Prerpints.org logo
Preprint
Article

This version is not peer-reviewed.

Complete Non-Retention in HIV Pre-Exposure Prophylaxis Care and Associated Factors Among Men at High Risk of HIV in Tanga, Tanzania

Submitted:

23 April 2026

Posted:

28 April 2026

You are already at the latest version

Abstract
Background: The effectiveness of HIV pre-exposure prophylaxis (PrEP) in preventing new HIV infection remains affected by non-retention globally; the extent and predictors remain unknown in Tanzania. This study determined the prevalence of complete non-retention in PrEP and associated factors among men who have sex with men in Tanga, Tanzania. Methodology: We included 369 men who have sex with men who were aged 18 years and above. Participants were recruited through a respondent-driven sampling and asked if they were interested in starting PrEP. If the answer was yes, baseline data were collected just after the PrEP onboarding session. Complete non-retention was defined as never coming back for follow-up during the follow-up of the PREPTA project. A modified Poisson regression model was used to determine the factors associated with complete non-retention. Results: About 64.5% of participants were completely not retained in PrEP care. Of all participants, 55.7% had either depressive symptoms, anxiety symptoms, probable alcohol use disorder, or their comorbidities. The prevalence of complete non-retention was 15.0% lower among participants with a steady partner (aPR= 0.85, 95% CI: 0.73-0.98, p= 0.034) and those with financial dependents (aPR=0.85, 95% CI: 0.73-1.00, p=0.045) compared to their respective counterparts. We found no significant statistical association (aPR=0.97, 95% CI: 0.83-1.17, p=0.725) between mental disorder symptoms and complete non-retention in PrEP care. Conclusion: The prevalence of complete non-retention was high, and more than half of the participants had mental disorder symptoms. We recommend integration of mental health services while embodying psychosocial support strategies in HIV prevention and treatment programming to have more robust program outcomes.
Keywords: 
non-retention
;  
pre-exposure prophylaxis
;  
men who have sex with men
;  
depressive symptoms
;  
anxiety symptoms
;  
alcohol use disorder
Subject: 
Public Health and Healthcare  -   Public Health and Health Services

1. Introduction

HIV pre-exposure prophylaxis (PrEP) is a modern, highly effective biomedical intervention against HIV acquisition [1]. The effectiveness of PrEP is as high as 99%, especially with good clients’ compliance, adherence as prescribed, and retention while receiving other psychosocial care [2]. In 2015, the World Health Organization (WHO) endorsed the use of PrEP, mainly tenofovir disoproxil fumarate (TDF) or tenofovir alafenamide (TAF) in combination with emtricitabine, after the medication had shown high effectiveness in preventing HIV infection among key and vulnerable populations [3].
Despite the introduction of PrEP, HIV infections continue to be a major public health challenge globally, and not the least with sub-Saharan Africa being home to 67% of the world’s population of people living with HIV [4]. In 2024, about 40.8 million people were living with HIV, and 1.3 million people were newly diagnosed [5]. More than half (55%) of the new infections occurred in members of key populations [6]. Men who have sex with men are one of the key population remains, accounting for 44% of all new infections and experiencing 26 times higher risk of acquiring HIV than members of the general population [6]. This is driven by both biological and psychosocial vulnerabilities, such as fragile rectal mucosa, high-density HIV target cells in rectal mucosa [7], multiple sexual partners, stigma, sexual violence, criminalization, etc., which overall shape the preventive services, including PrEP use [8,9].
PrEP intervention has shown high acceptability and uptake worldwide, including Tanzania, where more than 90% of engaged clients initiate PrEP [10,11]. However, retention in PrEP services remains a challenge [12], with local studies reporting retention rates between 23%[13] and 47% [14]. Some clients never return for follow-up visits after they are enrolled in PrEP care [15]. In addition, tracing of the clients may be difficult because of the health system's challenges, such as siloed systems [16], as clients might have transferred clinics or changed their HIV risk behaviors. In Tanzania, health information systems are still in development, and some do not communicate with one another, so that clients who shift between hospitals are likely to be missed in routine record systems [17,18]. While behavioral modification of risk, such as use of condoms and faithfulness in sexual relationships [19], may partly explain non-retention for the small proportion of clients, the phenomenon remains to be explored fully. Attendance at least one follow-up visit after PrEP initiation represents a critical level of engagement, enabling health risk screening, counseling, and ongoing client support [20], but some clients never come back. We consider those who never came back as completely not retained. This study defined complete non-retention as a proportion of clients who never attended any follow-up visit after initiating PrEP [21] during the six-month period of follow-up.
In understanding complete non-retention in PrEP, we put forward that a phenomenon can be driven by a complex interplay of biopsychosocial factors such as stigma, relationship status, having financial dependants, mental disorder symptoms, and substance use in this population [22]. In Tanzania, 70% of the specific population of men who have sex with men were found to have depressive symptoms, and 45% of probable alcohol use disorder [23]. It is important to note that depressive symptoms, anxiety symptoms, and alcohol use disorder are closely related; often, they co-exist as comorbidities [24,25] with very high multicollinearity and confounding among themselves. Some people with depression, anxiety, or other disorders, particularly the marginalized population, those without knowledge of mental disorders uses alcohol and other drugs as a self-medication to get relief from specific symptoms [26], introducing another layer of complexity.
Depressive symptoms, anxiety symptoms, and substance use may be a response to the stigma, discrimination [27,28] of same sex affiliations [9]. In Tanzania, only heterosexual relationship is formal, while homosexual behaviors are criminalized with up to 30 years or life imprisonment [9]. This contributes to a more hostile environment and minority stress, one of the possible explanations of non-retention in PrEP, as explained by biopsychosocial minority stress theory [8]. According to this theory, men who have sex with men experience chronic stress due to societal stigma, discrimination, and prejudice related to same-sex affiliations [29,30]. These forms of stigma may contribute to the onset and exacerbation of mental health challenges [28], including depression symptoms, anxiety symptoms, and substance use, such as alcohol use, which is a maladaptive coping mechanism [31]. These mental health challenges from a discriminatory environment may reduce self-efficacy in health-seeking behaviors, including retention in PrEP and consequently non-retention [32]. Studies elsewhere have determined an association between retention in PrEP and different mental disorder symptoms, some of them showing no association [33,34,35] while others reporting an association [35,36].
In Tanzania, previous works have assessed PrEP uptake, retention in PrEP care [37], and one study assessed depression among seroconverted men who have sex with men [38]. No retrievable study in Tanzania has determined complete non-retention and associated factors among men who have sex with men on PrEP. This study determined the prevalence of complete non-retention and associated factors among men who have sex with men enrolled in PrEP care in Tanga, Tanzania.

2. Methods

2.1. Study Design and Setting

This paper presents a complete case analysis of men who have sex with men recruited into the pre-exposure prophylaxis rollout in Tanzania (PREPTA) project. The trial details have been provided in previous papers [39,40,41]. The participants were followed up for six months; the analysis focused only on those who never came back for follow-up for six months, and no time factor was counted for the rest. Furthermore, we included only participants from Tanga (control arm), for whom data regarding both alcohol use disorder, depressive symptoms, and anxiety symptoms were available and complete.

