Characterising People who inject drugs, and association with HIV infection: A Situation Analysis in Kampala city, Uganda

Study design & setting

This cross-sectional survey was implemented from August 2023 to January 2024 at selected hotspots in Kampala City. The study was conducted with the active involvement of the Uganda Harm Reduction Network(UHRN), a PWID community-based organization that provides harm reduction services to PWID at different drop-in centers and hotspots in Kampala.

Mental health illnesses

In total, 243 individuals (48.7%) disclosed having at least one underlying mental illness, among whom 55 (11%) were undergoing treatment, and 64(12.8%) reported more than one existing mental condition. Among those who reported mental illness, 48 (9.6%) acknowledged co-existing anxiety, with five of them receiving treatment; 154 (30.8%) indicated co-existing depression, with 39 (7.8%) undergoing treatment. Furthermore, out of the 33(6.6%) individuals who reported co-existing Attention Deficit Hyperactivity Disorder (ADHD), three were receiving treatment (Figure 1). Other reported mental illnesses included schizophrenia (3 cases), bipolar disorder (1 case), and personality disorder (4 cases) (Fig. 1).

• Download figure
• Open in new tab

Figure 1: Mental Health Illnesses(self-reported) among people who use drugs in Uganda (n=499)

Physical health illnesses

In terms of physical health, a total of 164(32.9%) participants reported experiencing fever within the three months preceding the interview, while 162(32.5%) mentioned having a cough, 21(4.2%) reported frequent urination, 34(6.8%) experienced painful urination, and 10(2%) had groin wounds. Additionally, 110 (22%) individuals had been treated for malaria, 75(15%) for sexually transmitted infections (STIs), 18 (3.6%) for urinary tract infections (UTIs), and 32(6.4%) for upper respiratory tract infections (URTIs). Five participants had received treatment for tuberculosis, and 20(4%) for skin diseases.

Drug use-related characteristics

The median duration of drug use was .4 years (IQR: 3.9-13.8 years). Four hundred and twelve participants (82.6%), revealed that they were introduced to drug use by close friends or relatives, with 72.5% citing peer influence as the primary reason for continued drug use. Other reported reasons included coping with the loss of a loved one in 57 individuals (11.4%), coping with work-related stress(six participants), and personal illness (six) as reasons for continued drug use. Notably, 354 participants (70.9%) were categorized as PWID based on their reported history of injection drug use. Among them, 178, (50.3%) had last injected more than a month before the interview, 62 (17.5%) within the last month, 61 (17.2%) within the last week, and 21 (5.9%) on the day of the interview. Regarding injection frequency, 199 participants (57.2%) reported occasional injections, 54 (15.5%) injected at least monthly, 67 (19.3%) injected weekly, and 28 (8%) injected daily. Of the 55 participants who reported being on Medications for Opioid Use Disorder (MOUD), 35 (%) injected drugs only occasionally (Table 2).

HIV Risk behavior

Key HIV risk behaviors were identified, particularly related to both drug-injection and sexual risky practices in the past six months. Sharing of injecting equipment was a significant concern, with 33 participants (9.3%) admitting to using a shared needle or syringe, and a similar proportion acknowledging the sharing of other injecting materials. Additionally, 22 participants (6.2%) reported a risky practice of flashing blood, and using unsterilized syringes and needles, either on themselves or another person. Despite these risky behaviors, a majority of 268 participants (75.7%) stated that obtaining clean needles and syringes was easy, which may potentially mitigate some risks associated with injecting drug use(Table 3).

