Introduction

Social and structural vulnerabilities, including stigma and discrimination, increase HIV acquisition risk and impede access to HIV services for key populations (KP) including sex workers, gay, bisexual, and other men who have sex with men (MSM), transgender individuals such as transgender women (TGW), and people who inject drugs (PWID) [1–3]. In sub-Saharan Africa, half of all new HIV infections occur among KP and their sexual partners [4]. Risk of acquiring HIV is 28 times higher among MSM than adult men, 14 times higher among TGW than adult women, and 35 times higher among PWID than people who do not inject drugs [4].

As COVID-19 has reinforced health inequities among marginalized groups, social and structural barriers affecting HIV risk and access to services for KP could be worsened by the COVID-19 pandemic or other emerging pandemic threats. Disruptions in health commodities and routine health services for HIV, as have been documented during periods of COVID-19 lockdown across Africa and Asia [5], may disproportionately affect KP given their high HIV burden. While there is no clear evidence that people living with HIV have higher risk of SARS-CoV-2 infection, people living with HIV may be more likely to suffer severe illness from COVID-19 and have higher in-hospital mortality than people not living with HIV based on global hospitalization data across 38 countries [6]; given their HIV burden, KP may disproportionately experience COVID-19 complications and adverse outcomes relative to the general population.

In Zambia, available data on population-level HIV prevalence and population size estimates of KP are limited and there are few legal protections for KP. Consensual same-sex conduct is criminalized [7], and punitive drug and drug equipment laws pose challenges to harm reduction service delivery [8,9]. Moreover, legal structures and cultural contexts contribute to stigma and discrimination and increase risk of violence and abuse for KP [10].

From March-September 2020, in response to the COVID-19 pandemic, Zambia instituted a partial lockdown and closed establishments including sports clubs, nightclubs, bars, casinos, and restaurants, as well as implemented social distancing measures and restrictions. Government mitigation strategies or mandates in response to COVID-19 or future pandemic threats, like those implemented in Zambia, have the potential for further stigmatization of KP, including police targeting and loss of income, and may be difficult to adhere to for KP, particularly those who are homeless or those with overcrowded living conditions [11–13].

We qualitatively explored the impact of COVID-19 and government restrictions to manage COVID-19 on HIV services and sexual and substance use behaviors among MSM/TGW and PWID as part of a formative assessment to inform two upcoming biobehavioral surveys (BBS), the first respondent-driven sampling (RDS) surveys to be conducted with these groups, in Zambia.

Materials and methods

Data were collected from November-December 2020, two months after the partial lockdown was lifted. Survey towns included Lusaka (MSM/TGW, PWID), Livingstone (MSM/TGW, PWID), Ndola (PWID), Solwezi (MSM/TGW), and Kitwe (MSM/TGW), Zambia. MSM/TGW survey towns were selected where prior MSM population size estimates were highest according to an unpublished study [14]. PWID survey towns were selected to include three highly populous cities in Zambia, which have a high number of people who use drugs according to population size estimates from the same unpublished study [14]. In each survey town, 3 focus groups discussions (FGDs) and 15 in-depth interviews (IDIs) with each KP were conducted. Overall, 45 PWID and 60 MSM/TGW participated in IDIs and 70 PWID and 89 MSM/TGW participated in FGDs.

Participants were recruited using purposive sampling techniques and through snowball sampling to ensure diversity across demographic characteristics. Recruitment efforts were supported by KP community mobilizers and KP civil society organizations. MSM/TGW were eligible to participate if they were assigned male at birth, self-reported anal or oral sex with a man in the past 6 months, were at least 16 years of age, lived in the survey town for at least 3 months, spoke English or any other designated local language, and were able to provide informed consent. PWID were eligible to participate if they self-reported drug injection for non-medical purposes in the past month, were at least 16 years of age, lived in the survey town for at least 3 months, spoke English or any other designated local language, and were able to provide informed consent.