2.2. Study Population and Eligibility

The study included men who had sex with men who were enrolled in a PrEP care program in Tanga, Tanzania. It recruited those who were at least 18 years old, had sex with a man in the last 3 months, and lived in the city of Tanga for the past 6 months. Other inclusion criteria were having creatinine clearance exceeding 60 mL/min and consenting to initiate PrEP according to the national HIV guideline [42].

2.3. Sample Size

The study included men who have sex with men with complete data on alcohol use disorder, depressive symptoms, and anxiety symptoms, collected as part of the PREPTA project. The PREPTA project sample size was estimated using a formula for cohort studies that accounted for respondent-driven sampling (RDS) [43]. For the current paper, initially a total sample size of 384 participants was anticipated, based on the Cochran formula using an assumed proportion of 50%, desired precision of 5% [44], which was adequate to achieve the statistical power of 80% when using the design effect of 2. We used a proportion of 50%, which has shown the ability to provide an optimum sample size when the actual proportion is between 10 to 90% [45]. In this case, we used it because no retrievable previous studies have determined complete non-retention in the same population in Tanzania. After data cleaning, only 369 participants had complete data, which was 4% short of the expected sample size. We did post-Hoc power analysis by using the available sample size (369) to see if our study was powered enough to get statistically significant results, where we found the power to have decreased to 79%. This shows that the study retained the significant power to detect the true association between mental disorder symptoms and complete non-retention [46].

2.4. Recruitment Procedures

For this study, we detail the recruitment of the PREPTA project from which our data are emanating. During the PREPTA project, the recruitment procedure applied respondent-driven sampling (RDS). The process began with three initial participants, referred to as “seeds.” These seeds were purposefully selected to ensure diversity in age, geographic location, and educational background. Peer educators assisted in identifying seeds with extensive social networks. Seeds received up to three unique coupons to invite peers upon completing interviews and providing biological samples. The research participants recruited by the initial seeds made up the first wave. Those individuals then recruited others, creating a second wave, and this process continued until the desired number of participants (369) was reached. Participants received information from the research team about PrEP, what it is, how it works, its benefits, and how often it needs to be refilled. They were also reminded about the importance of attending monthly clinic visits to get their refills. As part of the recruitment process, participants were also screened for alcohol use. They remained under follow-up for the six months.

3. Measurements

3.1. Dependent Variable

The primary outcome was complete non-retention in PrEP. To determine the prevalence of complete non-retention, we counted all participants who did not come for any visit during the follow-up period of six months; this number was taken as the numerator over the total number of men who have sex with men enrolled in the project.

3.2. Independent Variables

The independent variables included in this paper were depressive symptoms, anxiety symptoms, probable alcohol use disorder, and sociodemographic, structural, and behavioral characteristics. We assessed depressive and anxiety symptoms using patient health questionnaire (PHQ-2) and the Generalized Anxiety Disorder Questionnaire (GAD-2), respectively. The probable alcohol use disorder was assessed by using the Alcohol Use Disorders Identification Test (AUDIT), a 10-item screening tool developed by the World Health Organization to help healthcare providers identify individuals who may benefit from interventions aimed at reducing alcohol use [47]. Other sociodemographic, behavioral, and structural characteristics, such as age, having financial dependants, and having a steady partner, were assessed using a semi-structured questionnaire. The details of PHQ-2, GAD-2, AUDIT, and semi-structured questionnaire used on scoring systems and their validation in Tanzania have been detailed in our previous manuscripts on mental distress and probable alcohol use disorder, respectively, in the same population [23,48].

3.3. Data Analysis

Descriptive statistics for continuous variables were summarized using the mean and standard deviation, while categorical variables, such as education level, were presented as proportions. In bivariate analysis, the chi-square test was used to screen for the factors associated with complete non-retention, which will be selected for the final regression analysis. Factors were selected to be included in the modified Poisson regression analysis if they achieved a p-value <0.2 in the bivariate analysis, and those identified in previous literature to be related to complete non-retention. Given that the prevalence of complete non-retention (64.5%) exceeds 10%, the modified Poisson logistic regression with robust standard errors was used to determine independent factors associated with complete non-retention, with the measure of association a prevalence ratio. We did four different modified poison regression models to control the effect of potential confounders for both mental disorder symptoms. A mental health expert identified potential confounders by the common knowledge of mental health, the biopsychosocial model in causal pathways of depression, anxiety, and alcohol use disorder. The modified Poisson regression model was chosen because it minimizes the risk of overestimating effects, which can occur when using a conventional logistic regression model [49]. Variables with a p-value <0.05 in the multivariable modified Poisson regression model were considered statistically significant. All analyses were conducted using STATA, version 18.

4. Results

4.1. Prevalence of Complete Non-Retention

The prevalence of complete non-retention to HIV pre-exposure prophylaxis was found to be 64.5% with a margin of error of 4.5% at the 95% confidence interval.

4.2. Distribution of Socio-Demographic Characteristics by Retention Status

A total of 369 participants were included in the analysis. They had a mean (±SD) age of 24.7±5.5 years. More than half (58.81%) were in the age category between 18 and 24 years. Most had never been married (87.53%), and almost two-thirds (65.58%) had financial dependents. Age group (p=0.008), marital status (p=0.028), and having financial dependents (p=0.006) were associated with non-retention (see Table 1 below). The details of other sociodemographic characteristics are detailed in our previous papers [23,48].

4.3. Distribution of Sexual Behavior Characteristics by Complete Non-Retention Status

The mean age at sexual debut among the participants was 15.9 years, with a standard deviation of 2.7 (µ±SD: 15.9 ± 2.7). About 70.73% of participants had their sexual debut before 18, with over a third (37.10%) involving anal, oral, or thigh contact as the first contact. About 40.11% of the participants engaged in anal sex before 18 years of age, and 31.98% of them had a man as their first sexual partner. Just above half, 51.10% of the study participants had a steady partner. About 76.37% of the study participants used lubricants, and half of them reported using them always. Around 64.84% reported having ever tested for HIV, and 72.60% of participants perceived high PrEP self-efficacy. Among all factors, a steady male partner (p=0.031), frequency of lubricant use (p=0.001), lubricant use in the last anal sex (p=0.011), and history of ever tested HIV (p=0.014) were associated with complete non-retention, while other factors were not (Table 2). Other behavioral and structural characteristics have been detailed in previous papers [23,48].

