care engagement, specifically among men in sub-Saharan Africa. The relationship between PTSD and engagement in care among men with HIV warrants further exploration. Potential mediators and modifiers of this pathway should be explored, including coping mechanisms. It has been theorized that avoidant behavior, common among individuals with PTSD, may mediate the relationship between PTSD and poor HIV care outcomes [29]. In non-HIV related medical settings, a negative relationship between avoidant behaviors and care engagement has been observed, in which decreases in avoidant behaviors have been associated with increased care engagement and adherence [29]. Research should examine if similar pathways exist among PTSD, avoidant coping, and HIV care engagement behaviors among men with HIV in sub-Saharan Africa. Greater understanding of the pathways and relationships among trauma, PTSD, and HIV care outcomes, particularly among men with HIV in sub-Saharan Africa, is needed. In the current analysis, symptoms of depression and anxiety were not associated with loss to clinic six months after HIV care initiation. This is in contrast to prior research which has found depression to be associated with worse retention in HIV care. Among PWH in South Africa, symptoms of depression, but not anxiety were associated with lower odds of being retained in HIV care [30, 31]. Similarly, a study with PWH in Rwanda found that symptoms of depression were associated with time to attrition from HIV care among PWH [32]. Future research to understand the relationships among depression, anxiety, and retention in HIV care in Cameroon is needed. Greater understanding of potential protective factors that facilitate retention in HIV care among individuals with symptoms of depression or anxiety use could yield meaningful insights. Among those retained in care at the original clinic, symptoms of anxiety, but not PTSD or harmful alcohol use were associated with unsuppressed viral load among the population overall and among men. The relationship between symptoms of depression and viral non-suppression approached significance among the study population overall, and among men specifically. Prior research from South Africa that found that neither depression nor PTSD was associated with unsuppressed HIV viral load [33]. However, in contrast to our findings, research with PWH in South Africa found that alcohol use disorder was associated with unsuppressed HIV viral load [33]. Similarly, a study with men with HIV in Uganda found that hazardous drinking was associated with unsuppressed HIV viral load [34]. More research is needed to understand the relationship between harmful alcohol use and viral suppression among PWH in Cameroon. Prior research on the relationship between anxiety symptoms and viral suppression remains limited and equivocal. For example, a study with PWH in Vietnam who reported hazardous drinking found that anxiety symptoms were not associated with viral suppression [35]. However, a study with PWH in the United States found that anxiety symptoms were significantly associated with lower likelihood of sustained viral suppression [36]. More research is needed to understand the relationship between anxiety, depression, and viral suppression, particularly among men with HIV in Cameroon and throughout the Global South. Potentially modifiable mediators and moderators of these relationships should be explored. Men initiating HIV care with poor mental health may be particularly vulnerable to loss to care and HIV viral non-suppression. While research on the relationship among gender, mental health, and HIV care outcomes remains limited, our findings are broadly consistent with a study of adolescents with HIV in Mozambique that found that symptoms of anxiety, depression, and PTSD were associated with higher odds of suboptimal ART adherence among male but not female adolescents with HIV [37]. In addition, previous research with this cohort of PWH initiating HIV care in Cameroon found that depressive symptoms were associated with greater prevalence of entry into HIV care with advanced HIV among men, but not among women [17]. The extent to which gender modifies the relationship between untreated mental health disorders and HIV care continuum outcomes should be further investigated. Gender-specific pathways between mental health and HIV care continuum outcomes should be examined. Interventions to foster improved mental health and HIV treatment outcomes specifically among men with HIV in sub-Saharan Africa are needed. A scoping review of interventions to improve men’s engagement in HIV care in sub-Saharan Africa found that among 71 articles included in the review, one-quarter of included interventions focused exclusively on male partners and just 7 interventions focused exclusively on men [38]. Further, half of the interventions included in this review focused primarily on HIV testing [38]. There remains limited evidence on strategies to engage men in HIV care effectively and sustainably, particularly beyond HIV testing. Such research should consider the role of HIV-related stigma as such stigma has been consistently associated with poor mental health and suboptimal HIV care outcomes throughout sub-Saharan Africa, including in Cameroon [39, 40]. Emerging research suggests the relationship between HIV-related stigma and HIV care outcomes may be mediated by mental health symptoms [41–43]. However, little is known about whether such relationships and pathways differ by gender. This study has limitations worth noting. Data were obtained from three HIV clinics in Cameroon and may not be representative of other regions of Cameroon or sub-Saharan Africa. In addition, our study sample was relatively small and may not be adequately powered for statistical inference. Loss to clinic and viral suppression were obtained from HIV clinic records and were not independently verified. Importantly, loss to clinic is not synonymous with loss to care, as other studies have documented ‘silent transfers’ are common among PWH, including in the Global South and Cameroon specifically [25, 44–46]. Additional research is needed to understand the magnitude of silent transfers among PWH in Cameroon