Where possible, FGDs for each KP group were stratified by participant characteristics (i.e., gender and age) to encourage participants to freely share ideas and perceptions. After a brief introduction of the upcoming BBS, interviewers obtained verbal informed consent from each eligible participant. As part of the consent process, potential benefits of participating in the formative assessment were discussed, including potential to ensure KP perspectives and experiences informed BBS methods and KP service delivery. The legal age of consent in Zambia is 16 years hence parental/guardian consent was not sought nor required for participation. However, referrals to local resources were provided to participants less than 18 years who reported sex work or who reported being a victim of violence or trafficking. FGD participants were instructed not to use their name, the name of other participants, or other KP that could suffer negative consequences if identified through FGDs. IDIs and FGDs were conducted by two trained interviewers (a moderator and a note taker) with the aid of semi-structured IDI/FGDs guides at separate, private locations for MSM/TGW and PWID. Verbal informed consent and IDIs/FGDs were conducted in the participant’s preferred language (Chinyanja, Chitonga, Cibemba, Kikaonde, Silozi, or English). All participants were reimbursed Zambian kwacha equivalent of $13 to cover transportation costs and time. Refreshments including a bottle of water, a snack, and/or soft drink were offered and provided to participants.

IDI and FGD guides were adapted from formative assessment guides used with MSM and others assigned male who have sex with men in neighboring Zimbabwe [15]. During IDIs/FGDs, interviewers generated memos as the discussion unfolded to help formulate follow-up questions and probes. Interviewers documented their impressions about the sessions, its main themes and participant quotes using a deductive interview notes template according to IDI/FGD guide categories. Following IDIs/FGDs, teams also held debrief sessions to discuss emerging themes. Data from the deductive notes template were subsequently populated into an Excel matrix; categories were analyzed separately for each KP group, survey town, and data collection method by two team members [16]. For each KP, across case and within case comparisons were made and additional comparisons by age and gender were reviewed for select categories [17]. Themes presented reached data saturation—the point at which further inquiry did not yield any additional information [18,19].

Ethical and administrative approvals were received from the Columbia University Institutional Review Board, the Tropical Diseases Research Centre Ethics Review Committee, and the Zambia National Health Research Authority. The protocol was also reviewed in accordance with the United States Centers for Disease Control and Prevention (CDC) human research protection procedures and was determined to be research, but CDC investigators did not interact with human subjects or have access to identifiable data or specimens for research purposes. Additional information regarding the ethical, cultural, and scientific considerations specific to inclusivity in global research is included as (S1 Questionnaire).

Results

Impacts of COVID-19 among MSM and TGW

Duration of sessions ranged from 55–150 minutes for IDIs and 90–230 minutes for FGDs. MSM/TGW IDI participants ranged in age from 16–55 years (Table 1). More than a quarter of participants identified as TGW (28%). Most had received a secondary school education (48%) and worked in the informal sector (57%). FGDs included between six to nine participants. Nine FGDs consisted of participants 25 years or older (75%) and three FGDs consisted of participants less than 26 years (25%).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Demographic characteristics among MSM/TGW IDI participants, Zambia, 2020.

https://doi.org/10.1371/journal.pone.0289007.t001

Six primary themes related to the impacts of COVID-19 on MSM and TGW emerged from IDIs and FGDs: economic impact, mental health, HIV services, stigma and discrimination, sexual behavior, and, relevant to the objectives of the formative assessment, willingness to participate in the upcoming BBS. Illustrative quotes for themes and sub-themes are provided in Table 2.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Themes, sub-themes, and illustrative quotes on the impacts of the COVID-19 pandemic among MSM and TGW, Zambia, 2020.

https://doi.org/10.1371/journal.pone.0289007.t002

Across towns, MSM and TGW described severe economic impacts resulting from the COVID-19 pandemic, including loss of formal employment and loss of income for those who were self-employed and/or working as sex workers. Movement restrictions and the temporary closure of public places such as bars and restaurants impeded business for participants with formal employment and limited opportunities for MSM and TGW sex workers to meet clients.