4.4. Distribution of Mental Disorder Symptoms by Complete Non-Retention Status

Regarding the distribution of mental disorders symptoms, 15.72% of study participants had depressive symptoms, while 10.57% presented anxiety symptoms, and about 45.26% of the study participants met the criteria for probable alcohol use disorder. Overall mental disorder symptoms (depressive symptoms, anxiety symptoms, and probable alcohol disorder) were present in more than half (55.56%) of the study participants. No mental disorder was associated with complete non-retention.
Table 3. Distribution of mental disorder symptoms by complete non-retention status.
Table 3. Distribution of mental disorder symptoms by complete non-retention status.
Variable
Total (N)		 Complete non-retention status
p-value
No n (%) Yes n (%)
Depressive symptoms 0.308
Yes 58 (15.72) 24 (41.38) 34 (58.62)
No 311 (84.28) 107 (34.41) 204 (65.59)
Anxiety symptoms 0.683
Yes 39 (10.57) 15 (38.46) 24 (61.54)
No 330 (89.43) 116 (35.15) 214 (64.85)
Probable alcohol use disorder 0.638
Yes 167 (45.26) 65 (36.72) 112 (63.28)
No 192 (54.74) 66 (34.38) 126 (65.62)
Any mental disorder symptoms 0.481
Yes 205 (55.56) 76 (37.07) 129 (62.93)
No 164 (44.44) 55 (33.54) 109 (65.74)

4.5. An Association Between Mental Disorder Symptoms and Complete Non-Retention

We conducted a series of four modified poison regression analyses for all mental disorders, starting with model 1 for depression symptoms, model 2 for anxiety symptoms, model 3 for probable alcohol use disorder, and model 4 for any mental disorder symptoms. In each model, we controlled for possible confounders such as age, level of education, marital status, having financial dependants, PrEP stigma, experienced stigma, history of being arrested, history of forced sex, level of social support, having a steady partner, lubricant use, condom use and HIV knowledge. In all four models, we controlled for similar confounders due to shared pathophysiological pathways and biopsychosocial underpinnings for depression, anxiety, and alcohol use disorders. In addition, in model 1, we controlled for anxiety and alcohol use disorder, in model 2, we controlled for depression and alcohol use disorder, and in model 3, we controlled for depression and anxiety as well. From all these models, none of the mental disorders showed any significant statistical association with complete non-retention to pre-exposure prophylaxis. See Table 4 below, which combines all models.

4.6. Factors Associated with Complete Non-Retention

After controlling for all possible confounders, we did a final modified Poisson regression analysis to determine the association between any mental disorder and complete non-retention. We found no significant statistical association (aPR=0.97, 95% CI: 0.83-1.17, p=0.725) between mental disorder symptoms and non-retention to PrEP. Apart from mental disorder symptoms other factors which showed significant statistical association were, having a steady partner with 15% (aPR= 0.85, 95% CI: 0.73-0.98, p= 0.034) at 15.00% lower prevalence of non-retention compared to their counterparts, and having financial dependents with 15.00% (aPR=0.85, 95% CI: 0.73-1.00, p=0.045) at low prevalence of non-retention compared to those without.
Table 5. Modified Poisson logistic regression analysis of factors associated with complete non-retention.
Table 5. Modified Poisson logistic regression analysis of factors associated with complete non-retention.
Variables Crude prevalence ratio Adjusted prevalence ratio
cPR (95%CI) p-value aPR (95%CI) p-value
Age groups (years)
18-24 Ref. Ref. 0.506
25+ 0.81 (0.69-0.95) 0.028 0.97 (0.78-1.13)
Level of education
Primary and below Ref. Ref.
Secondary and above 0.89 (0.77-1.04) 0.154 0.86 (0.74-1.01) 0.064
Marital status
Never married Ref. Ref.
Married currently or previously 0.75 (0.56-1.01) 0.061 0.90 (0.66-1.26) 0.579
Having financial dependents
Yes 0.80 (0.69-0.93) 0.004 0.85 (0.73-1.00) 0.045
No Ref. Ref.
Type of sex at sex debut
Anal/Oral/Thighs 0.89 (0.76-1.05) 0.165 0.84 (0.70-1.03) 0.090
Vaginal Ref. Ref.
Age at first anal sex
<18 Ref. Ref.
18+ 1.15 (0.99-1.33) 0.075 1.19 (0.98-1.44) 0.082
Steady partner
Yes 0.85 (0.73-0.99) 0.032 0.85 (0.73-0.98) 0.034
No Ref. Ref.
Lubricant use in the last anal sex
Yes 0.79 (0.69-0.92) 0.002 0.85 (0.69-1.04) 0.112
No Ref. Ref.
Frequency of lubricant use
Never used Ref. Ref.
Sometimes 1.04 (0.87-1.23) 0.693 1.16 (0.94-1.42) 0.162
Always 0.77 (0.64-0.92) 0.004 1.01 (0.77-1.31) 0.956
History of group sex
Yes 1.15 (0.94-1.41) 0.067 1.11 (0.91-1.37) 0.317
No Ref. Ref.
Perceived high risk
Yes 0.95 (0.81-1.12) 0.556 0.98 (0.92-1.26) 0.348
No Ref. Ref.
Experienced stigma
No Ref. Ref.
Yes 1.13 (0.92-1.39) 0.231 1.19 (0.95-1.27 ) 0.216
PrEP stigma
Low Ref. Ref.
High 0.86 (0.70-1.07) 0.178 0.84 (0.68-1.03) 0.097
Level of social support
Inadequate 1.10 (0.91-1.34) 0.315 0.90 (0.75-1.08) 0.271
Adequate Ref. Ref.
Any Mental disorder symptoms
Yes 0.95 (0.81-1.10) 0.180 0.97 (0.83-1.17) 0.725
No Ref. Ref.

