Mental illnesses are more prevalent in people living with HIV (PLHIV) than in the general population, highlighting the need to integrate mental health services into primary healthcare (PHC). Health workers are key players in integrating these services.
This study assessed healthcare workers’ perceptions, opinions, and experiences on barriers to integrating mental health services into antiretroviral therapy (ART) services.
The study was conducted in public health facilities in Windhoek, Namibia.
A descriptive mixed-methods study using a convergent parallel design was conducted among 44 healthcare workers (
The study identified several barriers, including inadequate facility space, limited knowledge and skills for managing mental illness, insufficient in-service training, a lack of guidelines and standard operating procedures, and staff shortages.
Significant structural and workforce-related barriers hinder effective mental health screening, diagnosis, and treatment for PLHIV in PHC settings. Addressing these challenges requires facility restructuring, workforce training, task shifting, and strong political commitment to strengthen healthcare infrastructure.
The study provides valuable insights into healthcare workers’ experiences and contributes much-needed regional evidence on barriers to integrating mental health services into HIV care in Namibia.
People living with human immunodeficiency virus (PLHIV) experience higher rates of mental health issues compared to the general population.
Evidence on healthcare workers’ (HCWs) perceptions of integration is limited,
This study aimed to assess healthcare workers’ perceptions, opinions, and experiences on barriers to integrating mental health services into ART services in Namibia.
This study was conducted in Windhoek, Namibia’s capital. In 2023, Windhoek’s population was 494 605.
A descriptive mixed-methods study with a convergent parallel design was conducted to explore healthcare providers’ opinions and experiences regarding mental health services for PLHIV at ART clinics and the National Psychiatry Hospital in Windhoek, Namibia. This design enabled researchers to gather deeper insights, triangulate, and corroborate on challenges and gaps in mental health service delivery by integrating qualitative and quantitative findings.
Ten out of 12 public health facilities were selected based on the number of patients on ART. Healthcare workers providing ART care to PLHIV at ART sites and psychiatric outpatient departments were eligible to participate in the study. The study employed a non-probability snowball sampling method. Snowball sampling was used to recruit participants because the study targeted a specific group of healthcare workers experienced in providing ART and mental health services, which is difficult to identify through random sampling. A total of 44 (
The required sample size for the study was determined by considering findings from prior research, guidance from research books, and the specific phenomena under investigation. Literature suggests an appropriate sample size for a phenomenological study is 3 to 25 participants.
Data collection tools were developed in alignment with the study objectives and informed by a review of relevant literature on health workers’ perspectives regarding mental health and HIV service integration. The data collection tools gathered demographic data and items addressing key domains, including knowledge, attitudes, perceived barriers, and experiences related to integrating mental health services into HIV care. The data collection tools were reviewed for clarity and relevance by public health and HIV programme implementation experts. Subsequently, the questionnaire and interview guide were piloted with ART clinic healthcare workers to ensure clarity, and feedback was incorporated into the final versions of the tools. Data were collected in September 2023 and healthcare workers in urban ART clinics in Namibia, providing a snapshot of experiences and perspectives during that period. Data collection began with an information session for HCWs and the distribution of an information sheet, with permission from the department head. A questionnaire with open- and closed-ended questions assessed HCWs’ perspectives on mental health screening and management for PLHIV, training background, availability of mental health providers, facility conduciveness, standard operating procedures (SOPs) and guidelines, comfort in managing mental disorders, and views on mental health prevalence among PLHIV. The Drop-off/Pick-up (DOPU) method was used to improve response rates.
Questionnaire data were captured in Microsoft Excel and analysed in STATA version 15 (College Station, TX, United States), with data preparation and cleaning. Descriptive statistics, including frequencies and proportions for categorical variables, were used to explore the data. Responses were summarised in tables, presenting means for continuous variables and proportions for categorical variables.
Qualitative data were transcribed in English, and translation was unnecessary. Transcribed scripts were coded using NVivo 12 (QSR International Pty Ltd, Victoria, Australia) using thematic analysis. We employed inductive methods to analyse the data. Key features of a general inductive approach include condensing large volumes of raw text data into a summarised form and establishing clear connections between research objectives and findings.