Drug use and sexual-related HIV risk factors

Regarding sexual behaviors, 189(53.4%) participants reported engaging in sexual encounters with non-regular partners within the last three months, indicating a potential pathway for HIV transmission beyond injecting drug use. Only (19.8%) knew the HIV status of their non-regular partners, suggesting a lack of awareness regarding potential HIV risks in sexual encounters. Condom use during sexual encounters with non-regular partners was low, with only 32.6% and 25.2% of participants reporting consistent condom use during sex with each non-regular partner and all their last three sexual encounters, respectively. Transactional sex was prevalent, with 128 participants (67.4%) engaging in paid or paid-for sexual encounters, and 45.3% exchanging sex for drugs. Additionally, 191(53.9%) participants reported using drugs before engaging in sexual encounters with others, indicating a heightened risk of HIV transmission due to impaired judgment and decision-making associated with drug use(Table 3).

HIV positivity among people who inject drugs

Overall, HIV status was ascertained in 354(70.9%) participants, with 13 (3.7%) self-reported to be HIV positive. HIV positivity was higher among females (8.4%) compared to males at 1.9%. A higher proportion of young persons (18-24 years) were HIV-positive (7.1%) compared to older men aged 25-34 years (2.1%) and those above 34 years (4.5%). The proportion of HIV positives was higher (16.7%) among non-Ugandans compared to Ugandan participants (3%). There was no variation in positivity based on participants’ marital status and income.

Regarding HIV risk behaviors related to drug use, HIV was more prevalent among participants who reported difficulty in accessing clean injection materials. Approximately 5.3% of those with difficult-to-access clean injection materials compared to 1% of those with ease of accessing clean injection materials reported to be HIV positive. The reported proportions of HIV positives did not vary based on duration of injection, and frequency of injecting.

In terms of sexual risk behaviors, HIV occurrence was higher among participants who reported engaging in sexual activities with non-regular partners. No variation in HIV status was reported based on other sexual risk factors such as consistent condom use, awareness of partner’s HIV status, paying money or other goods for sex, and taking drugs for sex(Table 3).

Association between socio-demographics, risk behaviors, and HIV

At bivariate analysis, female sex, non-Ugandan by nationality, difficulty in accessing sterile injection materials, and a history of taking drugs before sex, were more likely to be HIV positive. The age of the participant, having sex with a non-regular partner in the last three months, and consistent condom use with non-regular partners were not statistically significant to an individual’s HIV status (Pavlues>0.05). Multivariate analysis showed that females compared to the males were over four times more likely to report HIV-positive status {Adjusted Odds ratio(AOR)=3.8, 95% confidence interval (CI):1.01-15.5, P value=0.049}. Participants who reported having difficulty in accessing sterile injection materials were likely to be HIV positive, {AOR=4.69, 95% (CI): 1.15-19.16, p=0.031}. The other factors showed no significant association with the HIV status of the participants (Table 4).

Mental health

We found a high self-reported occurrence of mental health disorders, coupled with limited access to treatment services. Nearly half of the participants reported a diagnosis of co-existing mental disorder, with only one in ten were receiving treatment. Depression, Anxiety, and Attention Deficit Hyperactivity Disorder were the most prevalent co-morbidities. Schizophrenia, post-traumatic stress disorder (PTSD), and attention deficit hyperactivity disorder (ADHD) were less common in this population. The prevalence of mental disorders in this study was high compared to what was found in a systematic review of 24 studies from Uganda, which reported a prevalence of 24.2%, in adults (27, 28). In the same review, the combined prevalence of anxiety disorders and depressive disorders was approximately one in four adults. Globally, the prevalence of mental disorders has been found to vary widely based on the method of ascertainment, with recent studies reporting a prevalence of depressive disorders varying from as low as 3·4% to a high 66% in the general population (2). Despite potential social desirability bias associated with self-reported data, our findings suggest that the escalating injecting drug use is likely exacerbating the ongoing mental health crisis in Uganda. Diagnosing mental disorders can be challenging, particularly in low-resource settings like Uganda where the inadequate human resources to address the growing mental disorder and drug use disorder burden are limited(29). The signs and symptoms of acute intoxication and withdrawal syndrome are similar to those of other psychiatric disorders, complicating the accurate diagnosis of these conditions. Furthermore, mental disorders can be risk factors for and can exacerbate, substance use, and vice-versa, creating a complex interplay between the two(30). This interplay highlights the need for integrated approaches to address both mental health disorders and opioid use disorders in Uganda.