5. Discussion

We analyzed complete non-retention in PrEP and its association with mental disorder symptoms among men who have sex with men enrolled in the PREPTA project in Tanga, Tanzania. We found that 64.5% of participants did not come for any follow-up visit at all following their enrollment in PrEP care. Having a steady sexual partner and financial dependence was associated with better retention in care. Mental disorders did not influence retention in PrEP care.
The high level of complete non-retention is relatively high compared to 21% reported by Martinson and colleagues [50] and 44% PrEP discontinuation reported within 30 days in a study conducted in Washington, USA [51]. This difference may partly be explained by differences in follow-up duration and study population difference as a study in Washington included male, female, and transgender individuals, while our study exclusively focused on men who have sex with men. Studies conducted in Brazil reported a non-retention rate at 9.2% [52] and 17% [53], respectively, compared to the rate observed in the current study; however, these studies followed participants up to 48 weeks, whereas our follow-up period was relatively shorter at six months. Apart from these studies from Washington and Brazil, another study in Australia reported that discontinuation from the PrEP was more than half of the participants[54], which is still lower compared to the current study findings. Furthermore, the prevalence of non-retention in this study is higher than 47% reported in a meta-analysis from sub-Saharan Africa, which included studies with different follow-up times up to six months [12]. Our study findings show a significantly high level of complete non-retention in PrEP care, which is relatively high compared to comparative studies, even though follow up period varied. With all these, the consistently reported higher prevalence of no-retention, it calls for context-specific and population-based strategies to improve the PrEP continuum of use.
This study found that about 55.56% of all study participants had either depressive symptoms, anxiety symptoms, probable alcohol use disorder, or their comorbidity, which cannot be ignored. This prevalence is nearly four times compared to general population [55] according to WHO, calling attention from health stakeholders. The higher prevalence of mental disorder symptoms is largely explained by minority stress theory [30], which entails a role of chronic stress due to stigma, and other prejudices against sexual minorities like men who have sex with men in different countries, including Tanzania, where same sex is criminal [9]. Other biological vulnerabilities, such as high susceptibility of rectal mucosa to HIV infection [7], can also account for this elevated prevalence. Addressing mental health matters among men who have sex with men is therefore an important step towards a holistic care approach in fulfilling the Millennium Development Goal number three (MDG3), targeting maintaining quality life and better well-being for all [56]. Evidence-based practices show that the integration of mental health services in regular HIV care and treatments improves overall treatment outcomes while addressing both concerns [57,58].
Despite the high prevalence of mental disorder symptoms in this study, we found no association between mental disorder symptoms (depressive symptoms, anxiety symptoms and probable alcohol use disorder) and non-retention in PrEP prophylaxis among men who have sex with men. The study findings are consistent with what was found by Mehrotra et al on 2016, in a study which involved six countries, where they found no significant association between depression and retention in PrEP care [59]. In addition, findings from our study are in line with findings in studies conducted in Philadelphia, USA [33,34] and in China [35]. However, the comparison studies dealt with individual mental disorders such as anxiety, depression, and substance use symptoms separately. Also, a study conducted in Kenya and Uganda among women and men initiated on oral pre-exposure prophylaxis found no association between depression and continued PrEP use, which is consistent with our findings, even though the study involved a larger and more diverse population [60]. Furthermore, our study findings are contrary to other studies which reported a significant association between mental disorders and retention in PrEP [61,62], including a scoping review of literatures around the world which involved five databases [63]. It is important to note that the comparative studies dealt with separate mental disorder symptoms such as depression alone, but our study included four different disorders, including depression, anxiety, alcohol disorder, and their comorbidities, limiting full comparison. Not finding an association between mental disorder symptoms and complete non-retention may partly be explained by enrollment screening that captured men having sex with other men who were motivated to seek care [64], screening of mental disorder symptoms and enrollment being done at the same time, and different motivations towards PrEP use beyond the project’s objective. Additionally, mental disorder symptoms may heighten the perceived HIV risk that encourages continued use of PrEP [65]. Furthermore, non-retention has been common to nearly two-thirds of all participants, which could affect the distribution of mental disorder symptoms and other characteristics that can affect the association. To complete this study's findings, future studies could assess subjective experiences of mental disorder symptoms and non-retention phenomenologically.
The current study found that men with a steady partner had 15% lower prevalence of non-retention compared to their counterparts. This finding is contrary to what was reported by Mubezi and colleagues in Rwanda in 2024, that men who have sex with men who have multiple sexual partners were more likely to be retained in PrEP care [66]. In explaining this, the relationship dynamics should be considered, including good communications, encouragements, and even financial support, as the perception about relationship support has been associated with PrEP use [67]. The association can also be explained by perceived or experienced good social support and emphasis from the partner, driven by the commitments and futures in their steady relationships, compared to people who have multiple sexual partners [68,69].
Lastly, our study found that participants who had financial dependents had 15% lower prevalence of non-retention compared to those who did not have. These findings are in line with reports from different studies, which have shown a strong association between financial status and pre-exposure prophylaxis use, such as people stopping PrEP due to financial hardships [70,71,72]. On the other hand, this association can be explained by the fact that individuals with financial dependents are more likely to engage in positive health-seeking behaviors, such as continued PrEP use in order to remain healthy and able to work, compared to those without these kinds of responsibilities. This would be in line with the role to care for others theory, which has been reported as an important theme from the previous studies that analyzed motives of adherence to treatments among people living with HIV [73,74,75]. Studies that have analyzed the found association are scarce, but it presents an important socioeconomic finding that is protective and enhances the PrEP care in key and vulnerable populations.

Study Limitations

This study provides the first known estimate of complete non-retention in PrEP care among men who have sex with men in Tanzania. The study had several limitations, which have been discussed in detail in previous papers [23,48]. Given that we only did a complete case analysis, the study was prone to selection bias, healthcare negligence, and ethical implications for those who might have dropped out due to mental disorder symptoms. However, this was respondent driven sample and self-motivated, with no severe mental disorder case detection during baseline assessment. This study remains a background study estimating non-retention in the same environment and population.

6. Conclusions and Recommendations

The prevalence of complete non-retention was high, and more than half of the participants had mental disorder symptoms. We recommend integration of mental health services while embodying social-economic and partner-support strategies in all HIV prevention and treatment programming to promote holistic care and improve program outcomes.

Author Contributions

NK: Conceptualization, Data curation, Formal analysis, Visualization, Writing – original draft. EM: Supervision, Project administration, Writing – review & editing, Conceptualization, Methodology. ML: Resources, Writing – review & editing, Funding acquisition, Validation, Project administration, Data curation, Supervision, Methodology, Conceptualization. KM: Validation, Project administration, Methodology, Conceptualization, Supervision, Funding acquisition, Writing – review & editing, Investigation, Resources. CHM: Data cleaning, technical support, statistical modeling, and analysis. EJM: Conceptualization, Funding acquisition, Methodology, Resources, Software, Validation, Writing – review & editing.

Funding

The author(s) declared that financial support was received for this work and/or its publication. This study was funded by the Research Council of Norway through the Global Health and Vaccination Programme (GLOBVAC) (project number 285361).

Institutional Review Board Statement

This study was conducted in accordance with the Declaration of Helsinki and was approved by the Muhimbili University of Health and Allied Sciences Ethical Review Committee in Tanzania, Regional Committee for Medical and Health Research Ethics (REK) in Norway. The study was conducted in accordance with the local legislation and institutional requirements.

Data Availability Statement

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation. Contact details: Prof. Elia J Mmbaga Email: elia.mmbaga@medisin.uio.no.

Acknowledgments

We sincerely thank the DOCEHTA project team for their dedication to fulfilling the project and supporting us throughout. We are also deeply grateful to the participants, as their involvement made this study possible.

Conflicts of Interest

The authors declare no conflicts of interest.