All healthcare workers reported being fluent in English and were interviewed in English. The PI conducted all IDIs. The principal investigator has prior research expertise in the health sector and is knowledgeable about HIV service delivery in Namibia. Furthermore, the principal investigator has academic training and expertise in conducting qualitative interviews. Peer debriefing with a mentor was done to enhance reflexivity. To strengthen credibility, interviews were carefully transcribed through repeated reading of the transcripts and independent coding by two researchers, followed by regular discussions of the analysis. Transferability was supported by clear descriptions of the study setting and participants, enabling readers to assess whether the findings may apply to similar contexts. Dependability was supported through a systematic, transparent data analysis process, including the use of NVivo for coding and regular discussions between the researchers to refine themes. Confirmability was promoted through reflexive practices, including field notes and debriefing with a mentor to reflect on potential researcher bias.
Ethical clearance to conduct this study was obtained from the University of KwaZulu-Natal Biomedical Research Ethics Committee (No. BREC/00002904/2021) and the Ministry of Health and Social Services’ ethical committee (No: 17/3/3/NKC). The study involved minimal risk, and respondents gave verbal consent to participate and have the interviews recorded. Participants were allowed to withdraw from the study at any time and skip any questions they were uncomfortable answering. No personally identifying information, such as first and last names or identity numbers, was collected to ensure participants’ anonymity. All data were transferred daily from the recorder to a password-protected computer and then deleted from the recorder. Only the principal investigator (PI) had access to the password-locked computer. Questionnaires were kept in a lockable drawer accessible only to the PI.
Demographic characteristics of survey respondents (
| Characteristics | ART site ( |
Mental health site ( |
Grand total |
|||
|---|---|---|---|---|---|---|
| % | % | % | ||||
| Clinical mentor | 2 | 6 | - | 0.0 | 2 | 4.5 |
| Doctor | 1 | 3 | - | 0.0 | 1 | 2.3 |
| Health assistant | 5 | 14 | - | 0.0 | 4 | 9.1 |
| Medical intern | 0 | 0 | 5 | 55.6 | 5 | 11.4 |
| Medical officer/doctor | 5 | 14 | 3 | 33.3 | 8 | 18.2 |
| Nurse | 19 | 54 | - | 0.0 | 19 | 43.2 |
| Nurse mentor | 3 | 9 | - | 0.0 | 3 | 6.8 |
| Specialist | - | 0 | - | 0.0 | 1 | 2.3 |
| Total | 35 | 100 | 9 | 100.0 | 44 | 100.0 |
| Certificate | 2 | 8 | - | 0.0 | 2 | 5.9 |
| Degree | 11 | 44 | 6 | 66.7 | 17 | 50.0 |
| Diploma | 4 | 16 | - | 0.0 | 4 | 11.8 |
| Honours degree | 3 | 12 | - | 0.0 | 3 | 8.8 |
| Master’s degree | 5 | 20 | 3 | 33.3 | 8 | 23.5 |
| Missing | 10 | 29 | 0 | 0.0 | 10 | 23.7 |
| Total | 25 | 71 | 9 | - | 34 | 77.0 |
| Clinic | 14 | 40 | - | 0.0 | 14 | 31.8 |
| District hospital | 7 | 20 | 9 | 100.0 | 16 | 36.4 |
| Health centre | 8 | 23 | - | 0.0 | 8 | 18.2 |
| Intermediate hospital | 1 | 3 | - | 0.0 | 1 | 2.3 |
| National level | 5 | 14 | - | 0.0 | 5 | 11.4 |
| Total | 35 | 100 | 9 | 100.0 | 44 | 100.0 |
| < 1 | 2 | 8 | 5 | 55.6 | 7 | 20.6 |
| > 10 | 6 | 24 | 1 | 11.1 | 7 | 20.6 |
| 1–4 | 9 | 36 | 2 | 22.2 | 11 | 32.4 |
| 5–7 | 6 | 24 | 1 | 11.1 | 7 | 20.6 |
| Unknown | 2 | 8 | 0 | 0.0 | 2 | 4.5 |
| Missing | 10 | 28.5 | 0 | 0.0 | 10 | 22.7 |
| Total | 25 | 71.4 | 9 | 100.0 | 34 | 77.3 |
ART, antiretroviral therapy; HCW, healthcare worker.