Physical health profiles

Injecting drug users face a significant burden of physical health issues, including infectious diseases such as malaria, sexually transmitted infections (STIs), and HIV. The reported prevalence of malaria treatment (22%) is consistent with the high malaria burden in Uganda, as the leading cause of morbidity and mortality (31). The prevalence of STIs (15%) and HIV (4.38%) underscores the intersection of substance use and sexual health risks among this population. These findings align with other studies, which have similarly reported elevated rates of STIs, HIV and other communicable diseases, among people who use drugs. The high vulnerability has been associated with some factors including high-risk sexual behaviors associated with drug use, limitations in accessing healthcare services, and hygiene practices, among others (32, 33). The presence of tuberculosis (TB) and skin diseases among participants highlights the broader health challenges faced by PWID (25). Overall, these findings emphasize the need for integrated healthcare interventions that address both infectious and non-communicable diseases among injecting users in Uganda and similar settings.

HIV Prevalence

The prevalence of HIV among injecting drug users was 4.38%, far less than the 33% reported by the Uganda AIDS Commission and Uganda Ministry of Health (34). The recent crane survey and two other studies found the HIV prevalence among PWID in Uganda to be ranging between 3.6% to 17% (35–37). In this study, we primarily relied on reported participants’ self-report on their HIV status, with only one in five person who had not tested in the last six months agreeing to test. It is possible that most participants who reported a negative test may have tested well before the previous three months, and some may have reported a negative HIV test due to stigma. Stigma about HIV-positive status is highly prevalent in this population, and often influences the uptake of HIV testing services(38, 39). Even with the low prevalence finding, HIV prevalence is four times higher among females than males, consistent with studies in Tanzania where the HIV-positive sero-status was three times higher in females compared to males (40). Risky injection practices and sexual behaviors contribute to the heightened risk of HIV transmission among injecting drug users(25, 41). Sharing of injecting equipment and engaging in unprotected sexual encounters with non-regular partners were common. The high frequency of transactional sex and the use of drugs before sexual encounters further compound the risk of HIV transmission. However, despite the high occurrence of risk behaviors, our study found a low HIV prevalence, likely due to the self-reported nature of HIV status. Stigma and fear may prevent persons with injecting drug use disorders from getting tested or disclosing their status. Low HIV status awareness (only one in five) among injecting drug users also suggests inadequate coverage of HIV prevention services(41). At the time of data collection, only one MOUD clinic existed in Kampala city. This highlights the urgent need to expand harm reduction services, including HIV testing, to reach more injecting drug users in Uganda.

Risk factors for HIV infection

The study reveals the existence of risky drug injection and sexual practices in Uganda with the potential to increase the number of new HIV infections. The common risk drug injection practices reported include sharing needles and syringes, sharing mixing equipment, flushing blood, and reusing syringes multiple times. These findings reinforce earlier reports from studies in Uganda (17, 37). Prevoius studies associated highly prevalent drug risk behavior with the unaffordability of new syringes and the nadir existence of harm reduction programs in Uganda. Regarding to risky sexual practices, over half of the participants reported having sex with a nonregular partner, three of ten reported inconsistent condom use, and over two-thirds of the participants were engaging in transactional sex. These factors heighten the vulnerability of PWID to HIV infection (42). While taking drugs before sex was associated with an increased risk of HIV, this association was not significant in multivariate, similar to other sexual risk factors. A previous study by the Community Health Alliance Uganda (CHAU) examined discussed three main HIV risk perspectives: impaired judgment leading to unprotected sex, frequent sexual encounters with non-regular partners, and reduced libido due to drug use, which lowers sexual activity(37). Additionally, people who inject drugs may be forced to sell sex to acquire money to buy drugs (32, 43). In our study, nearly seven in ten of those who had multiple sexual encounters, paid money or other commodities in exchange of sex with another person, and nearly half reported having had sex in exchange for drugs.