Abbreviations

The following abbreviations are used in this manuscript:
AIDS Acquired immune deficiency syndrome
AUDIT Alcohol Use Disorder Identification Test
DOCEHTA Strengthening Doctoral Education for Health in Tanzania
GAD-2 Generalized anxiety disorder questionnaire- two-question version
HIV Human immunodeficiency virus
IRB Institutional Review Board
MDG 3 Millennium Development Goal number three
MUHAS Muhimbili University of Health and Allied Sciences
NORAD Norwegian Programme for Capacity Development in Higher Education and Research for Development
PHQ-2 Patient Health Questionnaire two-question version
PrEP Pre-exposure prophylaxis
PREPTA Pre-exposure Prophylaxis Project rollout in Tanzania
REC Research Ethics Committee
UNAIDS Joint United Nations Programme on HIV/AIDS
WHO World Health Organization

References

  1. World Health Organization. WHO implementation tool for pre-exposure prophylaxis (PrEP) of HIV infection: provider module for oral and long-acting PrEP [Internet]. Geneva: World Health Organization; 2024 [cited 2024 Sep 29]. Available from: https://www.who.int/publications/i/item/9789240097230.
  2. Munyaneza, A.; Bhutada, K.; Shi, Q.; Zotova, N.; Nsereko, E.; Muhoza, B.; Kubwimana, G.; Murenzi, G.; Nyirazinyoye, L.; Anastos, K.; et al. High retention among key populations initiated on HIV pre-exposure prophylaxis in Kigali City, Rwanda. J. Int. AIDS Soc. 2024, 27, e26392. [CrossRef]
  3. World Health Organization. Guideline on when to start antiretroviral therapy and on pre-exposure prophylaxis for HIV. Geneva: World Health Organization; 2015. p. 1–76.
  4. Moyo, E.; Moyo, P.; Murewanhema, G.; Mhango, M.; Chitungo, I.; Dzinamarira, T. Key populations and Sub-Saharan Africa's HIV response. Front. Public Heal. 2023, 11. [CrossRef]
  5. United Nations Programme on HIV/AIDS (UNAIDS). Fact sheet 2024: Global HIV statistics [Internet]. Geneva: UNAIDS; 2024 [cited 2024 Sep 29]. Available from: https://www.unaids.org/en/resources/fact-sheet.
  6. UNAIDS. New HIV infections data among key populations: proportions in 2010 and 2022 [Internet]. 2024. p. 1–14. Available from: http://www.wipo.int/amc/en/mediation/rules.
  7. Mayer, K.H.; Nelson, L.; Hightow-Weidman, L.; Mimiaga, M.J.; Mena, L.; Reisner, S.; Daskalakis, D.; A Safren, S.; Beyrer, C.; Sullivan, P.S. The persistent and evolving HIV epidemic in American men who have sex with men. Lancet 2021, 397, 1116–1126. [CrossRef]
  8. Meyer IH. Prejudice, social stress, and mental health in lesbian, gay, and bisexual populations: conceptual issues and research evidence. Psychol Bull. 2003 Sep;129(5):674–97.
  9. The United Republic of Tanzania. The Penal Code [Internet]. Dodoma: Government Press; 2022 [cited 2024 Apr 29]. Available from: https://www.parliament.go.tz/polis/uploads/bills/acts/1662102362-CHAPTER%2016-THE%20PENAL%20CODE.pdf.
  10. Mpirirwe, R.; Segawa, I.; Ojiambo, K.O.; Kamacooko, O.; Nangendo, J.; Semitala, F.C.; Kyambadde, P.; Kalyango, J.N.; Kiragga, A.; Karamagi, C.; et al. HIV pre-exposure prophylaxis uptake, retention and adherence among female sex workers in sub-Saharan Africa: a systematic review. BMJ Open 2024, 14, e076545. [CrossRef]
  11. Kagaayi, J.; Batte, J.; Nakawooya, H.; Kigozi, B.; Nakigozi, G.; Strömdahl, S.; Ekström, A.M.; Chang, L.W.; Gray, R.; Reynolds, S.J.; et al. Uptake and retention on HIV pre-exposure prophylaxis among key and priority populations in South-Central Uganda. J. Int. AIDS Soc. 2020, 23, e25588. [CrossRef]
  12. Zhang, J.; Li, C.; Xu, J.; Hu, Z.; E Rutstein, S.; Tucker, J.D.; Ong, J.J.; Jiang, Y.; Geng, W.; Wright, S.T.; et al. Discontinuation, suboptimal adherence, and reinitiation of oral HIV pre-exposure prophylaxis: a global systematic review and meta-analysis. Lancet HIV 2022, 9, e254–e268. [CrossRef]
  13. Kiondo, F.; Metta, E.; Mmbaga, E.J.; Leshabari, M.T.; Swai, C.; Mbotwa, C.H.; Moen, K. Retention in HIV Pre-Exposure Prophylaxis Among Men Who Have Sex with Men in Tanga, Tanzania. HIV/AIDS - Res. Palliat. Care 2025, ume 17, 185–194. [CrossRef]
  14. Mrisho A, Rutatinisibwa H, Msalale G, Shemtandulo R, Mturi B, Ambonisye S, et al. Reasons influencing PrEP continuity among key and priority populations in Tabora region, Tanzania: a cross-sectional study [Internet]. 2022 [cited 2025 Mar 27]. Available from: https://sjhresearchafrica.org/index.php/public-html/article/view/285/185.
  15. Lichtwarck, H.O.; Mbotwa, C.H.; Kazaura, M.R.; Moen, K.; Mmbaga, E.J. Early disengagement from HIV pre-exposure prophylaxis services and associated factors among female sex workers in Dar es Salaam, Tanzania: a socioecological approach. BMJ Glob. Heal. 2023, 8, e013662. [CrossRef]
  16. Rayanakorn, A.; Chautrakarn, S.; Intawong, K.; Chariyalertsak, C.; Khemngern, P.; Olson, D.; Chariyalertsak, S. Experiences and challenges of pre-exposure prophylaxis initiation and retention among high-risk populations: qualitative insights among service providers in Thailand. Front. Public Heal. 2024, 12, 1366754. [CrossRef]
  17. Nyongole, O.; Munishi, C.; Kimu, N.; Sangeda, R.Z.; Mseti, M.; Lugina, E.; Kiwara, F.; Sirili, N.; Chacha, F.; Nicholaus, B.; et al. Challenges of data quality for the management of prostate cancer in Tanzania: A hurdle towards improving the quality of service. PLOS Glob. Public Heal. 2025, 5, e0005579. [CrossRef]
  18. Shidende NH. Challenges in implementing patient-centred information systems: an activity theory perspective. 2014. p. 1–20.
  19. Vanbaelen, T.; Rotsaert, A.; Jacobs, B.K.; Florence, E.; Kenyon, C.; Vuylsteke, B.; Laga, M.; Thijs, R. Why Do HIV Pre-Exposure Prophylaxis Users Discontinue Pre-Exposure Prophylaxis Care? AIDS Patient Care STDs 2022, 36, 159–167. [CrossRef]
