This study utilised an exploratory, descriptive and qualitative design. It was chosen to obtain a deeper understanding of the under-researched context of HReH in the Free State. This method made it possible to generate rich, in-depth descriptions of participants’ lived experiences as well as to explore difficult, context-specific problems. The research articulated in this article formed part of a larger study aimed at determining access to eye health through HReH in the Free State Department of Health (Free State DoH). Furthermore, thematic data analysis adapted the CBF for vision and eye health in public health.¹²

This study was conducted in the Free State province of SA. The province has one large district called Mangaung and classified as a metro, and four district municipalities: Fezile Dabi, Lejweleputswa, Thabo Mofutsanyana and Xhariep. Geographically, the Free State expands over approximately 129 825 km².⁹ Within the Free State DoH, vision and eye health services are primarily provided in hospitals that have dedicated eye clinics; hence, these hospitals were selected for the study. A total of seven hospitals, categorised into two district, four regional and one tertiary eye care hospital, met the inclusion criteria and were included in the study.

Purposive sampling was used to identify information-rich participants who were HReH in hospitals with eye clinics. The target population in this study were all HReH cadres (clinical and non-clinical) in selected hospitals. Initially, all the public hospitals in the province were approached to establish which ones had eye clinics. This confirmed eligible hospitals and allowed the primary investigator (PI) to start inviting participants at the correct hospitals from the five districts. A non-random purposive sampling strategy was employed in the study because of the projected small and specialised HReH population employed in Free State DoH and expected saturation confirmed by no new themes emerging after the 24th interview.

All HReH employed in the Free State DoH eye clinics were eligible for the study. Following ethical approval, all eligible eye clinics were telephonically contacted, informed of the study, and emailed details thereof. Interested participants were told of the PI’s visit to conduct the study and obtain contact information of those not present. Each willing participant was briefed on the aim and objectives of the study. Voluntary consent was emphasised to all enrolled willing participants. In addition, an information sheet with details of the research team and a consent form to peruse prior to commencement of in-person one-on-one interviews. Amicable interview dates and times were centred around the participants’ preferences and interviews commenced on submission of the signed consent forms.

Data were collected between May 2021 and May 2022 in a well-ventilated room that was outside the clinic to facilitate personal comfort and open and honest expression of the participants. Confidentiality and anonymity were assured through data coding, ensuring that responses could not be traced back to any individual. Data collection tools included a semi-structured interview guide and two digital recording devices. The interview guide was developed by integrating the key domains of the CBF and used to structure the flow of the interviews, systematically addressing all key thematic areas of the CBF. This ensured uniformity and consistency in the scope and focus of the data collected. Probing questions were used wherever needed to extract thick descriptions and detail. Albeit guided in English, other interviews were conducted in a mixture of Sesotho, the most widely spoken language in the province, and English. However, all depended on the language preferred by the participant. To minimise bias and reinforce consistency in the data collection process, the interviews were conducted by the same trained qualitative researcher who had no prior relationship with the participants. In addition, a neutral and non-leading language was used during the interviews to avoid influencing the participants’ responses.

All recorded interviews were transcribed verbatim and further translated into English by the interviewer, who is a native Sesotho speaker and fluent in English. This ensured that translations maintained both linguistic and contextual accuracy. To enhance the reliability of the translated data, a second bilingual translator independently reviewed a sample of the translated transcripts. Although back-translation was not applied to all transcripts because of resource limitations, discrepancies identified during the review process were discussed and resolved between the PI and the independent reviewer. This process helped ensure the trustworthiness and consistency of the translated qualitative data. The interviewer, trained as an optometrist with experience in both public and private eye care, acknowledged their professional background through reflexive journaling and engaged in regular peer debriefing with the research team to minimise bias and ensure balanced interpretation of participants’ narratives. Prior to data collection, the interviewer received training in qualitative research methods, including the use of semi-structured interview guides, probing techniques, and maintaining neutrality to avoid leading participants’ responses. This background provided a strong foundation for understanding the clinical and systemic challenges within the study setting.

When data collection was concluded, the recordings were downloaded onto a folder on the research team’s cloud account and stored in a password-protected file. Data were transcribed verbatim, translated, and cleaned. Codes were allocated to each interview to ensure anonymity, and each file was stored in the cloud folder, together with the original recordings, prior to analysis.

A thematic analysis approach was employed, involving a systematic process of preparation, organisation and reporting of data.¹³ Using a deductive methodology informed by the CBF, the qualitative analysis method as outlined by Elo and Kyngäs was used to analyse the data.¹³ In the preparation phase, the transcripts were read repeatedly to ensure immersion in the data and to gain an overall understanding. Based on the CBF, a systematic categorisation matrix was developed during the organising phase. Following a line-by-line review, the data were categorised based on how well they fit into these predetermined categories. Within each primary category, subcategories were created inductively to capture subtle themes that emerged from participant responses. NVivo version 12 was used to support the coding and an independent coder assessed it to increase rigour.