The study findings were analysed using the Wakida et al. framework,
This study’s findings yielded common themes from the survey and key informant interviews. The main findings from the study are analysed and presented under four of the five SURE domains: (1) Motivation to change, (2) Knowledge and skills, (3) Management and leadership, and (4) Financial resources. Additionally, the findings are summarised and presented according to the SURE framework as well as the Wakida et al. framework of barriers.
‘… the spaces in the facility are limited, so maybe we need to find more space.’ (Health Assistant, 28 years)
‘We don’t talk much about mental health in our health education; we mostly concentrate on the physical well-being of the client.’ (Health Assistant, 28 years)
Routine screening was inconsistent across facilities. Most ART providers indicated that PLHIV are not regularly screened for mental illness, while psychiatric outpatient staff reported limited HIV screening among patients with mental disorders. Some respondents noted that underreporting is common and that symptoms may go unnoticed without proactive assessment. As one participant stated:
‘I think it is under-reported because the patients that we usually refer are the ones with signs and symptoms, and those that are going through depression or something, but not showing signs, we really don’t pick them up if we don’t ask.’ (Doctor, 32 years)
Screening practices were largely informal, often triggered by concerning behaviour or reports from relatives. One nurse explained:
‘Sometimes the relatives bring the patient … saying he has been hearing and seeing things, then you start to suspect something is not right.’ (Nurse, 38 years)
Alcohol use was the only area with a formal tool, with general mental health screening conducted ad hoc:
‘For alcohol, we have a screening tool … but for general mental health screening, we just see as we talk to the person.’ (Nurse, 54 years)
Human resource limitations further constrained service delivery. Few ART facilities had mental health professionals, and most relied on referrals to the psychiatric hospital. The absence of onsite social workers also hindered linkage to care:
‘… It would be easier if social workers were on-site so they [
Most ART facilities lacked SOPs or guidelines for mental health, and some psychiatric units lacked ART guidelines. Health workers emphasised the need for training, standardised tools, and job aids to support integration. As one noted:
‘Health care workers should be trained on how to assess whether someone has mental illness … and after training, they should put the tools in place.’ (HCW, 41 years)
Another highlighted the need to develop a training curriculum to facilitate integration.
‘We do not talk much about mental health in our health education; we mostly concentrate on the physical well-being of the client.’ (Health Assistant, 28 years)
Healthcare workers emphasised the need for treatment guidelines, job aids, and training to support integration efforts, with one manager stating:
‘I think we need to sensitise our leaders to understand the importance of integration. It’s very vital.’ (Health Manager, 34 years)
At the psychiatric hospital, 78% of respondents had been trained in HIV screening, diagnosis, and treatment, though some reported relying on referrals due to limited mental health diagnostic skills. As one nurse mentor explained:
‘I don’t have the proper training or knowledge of diagnosing a patient with mental illness, so we mostly refer.’ (Nurse Mentor, 35 years)
‘It would be time-consuming if we provide it [
Another added that employing the necessary experts requires funding and time:
‘It can be the experts that need to be employed, which will cost money.’ (Health Assistant, 28 years)
Human resource constraints were also highlighted as a barrier, particularly the high patient-to–health worker ratio, which compromises the quality of screening. A manager described this challenge:
‘You might want to do a proper screening, but looking at the long queue, you might not do justice to this one client.’ (Health Manager, 35 years)
This study explored healthcare workers’ perspectives regarding barriers to integrating mental health services into ART services, using a framework adapted from Wakida et al.,
Lack of space emerged as a major structural challenge affecting confidentiality and service provision. According to Wakida et al., the convenience of service provision is a key enabler for integration.