  20. Rusley J, Arnold T, Chan PA, Martin H, Parent H, Gashi E, et al. HIV pre-exposure prophylaxis (PrEP) toolkit for use with adolescents and young adults in primary care. 2025.
  21. Hovaguimian, F.; Martin, E.; Reinacher, M.; Rasi, M.; Schmidt, A.; Bernasconi, E.; El Amari, E.B.; Braun, D.; Calmy, A.; Darling, K.; et al. Participation, retention and uptake in a multicentre pre-exposure prophylaxis cohort using online, smartphone-compatible data collection. HIV Med. 2021, 23, 146–158. [CrossRef]
  22. Chen, L.H.; Law, W.; Chang, D.H.F.; Sun, D. Editorial: The bio-psycho-social approach to understanding mental disorders. Front. Psychol. 2023, 14, 1225433. [CrossRef]
  23. Kitomary, N.R.; Metta, E.; Leshabari, M.T.; Moen, K.; Mmbaga, E.J. Prevalence and Factors Associated with Probable Alcohol Use Disorder Among Men at Higher Risk of HIV Enrolled in the HIV Pre-Exposure Prophylaxis Program in Tanga, Tanzania. A Call for Integrated Care. Subst. Abus. Rehabilitation 2025, ume 16, 293–307. [CrossRef]
  24. Johannessen, E.L.; Andersson, H.W.; Bjørngaard, J.H.; Pape, K. Anxiety and depression symptoms and alcohol use among adolescents - a cross sectional study of Norwegian secondary school students. BMC Public Heal. 2017, 17, 1–9. [CrossRef]
  25. Gratzer, D.; Levitan, R.D.; Sheldon, T.; Toneatto, T.; A Rector, N.; Goering, P. Lifetime rates of alcoholism in adults with anxiety, depression, or co-morbid depression/anxiety: a community survey of Ontario. J. Affect. Disord. 2004, 79, 209–215. [CrossRef]
  26. Turner, S.; Mota, N.; Bolton, J.; Sareen, J. Self-medication with alcohol or drugs for mood and anxiety disorders: A narrative review. Depression Anxiety 2018, 35, 851–860. [CrossRef]
  27. Brandt, L.; Liu, S.; Heim, C.; Heinz, A. The effects of social isolation stress and discrimination on mental health. Transl. Psychiatry 2022, 12, 1–11. [CrossRef]
  28. O'DOnnell, A.T.; Foran, A.-M. The link between anticipated and internalized stigma and depression: A systematic review. Soc. Sci. Med. 2024, 349, 116869. [CrossRef]
  29. Harper GW, Hong C, Jauregui JC, Odhiambo EO, Jadwin-Cakmak L, Olango K, et al. Proximal and distal minority stressors and mental health among young GBMSM in Kisumu, Kenya. Am J Community Psychol. 2024 Dec 1.
  30. Christian, L.M.; Cole, S.W.; McDade, T.; Pachankis, J.E.; Morgan, E.; Strahm, A.M.; Dush, C.M.K. A biopsychosocial framework for understanding sexual and gender minority health. Neurosci. Biobehav. Rev. 2021, 129, 107–116. [CrossRef]
  31. Okonkwo N, Rwema JOT, Lyons C, Liestman B, Nyombayire J, Olawore O, et al. Relationship between sexual behavior stigma and depression among MSM and transgender women in Kigali, Rwanda. Int J Ment Health Addict. 2022;20(6):3228–43.
  32. Quinn K, Bowleg L, Dickson-Gomez J. Intersectional stigma and PrEP use among young Black GBMSM. Soc Sci Med. 2019;232:86–93. Available from: . [CrossRef]
  33. Drallmeier T, Garrett EK, Meyr A, Salas J, Scherrer JF. Factors associated with starting PrEP. Prev Med. 2022;164:107344. Available from: . [CrossRef]
  34. Wood SM, Morales KH, Metzger D, Davis A, Fiore D, Petsis D, et al. Mental health and PrEP utilization. AIDS Behav. 2021;25(2):524–31.
  35. Young LB, Lalley-Chareczko L, Clark D, Ramos MT, Nahan RA, Troutman GS, et al. Correlation of PrEP adherence and mental health diagnosis. Int J STD AIDS. 2020;31(5):440–6.
  36. Velloza J, Heffron R, Amico KR, Rowhani-Rahbar A, Hughes JP, Li M, et al. Effect of depression on PrEP adherence. AIDS Behav. 2020;24(7):2178–87.
  37. Faini D, Munseri P, Sandstrom E, Hanson C, Bakari M. Awareness and use of PrEP among FSWs in Dar es Salaam. AIDS Behav. 2023;27(1):335–43.
  38. Mgopa LR, Mbwambo J, Likindikoki S, Pallangyo P. Violence and depression among MSM in Tanzania. BMC Psychiatry. 2017;17(1).
  39. Mbotwa C, Kazaura M, Moen K, Leshabari M, Metta E, Leyna G, et al. Predictors of mHealth use for PrEP adherence. BMC Health Serv Res. 2022;22(1).
  40. Mbotwa CH, Kazaura MR, Moen K, Lichtwarck HO, Leshabari MT, Metta E, et al. Effect of mHealth intervention on retention. Digit Health. 2023;9.
  41. Lichtwarck HO, Kazaura MR, Moen K, Mmbaga EJ. Harmful alcohol use among FSWs initiating PrEP. Int J Environ Res Public Health. 2023;20(1).
  42. Ministry of Health, The United Republic of Tanzania. National guidelines for the management of HIV and AIDS [Internet]. 2019 [cited 2024 May 30]. Available from: https://differentiatedservicedelivery.org/wp-content/uploads/national_guidelines_for_the_management_of_hiv_and_aids_2019.pdf.
  43. Gile KJ, Handcock MS. Respondent-driven sampling: an assessment of current methodology. Sociol Methodol. 2012;40(1):285–327 [cited 2024 Jun 1]. Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3437336/.
  44. World Health Organization. Introduction to HIV/AIDS and sexually transmitted infection surveillance: Module 4 – Introduction to respondent-driven sampling [Internet]. Available from: https://applications.emro.who.int/dsaf/emrpub_2013_en_1539.pdf.
  45. Naing L, Nordin R Bin, Rahman HA, Naing YT. Sample size calculation for prevalence studies. BMC Med Res Methodol. 2022;1–8. Available from: . [CrossRef]
  46. Biau, D.J.; Kernéis, S.; Porcher, R. Statistics in Brief: The Importance of Sample Size in the Planning and Interpretation of Medical Research. Clin. Orthop. Relat. Res. 2008, 466, 2282–2288. [CrossRef]
  47. Babor TF, Robaina K. The Alcohol Use Disorders Identification Test (AUDIT): a review of graded severity algorithms and national adaptations. Int J Alcohol Drug Res. 2016;5(2):17–24. Available from: https://ijadr.org/index.php/ijadr/article/view/222.