and beyond, and the extent to which transferring care is associated with sub-optimal downstream treatment outcomes broadly. As described, viral load measurements were missing in 28% of individuals who were retained in care at their original clinic 6 months following care initiation. However, multiple sensitivity analyses were conducted to explore possible variations in findings related to the handling of missing data. Results remained largely consistent across analytic methods. Still, additional approaches, including sampling-based methods that have been widely used to improve estimates of loss to HIV clinical follow-up, may be needed to verify true viremia outcomes in patients with missing measurements [47–51]. Conclusions Loss to clinic and viral non-suppression were common six months after HIV care initiation among this cohort of PWH initiating HIV care in Cameroon. The integration of mental health screening into HIV care may help identify individuals at risk of suboptimal HIV care outcomes, particularly among men with HIV. The extent to which mental health screening and treatment is associated with improved HIV care outcomes among men with HIV in Cameroon warrants further investigation. Acknowledgements This work is solely the responsibility of the authors and does not necessarily represent the official views of any of the institutions mentioned below. Appendix Table 7Log-binomial regression estimates characterizing the relationship between depression, anxiety, PTSD, and harmful alcohol use and unsuppressed viral load (viral load ≥ 40 copies/mL) 6 months following entry into care overall (N=326)Reverse adjudicationa Adjusted Risk Ratio (95% CI)Imputed as unsuppresseda Adjusted Risk Ratio (95% CI) Multiple imputationAdjusted Risk Ratioa (95% CI) Depression1.27 (0.89, 1.83)1.34 (1.03, 1.76)1.54 (0.89, 2.65)Anxiety1.54 (1.12, 2.13)1.23, 0.94, 1.591.59 (0.99, 2.55)Post-traumatic stress disorder0.78 (0.48, 1.26)0.77 (0.56, 1.08)0.70 (0.34, 1.41)Harmful alcohol use1.17 (0.80, 1.72)1.05 (0.79, 1.39)1.31 (0.81, 2.12)aAdjusted for gender, hunger, clinic Log-binomial regression estimates characterizing the relationship between depression, anxiety, PTSD, and harmful alcohol use and unsuppressed viral load (viral load ≥ 40 copies/mL) 6 months following entry into care overall (N=326) aAdjusted for gender, hunger, clinic Table 8Log-binomial regression estimates characterizing the relationship between depression, anxiety, post-traumatic stress disorder, and harmful alcohol use and unsuppressed viral load (viral load≥40 copies/mL) 6 months following entry into care among men (N=128)Reverse adjudicationaAdjusted Risk Ratio (95% CI)Imputed as unsuppressedaAdjusted Risk Ratio (95% CI)Multiple imputationaAdjusted Risk Ratio (95% CI)Depression1.49 (0.87, 2.56)1.78 (1.17, 2.69)1.57 (0.69, 3.56)Anxiety1.85 (1.15, 2.98)1.58 (1.07, 2.31)1.93 (1.02, 3.68)Post-traumatic stress disorder0.63 (0.27, 1.49)0.70 (0.38, 1.28)0.47 (0.13, 1.71)Harmful alcohol use1.15 (0.71, 1.85)1.07 (0.74, 1.53)1.14 (0.62, 2.06)aAdjusted for hunger, clinic Log-binomial regression estimates characterizing the relationship between depression, anxiety, post-traumatic stress disorder, and harmful alcohol use and unsuppressed viral load (viral load≥40 copies/mL) 6 months following entry into care among men (N=128) aAdjusted for hunger, clinic Table 9Log-binomial regression estimates characterizing the relationship between depression, anxiety, post-traumatic stress disorder, and harmful alcohol use and unsuppressed viral load (viral load≥40 copies/mL) 6 months following entry into care among women (N=198)Reverse adjudicationa Adjusted Risk Ratio (95% CI) Imputed as unsuppresseda Adjusted Risk Ratio (95% CI) Multiple imputationa Adjusted Risk Ratio (95% CI) Depression1.13 (0.69, 1.86)1.14 (0.80, 1.63)1.41 (0.74, 2.68)Anxiety1.40 (0.87, 2.26)1.06 (0.73, 1.53)1.33 (0.61, 2.88)Post-traumatic stress disorder0.85 (0.48, 1.51)0.78 (0.62, 1.17)0.85 (0.37, 1.99)Harmful alcohol use1.17 (0.61, 2.24)1.03 (0.62, 1.68)1.80 (0.85, 3.81)aAdjusted for hunger, clinic Log-binomial regression estimates characterizing the relationship between depression, anxiety, post-traumatic stress disorder, and harmful alcohol use and unsuppressed viral load (viral load≥40 copies/mL) 6 months following entry into care among women (N=198) aAdjusted for hunger, clinic Authors’ contributions AMP was responsible for conceptualization, funding, analysis, and writing the original draft. LMF was responsible for conceptualization, analysis, and reviewing and editing the manuscript. PVE and EPY were responsible for project administration, and reviewing and editing. AD was responsible for project administration, supervision, and reviewing and editing. MY, DN, AN, and KA were responsible for reviewing and editing. All authors have read and approved the final manuscript. Funding This research was supported by NIMH grant K01MH114721, NIAID grant U01AI096299 and NICHD grant P2CHD050924 (Carolina Population Center). Data availability The dataset contains sensitive information and is not publicly available. However, it could be made available from the first author (AMP) on reasonable request, with approval from the IRB at the University of North Carolina at Chapel Hill to maintain confidentiality. Declarations Ethics approval and consent to participate This study was approved by the Institutional Review Board at University of North Carolina at Chapel Hill in the United States and the National Ethical Committee of Research for Human Health in Yaoundé, Cameroon. All methods were carried out in accordance with relevant guidelines and regulations. All participants received information about the study as well as information about the opportunity to withdraw from the study at any time. All participants provided written informed consent and agreed to be interviewed. Competing interests The authors declare no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement The dataset contains sensitive information and is not publicly available. However, it could be made available from the first author (AMP) on reasonable request, with approval from the IRB at the University of North Carolina at Chapel Hill to maintain confidentiality.