Limited knowledge and skills, and the absence of in-service training, further hindered mental health service provision, reflecting a capability gap within the COM-B framework. Although many HCWs received basic training during their academic studies, the lack of ongoing in-service training led to significant knowledge gaps. This aligns with findings from Ethiopia, where HCWs reported insufficient skills to diagnose and manage mental illness.
Limited human resources were also a key barrier. HCWs noted understaffing, heavy workloads, and insufficient financial resources to hire additional staff, consistent with global evidence showing a gap between mental health needs and workforce availability in low- and middle-income countries (LMICs).
These findings have several implications for policy and practice in Namibia. Integrating mental health services into PHC ART clinics will require investments in physical infrastructure and designated consultation spaces. National guidelines and SOPs for mental health screening and treatment should be developed and disseminated to support standardisation across facilities. Well-structured in-service training programmes and continuous professional development initiatives are critical to improving HCWs’ capabilities, while task-shifting strategies, leadership engagement, and the allocation of financial resources can enhance opportunity and motivation. Based on these insights, immediate actions should focus on providing in-service training, SOPs, job aids, and temporary private consultation spaces. In the medium term, permanent designated mental health areas, standardised screening tools, and supervision and mentoring should be established. Longer-term strategies should integrate mental health into national PHC and ART policies, allocate sustainable financial resources, and institutionalise task-shifting approaches. Collectively, these steps provide a practical, phased roadmap for the sustainable integration of mental health services into ART care in Namibia, addressing the critical domains of skills, resources, and leadership engagement.
This study has several limitations. Snowball sampling may have introduced selection bias, as participants were referred by colleagues who may share similar perspectives. The small qualitative sample limits generalisability, although in-depth interviews allowed for detailed exploration of healthcare workers’ experiences, and key themes reached saturation. Conducting the study only in urban ART facilities further restricts generalisability to rural or other settings in Namibia. Social desirability bias may have influenced participants’ responses, despite assurances of confidentiality and encouragement to share honest experiences. Additionally, as the principal investigator conducted the interviews, responses may have been influenced by the researcher’s presence; reflexive practices, including debriefing with a mentor, were employed to minimise potential bias. This study used qualitative data from surveys and interviews, with written responses potentially providing less depth than interviews or focus groups. Combining perspectives of doctors, nurses, and nurse mentors may have obscured subtle differences between their viewpoints. Additionally, the study did not explore attitudes related to programme acceptability, appropriateness, or credibility, warranting further research. Despite these limitations, the study offers valuable insights into healthcare workers’ perspectives on barriers to integrating mental health into ART services, which can inform policy and practice in Namibia.
The study found key barriers to mental health service provision in PHC settings in Namibia, such as staff shortages, limited facility space, and a lack of guidelines, knowledge, and skills among healthcare workers. These issues hinder effective screening, diagnosis, and treatment of mental illness in PLHIV, obstructing integration into primary care. Solutions include reorganising facilities, training healthcare workers, implementing task-shifting, and securing political commitment to invest in infrastructure. Importantly, these findings highlight the need for national policy initiatives to integrate mental health into HIV care guidelines, allocate resources for workforce development, and standardise service delivery across PHC facilities in Namibia.
The authors declare that they have no financial or personal relationships that may have inappropriately influenced them in writing this article.
Ndeshiteelela K. Conteh: Conceptualisation, Data curation, Formal analysis, Investigation, Methodology, Project administration, Software, Writing – original draft. Ozayr Mahomed: Methodology, Supervision, Visualisation, Writing – review & editing. All authors reviewed the article, contributed to the discussion of results, approved the final version for submission and publication, and take responsibility for the integrity of its findings.
The datasets generated and analysed during the current study are not publicly available due to privacy and confidentiality considerations involving healthcare workers, but are available from the corresponding author, Ndeshiteelela K. Conteh, on reasonable request.
The views and opinions expressed in this article are those of the authors and are the product of professional research. It does not necessarily reflect the official policy or position of any affiliated institution, funder, agency, or the publisher. The authors are responsible for this article’s results, findings, and content.