  48. Kitomary NR, Metta E, Leshabari MT, Moen K, Mmbaga EJ. Mental distress among men at higher risk of HIV infection enrolled in PrEP program in Tanga, Tanzania. 2026.
  49. Gnardellis C, Notara V, Papadakaki M, Gialamas V, Chliaoutakis J. Overestimation of relative risk. Diagnostics. 2022;12(11).
  50. Martinson T, Montoya R, Moreira C, Kuncze K, Sassaman K, Heise MJ, et al. Urine tenofovir test predicts PrEP discontinuation. AIDS. 2024;38(11):1671–6.
  51. Drezner K, Coleman M, Visconti A, Thomas C, Beverley J, Harold RE, et al. Predictors of PrEP retention. AIDS Behav. 2024;28(8):2598–606.
  52. Jalil EM, Torres TS, Luz PM, Monteiro L, Moreira RI, de Castro CRV, et al. Low PrEP adherence despite high retention. J Int AIDS Soc. 2022;25(3):1–10.
  53. Grinsztejn B, Hoagland B, Moreira RI, Kallas EG, Madruga JV, Goulart S, et al. Retention and adherence to PrEP. Lancet HIV. 2018;5(3):e136–45.
  54. Philpot SP, Murphy D, Chan C, Haire B, Wells N, Fraser D, et al. Patterns of discontinuing and recommencing PrEP. AIDS Behav. 2023;27(9):2891–901. Available from: . [CrossRef]
  55. World Health Organization. Mental disorders [Internet]. WHO fact sheet. Available from: https://www.who.int/news-room/fact-sheets/detail/mental-disorders [cited 2026 Jan 20].
  56. United Nations Population Fund (UNFPA). Transforming our world: the 2030 Agenda for Sustainable Development [Internet]. Available from: https://www.unfpa.org/resources/transforming-our-world-2030-agenda-sustainable-development.
  57. Conteh, N.K.; Latona, A.; Mahomed, O. Mapping the effectiveness of integrating mental health in HIV programs: a scoping review. BMC Heal. Serv. Res. 2023, 23, 1–14. [CrossRef]
  58. Bulstra, C.A.; Hontelez, J.A.C.; Otto, M.; Stepanova, A.; Lamontagne, E.; Yakusik, A.; El-Sadr, W.M.; Apollo, T.; Rabkin, M.; UNAIDS Expert Group on Integration ,; et al. Integrating HIV services and other health services: A systematic review and meta-analysis. PLOS Med. 2021, 18, e1003836. [CrossRef]
  59. Mehrotra ML, Glidden DV, McMahan V, Amico KR, Defechereux P, Mayer KH, et al. Marginal structural model analysis of the iPrEx OLE study. 2017;20(7):1527–34.
  60. Velloza, J.; Baeten, J.M.; Haberer, J.; Ngure, K.; Irungu, E.; Mugo, N.R.; Celum, C.; Heffron, R.; for the Partners Demonstration Project Team Effect of Depression on Adherence to Oral PrEP Among Men and Women in East Africa. Am. J. Ther. 2018, 79, 330–338. [CrossRef]
  61. Akolo M, Nyamache AK, Gichuki R, Gelmon L. Factors affecting retention of men having sex with men on pre-exposure prophylaxis within Nairobi City County, Kenya – experimental study [Internet]. 2019. Available from: https://www.researchgate.net/publication/333547220.
  62. Velloza, J.; Hosek, S.; Donnell, D.; Anderson, P.L.; Chirenje, M.; Mgodi, N.; Bekker, L.-G.; Delany-Moretlwe, S.; Celum, C.; HPTN 082 study group Assessing longitudinal patterns of depressive symptoms and the influence of symptom trajectories on HIV pre-exposure prophylaxis adherence among adolescent girls in the HPTN 082 randomized controlled trial. J. Int. AIDS Soc. 2021, 24, e25731. [CrossRef]
  63. Collins, P.Y.; Velloza, J.; Concepcion, T.; Oseso, L.; Chwastiak, L.; Kemp, C.G.; Simoni, J.; Wagenaar, B.H. Intervening for HIV prevention and mental health: a review of global literature. J. Int. AIDS Soc. 2021, 24, e25710. [CrossRef]
  64. Grootendorst, D.C.; Jager, K.J.; Zoccali, C.; Dekker, F.W. Screening: why, when, and how. Kidney Int. 2009, 76, 694–699. [CrossRef]
  65. Orser, L.; O’bYrne, P.; Holmes, D. Perceptions, motivations, and beliefs about HIV risk and pre-exposure prophylaxis (PrEP) among participants in a nurse-led PrEP service (PrEP-RN). BMC Infect. Dis. 2022, 22, 1–11. [CrossRef]
  66. Mubezi, S.; Malamba, S.S.; Rwibasira, G.N.; Uwineza, J.; Kayisinga, J.d.D.; Remera, E.; Ikuzo, B.; Ndengo, E.; Umuhoza, N.; Sangwayire, B.; et al. Factors associated with retention and adherence on Pre-Exposure Prophylaxis among men who have sex with men in Kigali, Rwanda. PLOS Glob. Public Heal. 2024, 4, e0004063. [CrossRef]
  67. Kahle, E.M.; Sharma, A.; Sullivan, S.; Stephenson, R. The Influence of Relationship Dynamics and Sexual Agreements on Perceived Partner Support and Benefit of PrEP Use Among Same-Sex Male Couples in the U.S.. AIDS Behav. 2020, 24, 2169–2177. [CrossRef]
  68. Felsher, M.; Dutra, K.; Monseur, B.; Roth, A.M.; Latkin, C.; Falade-Nwulia, O. The Influence of PrEP-Related Stigma and Social Support on PrEP-Use Disclosure among Women Who Inject Drugs and Social Network Members. AIDS Behav. 2021, 25, 3922–3932. [CrossRef]
  69. Mehrotra, M.L.; Amico, K.R.; McMahan, V.; Glidden, D.V.; Defechereux, P.; Guanira, J.V.; Grant, R.M. The Role of Social Relationships in PrEP Uptake and Use Among Transgender Women and Men Who Have Sex with Men. AIDS Behav. 2018, 22, 3673–3680. [CrossRef]
  70. Hudrudchai, S.; Suwanwong, C.; Prasittichok, P.; Mohan, K.P.; Janeaim, N. Pre-exposure Prophylaxis Adherence Among Men Who Have Sex With Men: A Systematic Review. J. Prev. Med. Public Heal. 2024, 57, 8–17. [CrossRef]
  71. Wheeler DP, Fields SD, Beauchamp G, Chen YQ, Emel LM, Hightow-Weidman L, et al. Pre-exposure prophylaxis initiation and adherence among Black men who have sex with men in three U.S. cities: results from the HPTN 073 study. 2019;8–12.
  72. van Dijk, M.; de Wit, J.B.F.; E Guadamuz, T.; E Martinez, J.; Jonas, K.J. Slow Uptake of PrEP: Behavioral Predictors and the Influence of Price on PrEP Uptake Among MSM with a High Interest in PrEP. AIDS Behav. 2021, 25, 2382–2390. [CrossRef]
  73. Knight, L.; Hosegood, V.; Timæus, I.M. Obligation to family during times of transition: care, support and the response to HIV and AIDS in rural South Africa. AIDS Care 2016, 28, 18–29. [CrossRef]