COVID-19 contributed to adverse mental health outcomes. MSM and TGW in all cities and across data collection methods reported experiencing anxiety due to isolation and/or quarantine during the pandemic. Participants reported feeling isolated from family and friends both physically and emotionally. The economic impacts of COVID-19 further exacerbated anxieties for participants. Participants reported experiences of depression and anxiety about meeting financial commitments and their ability to afford necessities. Fear of SARS-CoV-2 exposure was reported as an additional anxiety for MSM and TGW during the pandemic; participants reported this fear motivated them to change behavior, including remaining in their homes, avoiding healthcare settings, and influencing their sexual behaviors.

All participants across data collection methods reported impacts of COVID-19 on HIV service delivery. Participants reported limited supply of condoms and lubricants at health facilities and barriers to HIV testing due to COVID-19-related disruptions and operational changes; prioritization of COVID-19 patients at health facilities and being turned away when presenting with non-COVID-19 health issues; avoidance of HIV care-seeking due to fear of SARS-CoV-2 exposure; and antiretroviral therapy (ART) stockouts. Limited access to condoms also led to increased engagements in condomless sex. While participants described adverse impacts of COVID-19 on access to HIV services, they also described its benefits, including the implementation of innovations and adaptions to HIV service delivery in response to the pandemic such as HIV self-testing, mobile clinics, and appointment reminders.

In addition to COVID-19-related barriers, MSM/TGW reported experiencing stigma and discrimination and poor-quality health services within public health facilities irrespective of COVID-19 due to non-MSM/TGW friendly staff across cities. MSM/TGW participants reported that negative attitudes from healthcare workers and lack of confidentiality deterred them from accessing services at public facilities. Healthcare workers were reported to routinely stigmatize and mistreat patients who identified as MSM/TGW. As a result, many MSM/TGW avoided visiting public health facilities and preferred accessing health services at private clinics and non-governmental organizations before and during COVID-19. Those who accessed services at public facilities reported feeling comfortable interacting with staff who were a part of the KP community; many reported not disclosing anal sex behaviors with staff to prevent stigmatization and/or discrimination.

Restrictions in movement and closure of usual meet-up spots due to COVID-19 impeded opportunities to meet sex partners and clients for MSM/TGW who sold sex. In response, participants used social media to meet sexual partners/clients and met in homes/private locations. While some participants reported an increase in sexual behavior due to boredom, most reported reduced sexual activities due to fear of SARS-CoV-2 exposure, movement restrictions, and closure of lodges/bars.

Participants reported that restrictions in movement and limited socialization in public areas where stigma, discrimination, and violence ‘typically’ occur resulted in reduced experiences of stigma and discrimination in public settings during the pandemic. Yet, mobility restrictions had no impact on reducing experiences of stigma from family and/or household members as MSM and TGW spent longer periods of time in the home because of the lockdown.

In response to prompts around interest and willingness to participate in the upcoming BBS in the context of COVID-19, most participants reported that COVID-19 was unlikely to impede participation; participants reported COVID-19 to be the “new normal” and would not impact their interest or willingness to enroll in the BBS. Most participants reported that transmission mitigation strategies such as the distribution of masks and sanitizer and implementation of social distancing at survey sites may encourage MSM and TGW to participate in the survey. It was reported that ensuring transmission precautions are in place would promote participants to feel safe coming to the survey site and engaging with survey staff. While most participants felt COVID-19 would not impact their willingness to participate, some reported that a worsening epidemic and/or increased government containment measures may impact participation of their peers. Fear of SARS-CoV-2 infection and mobility restrictions were cited as factors that may impede recruitment.