Another crucial risk factor deserving attention is the status of being a non-Ugandan, primarily refugees, who may have recently imigrated. The risk of being HIV positive was found to be five times higher among non-nationals compared to Ugandans. This aligns with findings from studies conducted in Dar-es-Salaam and the United States, where recent immigrants exhibited an increased risk of HIV infection (6, 44). We attribute this heightened risk among non-nationals to various factors, including unawareness of safer needle use practices, financial constraints hindering the purchase of new needles, and exposure to circumstances that increase the likelihood of engaging in risky sex work. The small sample size and missing data on key parameters may partially explain the failure to demonstrate an association between injection drug risk and the HIV status. Despite efforts to ensure the confidentiality and safety of participants during interviews, it was not uncommon for participants to decline to respond to particular essential questions or take an HIV test. Another factor that we didn’t find significant was the frequency of injection; in Uganda, most drug users only injected occasionally. This affects the risk of exposure to HIV and programming for harm reduction interventions such as clean needle and syringe distribution. Nonetheless, our study finding reinforces current evidence on the relation between HIV serostatus, and injecting drug and sexual risk practices. Two studies conducted in Uganda found that participants reused syringes in some instances up to 2-4 times before disposal (17, 45).

Limitations

Selection bias may have emerged due to increased security enforcement in preparation for a Non-Aligned Movement (NAM) Summit meeting held in Kampala, during January 2024, coinciding with our data collection period. This security exercise drove more PWID into hiding, reducing our access to some community members. Selecting initial seeds on a first-come basis likely favored accessible or motivated individuals, leading to an unrepresentative sample. This may underrepresent those more stigmatized, severely addicted, or with poor health and less access to services, skewing results and underestimating the true prevalence of risk behaviors and mental and physical health conditions.

In some instances, participants declined to answer or did not respond to a question. Although this could have been because individuals deliberately preferred not to answer, we can’t rule out the fact that cognitive impairments, mood disorders, or other mental health issues suffered by people suffering from opioid use disorder, could compromise the ability to recall and report information accurately.

Relying on self-reporting by participants could have introduced recall bias, as participants might not accurately remember past events or might selectively report on such events, leading to misrepresentation of the true measures. To minimize this bias, the questions were designed with a maximum recall duration of three months, pretested, and simplified to reduce cognitive load and facilitate easier recall of past events. Additionally, interviews were rescheduled based on participants’ availability to enhance recall accuracy, avoid distractions, and allow for better concentration and memory recall. Furthermore, while measures were taken to ensure confidentiality and safety, the possibility of overreporting or underreporting drug and sexual risk behaviors remains, as participants may have been influenced by social desirability bias stemming from internal stigma. The extended data collection period(six months) allowed us to reach and enroll injecting drug users of different categories. Collecting data at 38 community hotspots where most injecting drug users congregate helped ensure diversity. Using ODK enabled real-time data correction, increasing accuracy. Based on these adjustments, we believe these results are valid and representative of injecting drug users in Kampala.

Conclusion

Our study provides a comprehensive insight into the socio-demographic, mental, physical health, and HIV-associated risk of persons with drug user disorders in Kampala, Uganda. The findings indicate a significant vulnerability to injecting drug use, mental health disorders, and high-risk behaviors that predispose this population to HIV infection. Despite a low HIV prevalence compared to previous estimates, the interplay between drug use, risky injecting practices, and sexual behaviors suggests an urgent need for targeted interventions to address these intertwined challenges. Implementing a multi-sectoral approach that combines, harm reduction, physical healthcare, and mental health services is crucial for improving the well-being of persons who inject illicit drugs, and reducing the burden of HIV and other associated health issues in Uganda.