  74. Skovdal, M.; Maswera, R.; Kadzura, N.; Nyamukapa, C.; Rhead, R.; Wringe, A.; Gregson, S. Parental obligations, care and HIV treatment: How care for others motivates self-care in Zimbabwe. J. Heal. Psychol. 2018, 25, 2178–2187. [CrossRef]
  75. Wamoyi, J.; Renju, J.; Moshabela, M.; McLean, E.; Nyato, D.; Mbata, D.; Bonnington, O.; Seeley, J.; Church, K.; Zaba, B.; et al. Understanding the relationship between couple dynamics and engagement with HIV care services: insights from a qualitative study in Eastern and Southern Africa. Sex. Transm. Infect. 2017, 93, e052976. [CrossRef]
Table 1. Distribution of sociodemographic characteristics by complete non-retention status.
Table 1. Distribution of sociodemographic characteristics by complete non-retention status.
Variables Complete non-retention status p-value
No n (%) Yes n (%)
Age groups (years) (24.7 ± 5.5) 0.008
18-24 217 (58.81) 65 (29.95) 152 (70.05)
25+ 152 (41.09) 66 (43.42) 86 (56.58)
Marital status 0.028
Never married 323(87.53) 108 (33.44) 215 (66.56)
Married currently or previously 46(12.47) 23 (50.00) 23 (50.00)
Has own children 0.354
Yes 124 (33.60) 54 (43.55) 70 (56.45)
No 245 (66.40) 129 (52.65) 116 (47.35)
Level of education 0.167
Primary and below 121 (32.79) 37 (30.68) 84 (69.42)
Secondary and above 248 (67.21) 94 (37.90) 154 (62.10)
Having Financial dependants 0.006
Yes 242 (65.58) 98 (40.50) 144 (59.50)
No 127 (34.42) 23 (25.98) 94 (74.02)
Level of social support 0.292
Inadequate 282 (76.42) 96 (34.04) 186 (65.96)
Adequate 87 (23.58) 35 (40.23) 52 (59.77)
Table 2. Distribution of sexual behavior characteristics by complete non-retention status.
Table 2. Distribution of sexual behavior characteristics by complete non-retention status.
Variable
Total N (%)		 Complete non-retention status
p-value
No n (%) Yes n (%)
Age at sex debut (15.9 ± 2.7) 0.265
<18 261 (70.73) 88 (33.72) 173 (55.28)
18+ 108 (29.27) 43 (39.81) 65 (60.19)
Type of sex at sex debut 0.152
Anal/Oral/Thighs 137 (37.12) 55 (40.15) 82 (59.85)
Vaginal 232 (62.88) 76 (32.76) 156 (67.24)
Age at first anal sex (18.7 ± 4.5) 0.080
<18 149 (40.38) 45 (30.20) 104 (69.80)
18+ 220 (59.62) 86 (39.09) 134 (60.91)
First Sex Partner 0.468
Male 118 (31.98) 45 (38.14) 73 (61.86)
Female 251 (68.12) 86 (34.26) 165 (65.74)
Steady male partner 0.031
Yes 189 (51.10) 77 (40.74) 112 (59.26)
No 180 (48.90) 54 (30.00) 126 (70.00)
History of using any lubricant 0.069
Yes 278 (76.37) 107 (37.81) 176 (62.19)
No 85 (23.63) 23 (27.06) 62 (72.94)
Frequency of lubricant use 0.001
Never used 85 (23.35) 23 (27.06) 62 (77.94)
Sometimes 90 (24.73) 22 (24.44) 68 (75.66)
Always 194 (51.92) 85 (44.04) 109 (65.96)
Lubricant last time had anal sex 0.011
Never used 85 (23.35) 23 (27.06) 62 (72.94)
Yes 241 (66.66) 98 (40.66) 143 (59.44)
No 42 (11.54) 9 (21.43) 33 (78.57)
Ever paid for having oral or anal sex 0.134
Yes 276 (74.45) 92 (33.33) 184 (66.77)
No 93 (25.55) 39 (41.94) 54 (58.06)
Forced sex lasts 12 months 0.551
Yes 139 (37.77) 52 (37.41) 87 (62.59)
No 230 (62.33) 79 (34.35) 151 (65.65)
Arrested last 12 months 0.405
Yes 57 (15.45) 23 (40.35) 34 (59.65)
No 312 (84.55) 108 (34.62) 204 (65.38)
Ever tested for HIV 0.014
Yes 236 (64.84) 96 (40.00) 144 (60.00)
No 128 (35.16) 35 (27.13) 94 (72.87)
Easy access to male condoms 0.768
Yes 284 (76.96) 102 (35.92) 182 (64.08)
No 85 (23.04) 28 (34.15) 54 (65.85)
Self-perceived HIV risk 0.784
High risk 262 (71.97) 97 (36.47) 169 (63.53)
Medium risk 32 (08.79) 10 (30.30) 23 (69.70)
Low/no risk 62 (17.03) 22 (35.48) 40 (64.52)
Experienced stigma 0.534
<=26 (Low) 188 (50.95) 70 (37.23) 118 (62.77)
27-38 (Moderate) 129 (34.96) 46 (35.66) 83 (64.34)
>=39 (High) 52 (14.09) 15 (28.85) 37 (71.15)
Comprehensive HIV knowledge 0.239
No 259 (70.19) 87 (33.59) 172 (66.41)
Yes 110 (29.81) 44 (40.00) 66 (60.00)
PrEP knowledge 0.641
Low 245 (66.40) 89 (36.33) 156 (63.77)
High 124 (33.60) 42 (33.87) 82 (66.13)
PrEP stigma 0.146
Low 291 (79.95) 99 (33.67) 195 (66.33)
High 73 (20.05) 32 (42.67) 43 (57.33)
PrEP self-efficacy 0.486
Low (<24=24) 101 (27.47) 33 (32.67) 68 (67.33)
High (>24) 268 (72.53) 98 (36.57) 170 (63.43)
Table 4. Multiple adjusted models of modified Poisson regression analysis of the association between mental disorder symptoms and complete no-retention.
Table 4. Multiple adjusted models of modified Poisson regression analysis of the association between mental disorder symptoms and complete no-retention.
Mental Disorder symptoms Crude
Prevalence ratio
cPR (95% CI)		 p-value Adjusted
prevalence ratio
aPR (95% CI)		 p-value
Depression (Model 1)
Yes 0.89 (0.71–1.13) 0.308 0.95 (0.78–1.17) 0.634
No Ref. Ref.
Anxiety (Model 2)
Yes 0.95 (0.73–1.23) 0.683 0.98 (0.78–1.22) 0.825
No Ref. Ref.
Alcohol use disorder (Model 3)
Yes 1.02 (0.88–1.18) 0.638 1.05 (0.89–1.25) 0.566
No Ref
Any mental disorder (Model 4)
Yes 0.95 (0.81–1.10) 0.481 0.97 (0.83–1.17) 0.725
No Ref. Ref.
NB: See the adjusted possible confounder for each model in the text above, regarding the association between mental disorders and complete non-retention.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
Copyright: This open access article is published under a Creative Commons CC BY 4.0 license, which permit the free download, distribution, and reuse, provided that the author and preprint are cited in any reuse.
Prerpints.org logo

Preprints.org is a free preprint server supported by MDPI in Basel, Switzerland.

facebook logotwitter logolinkedin logo
bsky
MDPI Initiatives
Important Links
Subscribe

Accessibility

Disclaimer

Terms of Use

Privacy Policy

Privacy Settings

© 2026 MDPI (Basel, Switzerland) unless otherwise stated