Impacts of COVID-19 among PWID

Duration of sessions ranged from 80–180 minutes for IDIs and 60–190 minutes for FGDs. PWID IDI participants ranged in age from 18–51 years (Table 3). Most PWID identified as male (82%, Table 3), had received secondary school education (49%), and worked in the informal sector (73%). Of the nine FGDs, four consisted of only males, one consisted of only females, three consisted of males and females, and one consisted of males and TGW. Across towns, the median number of participants in FGDs was 8 (range: 6–9).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Demographic characteristics among PWID IDI participants, Zambia, 2020.

https://doi.org/10.1371/journal.pone.0289007.t003

Five primary themes related to the impacts of COVID-19 on PWID emerged from IDIs and FGDs: economic impact, mental health, HIV services, injection practices, and willingness to participate in the upcoming BBS. Illustrative quotes for themes and sub-themes are provided in Table 4.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 4. Themes, sub-themes, and illustrative quotes on the impacts of the COVID-19 pandemic among PWID, Zambia, 2020.

https://doi.org/10.1371/journal.pone.0289007.t004

COVID-19 restrictions and border closures led to unemployment and loss of income for many participants, as well as shortages and increased price of drugs, needles, and syringes, which contributed to emotional distress. Participants across towns and data collection methods reported increased needle-sharing and re-use of needles due to COVID-19 economic effects and limited drug supply. Some participants also reported reduced frequency of drug injection due to limited drug supply.

Participants reported not accessing HIV services due to COVID-19-related fears, long queues, limited operating hours, reassignment of healthcare workers, crowded conditions, and mask and COVID-19 testing requirements. Chronic issues for accessing services irrespective of the pandemic were also reported, including stigma and discrimination by healthcare workers and transportation costs. Others reported accessing HIV services, including pre-exposure prophylaxis (PrEP), with no difficulty.

Like MSM/TGW, PWID were willing to participate in the upcoming BBS despite the COVID-19 pandemic; participants echoed that COVID-19 was the “new normal” and indicated it would have little impact on their willingness to participate in the survey. Participants were motivated to engage in the BBS as they felt it was important for their health. Many participants described the importance of adhering to transmission mitigation strategies at the survey site, including wearing masks and use of hand sanitizer.

Discussion

Findings from this formative assessment highlight areas in which COVID-19 and government restrictions in response to COVID-19 have impacted KP in Zambia, including impacts on sexual risk behaviors, safe injection practices, and access to HIV services, and can inform both KP programming as well as surveillance efforts among KP in the context of COVID-19 and/or other emerging pandemic threats. Importantly, while the focus of this analysis was primarily to explore COVID-19 impacts, results also elucidate structural barriers to HIV service delivery for KP irrespective of COVID-19 such as stigma and discrimination and violence. Data on KP in Zambia are sparse likely due to the punitive and non-protective legal and social context, and this assessment provides evidence of areas where improvements are needed to align programs with UNAIDS 2025 10-10-10 targets for societal enablers. To remove social and legal impediments towards an enabling environment these targets aim to ensure less than 10% of countries have punitive legal and policy environments that deny access to justice, less than 10% of KP and people living with HIV experience stigma and discrimination, and less than 10% of women, girls, people living with HIV and KP experience gender inequality and violence [20].

Reduced access to HIV prevention, care, and treatment services among KP during COVID-19 has been documented in other settings globally, including in other low- and middle-income countries [21–23]. Like findings elsewhere [21–23], KP in this formative assessment experienced barriers accessing HIV services, including HIV testing, as well as limited supply of condoms and lubricants and drug stockouts at the health facility. Both MSM/TGW and PWID described barriers related to health facility operational changes and COVID-19 requirements as well as avoidance of HIV care seeking due to fear of potential SARS-CoV-2 exposure at the health facility. As the COVID-19 epidemic evolves in Zambia or future pandemic threats are posed, scale-up of community-based work and improvements in supply chain may be important to ensure continued coverage of HIV services during a public health emergency.

COVID-19 has also disrupted illicit supply chains. These disruptions have impacted drug supply, access to safe injection equipment, and drug injection practices for PWID in some countries [24]. In Zambia, loss of income and lack of affordable needles led to an increase in unsafe needle use and sharing practices, highlighting a need for harm reduction strategies and PWID-tailored services. Few organizations provide services specifically for PWID in Zambia, and those that do are limited in their ability to provide comprehensive harm reduction services [9]. Ensuring an enabling environment for organizations to provide harm reduction services for PWID, particularly in the context of COVID-19 or other emerging pandemic threats, warrant prioritization. Additionally, human rights-based training approaches for healthcare workers may reduce stigmatization at the health facility, a chronic barrier to care described by our participants.