To ensure credibility and trustworthiness, several criteria were employed. Prolonged engagement was maintained with participants during data collection, allowing the interviewer to build rapport, foster trust and obtain deeper insights into the participants’ experiences and perceptions. Thick description was used throughout the analysis to capture the complexity of the findings, with illustrative quotes included to provide rich context and meaning. Furthermore, member-checking was conducted with participants by sharing their transcripts for verification, and all participants who reviewed their transcripts confirmed the accuracy of the captured responses, and no changes were requested.

Themes and sub-themes, supported by direct excerpts, were established using the adapted CBF (Figure 1) as a guide. The research team further peer-reviewed and analysed the process and contents for reliability, validity and credibility, further ensured by the input from the independent coder. The New South Wales (NSW) Health CBF was designed to help health systems improve their ability to deliver better health outcomes and strengthen the foundations for effective service delivery. The framework recognises that capacity-building strategies need the right context and features, infrastructure, skills and relationships for an effective health system. The adapted framework focused on five key areas:

Ethical clearance to conduct this study was obtained from the University of KwaZulu-Natal (No. BREC/00000609/2019) and the Free State Department of Health. The Helsinki Declaration was followed in conducting the study, considering voluntary participation, confidentiality during data collection and reporting of findings.¹⁴ The head of Free State DoH granted permission to implement the study, and Chief Executive Officers (CEOs) and clinical managers of the respective health facilities were informed accordingly. Ethical clearance was obtained from the Biomedical Ethics Research Committee at the University of KwaZulu-Natal and the Free State Research and Ethics office. The authors informed all participants that participation was voluntary, unremunerated and could be withdrawn at any time. To confirm voluntary participation, each participant gave written consent prior to data collection. All data were de-identified and safely stored in password-protected digital files that were only accessible by the research team in order to maintain confidentiality and anonymity. In reporting and transcripts, participant codes were used in place of names. These procedures were put in place to respect ethical standards and protect research participants’ privacy.

The sample comprised 27 participants, with 70.4% (n = 19) females and 29.6% (n = 8) males (Table 1). The age range was from 25 to 61 years, with the mean age being 44.5 years (standard deviation [s.d.] = 11.3). The most common qualification among participants was a B. Optometry (10 participants), with additional qualifications such as Matric (2), MBChB (2) and various specialised diplomas and degrees (e.g. Postgraduate Diploma in Ophthalmology, BSc Honours in Nursing Science), highlighting a skilled and well-rounded workforce. Experience in public sector roles varied significantly between 12.5 and 29 years (s.d. = 8.2).

Data are presented under the six themes centralised within the CBF (NSW Health Department 2001) for building capacity to improve health: organisational development, workforce development, resource allocation, partnerships, leadership and building capacity (Table 2).

A major theme among participants was associated with organisational development is halted by inefficiencies such as unclear referral protocols and inadequate policies, disrupting workflow and patient care, evidenced by the following excerpts: ‘There is no specific protocol for eyecare’ (P17, F, 37) and ‘I don’t know if there is any policy, but for our internal referrals we got the referral forms’ (P18, F, 37). Further expressed was the need for clearer communication channels as well as standard protocols to enhance eye clinic operations and efficiency. Others emphasised the need for policies and better-defined roles within the clinic to improve both staff experience and patient outcomes: ‘The last time eyecare policy was updated was many years ago’ (P10, F, 44).

Extracted from responses to the last three questions of the interview guide, staffing issues emerged as a crucial matter in the province’s eye clinics. Some participants reported that staff shortages are a primary concern in eye centres. In addition, others lamented overburdened staff that struggle to meet patients’ needs, thus impacting negatively on the quality of patient care and student training as an extended role. Henceforth, staff multitasking and stretching of resources are performed:

This is the first study in Free State province to explore and map HReH, and its use of the CBF to organise and interpret findings is a notable strength. It provides a comprehensive lens through which to examine challenges in organisational development, resource allocation and workforce development. Moreover, the qualitative approach provided rich and significant insights into the lived experiences of HReH, enabling a grounded understanding of systemic inefficiencies. The qualitative data have strong trustworthiness measures where member-checking and independent coding were used. However, its focus on a single province may limit the broader generalisations of findings to other regions in South Africa or sub-Saharan Africa with different demographic and resource profiles. In addition, the reliance on participant self-reporting could introduce bias, potentially over-representing negative aspects of the health system. The bilingual interviews enriched data, but translation may have affected nuance, which was mitigated by a bilingual reviewer cross-check. Despite these limitations, the findings provide valuable insights that contribute to the evidence base for policy and workforce development.

Systemic improvements within the eye health service are recommended through clear communication, updated policies and adequate staffing to help reduce burnout and enhance service delivery. It is important to invest in essential equipment and consider equitable distribution of resources to develop efficiency. Moreover, the call for a stronger multidisciplinary approach or collaborations would ensure comprehensive care for the patients. Advocacy for HReH, transparent leadership and staff participation in decision-making can enhance morale and retention, particularly in underserved areas. Resource gaps can be overcome through public–private partnership initiatives and by placement of more optometrists in rural areas to improve access to care. Longitudinal studies might also yield rich insights into the impact of targeted interventions to advocate for developing eye health systems that are equitable and sustainable.