MSM/TGW participants highlighted the implementation of innovations in response to COVID-19; KP organizations may consider scaling up these types of innovations, including HIV self-testing, mobile clinics, and appointment reminders, as well as scaling up or strengthening other innovative delivery approaches such as multi-month dispensing of PrEP and ART, community ART groups, and other differentiated service delivery models, telehealth support across the cascade of services, and mailed testing and prevention commodities [25]. Moreover, as MSM/TGW in this formative assessment adapted to meet sexual partners/clients via social media in response to COVID-19 restrictions, KP organizations may utilize similar platforms to engage MSM/TGW and promote HIV services in the context of COVID-19 or other emerging pandemic threats; this may be appropriate given the legal context and can reduce risk of exposure.

In this formative assessment, COVID-19 had little impact on KP’s willingness to participate in the upcoming BBS. While findings that KP can successfully be recruited for BBS participation are promising, considerations for implementing BBS among KP in the context of COVID-19 or future pandemic threats must extend beyond willingness to participate. Adherence to transmission mitigation strategies including social distancing and mask wearing is important as emphasized by participants in this study; BBS methods may warrant adaptations including use of audio computer-assisted self-interview, limits in the number of KP visiting data collection sites, use of appointments in lieu of walk-ins, routine testing of BBS staff, and/or outdoor interview administration to minimize potential for transmission, among others. Moreover, as the COVID-19 epidemic or future epidemic threats evolve, severity of restrictions during BBS data collection may fluctuate which may impact participants’ willingness to participate, their ability to recruit, and/or their self-reported personal network sizes. As network size is used to generate weights for BBS using RDS, weighting approaches for RDS surveys conducted in the context of COVID-19 or future pandemic threats may require adjustment. Additionally, as described in this assessment, COVID-19 had impacts on access to HIV services, sexual behavior, and injection practices among KP and BBS estimates should be interpreted with this considered.

The primary limitation of this study is that data were not audio recorded or transcribed due to local ethical concerns related to participant confidentiality. We sought to overcome this limitation by providing a five-day intensive didactic training on qualitative research methods to all formative assessment staff and assigning two trained interviewers to document IDIs/FGDs sessions including observations and verbatim participant quotes. Moreover, while this is a limitation, the absence of transcription in response to confidentiality concerns have been documented in other formative assessments in similar contexts [15], and results remain important contributions to the literature nonetheless. Moreover, participants in this formative assessment were recruited with support from KP community mobilizers and civil society organizations, and results represent perspectives from individuals who were purposively recruited. Despite limitations, we feel these results are necessary to inform programming for and surveillance of MSM/TGW and PWID in Zambia, where little information on these KP have been published.

Taken together, findings elucidate COVID-19’s impacts on sexual risk behaviors, safe injection practices, access to HIV services, and willingness to participate in BBS for these groups, and have important programming implications. KP overall reported experiencing barriers accessing HIV services due to COVID-19 as well as structural barriers accessing care. MSM/TGW reported limited supply of condoms and lubricants at health facilities which led to increased engagements in condomless sex, and PWID attributed unsafe needle use and sharing to loss of income and lack of affordable needles during pandemic-related restrictions. Participants also noted innovations to address barriers to care in response to COVID-19. To maintain gains in the HIV response in this context and in the context of future pandemic threats, strengthening harm reduction strategies and improvements in access to HIV services, through scale-up of innovations, are and will continue to be necessary.

Supporting information

Acknowledgments

We thank all formative assessment participants for their time and valuable feedback. We appreciate the support from our partners as well as the contributions from the Zambia Survey Advisory Group including representatives from CDC, NAC, TDRC, ICAP, and the Zam-KPC.