Workplace-based learning opportunities in a South African family medicine training programme

Background Workplace-based learning (WBL) provides authentic learning opportunities to develop fit-for-practice healthcare workers. Different types of WBL opportunities have been described in high-income countries, but the opportunities in the district health systems of sub-Saharan Africa have not been characterised. Aim This study explored family physicians’ (FPs) and registrars’ perceptions of WBL opportunities in a decentralised postgraduate family medicine registrar training programme. Setting The study was conducted at five decentralised training sites across two provinces affiliated with the University of the Witwatersrand in South Africa. Methods This instrumental case study involved semi-structured qualitative interviews with 11 FPs and 11 registrars purposively sampled across the training sites. The verbatim transcripts were thematically analysed using Braun and Clark’s six-step approach. Results Workplace-based learning opportunities were grouped into four themes: Learning from interpersonal interactions, learning from district activities, self-directed learning and contextual influences on learning opportunities. Registrars learnt from patients, peers, FPs and other professionals. Feedback, self-reflection, portfolio use, involvement in various district events, such as student and staff teaching, and continuous medical education augmented learning. Contextual influences originated from health facilities, resource availability, district management and university support, excessive workload and a need for standardised district learning opportunities. Conclusion Registrars are exposed to several types of WBL opportunities in district health systems. Effective engagement with available opportunities and addressing contextual challenges could enhance registrar learning. Contribution Maximising learning opportunities to promote registrars’ acquisition of required skills and competencies to efficiently address community needs in a middle-income country such as South Africa.


Introduction
The World Health Organization's (WHO) Astana declaration (2018) advocated strengthening primary health care (PHC) services and prioritising universal health coverage, focusing on preventive, promotive, curative and rehabilitative care services. 1The training of skilled health workers to address community needs is a global priority. 1,2Medical training institutions continually modify curricula to equip postgraduate trainees with the knowledge, skills and professional attitudes needed to address community needs. 2,3,4,5,6,7Workplace-based learning (WBL) involves trainees learning in the workplace by meeting the clinical service delivery requirements and utilising authentic contexts for optimal training and skills development. 8Studies on WBL have been conducted in high-income countries such as the Netherlands and the United Kingdom, but very few studies have been conducted in sub-Saharan Africa, including South Africa (SA), on workplace learning experiences.An in-depth exploration of views of WBL opportunities in postgraduate training based in an SA district health context is still lacking.This article describes the various types

Workplace-based learning opportunities in a South African family medicine training programme
Read online: Scan this QR code with your smart phone or mobile device to read online.
of WBL opportunities and workplace factors influencing WBL in the district health context of SA.

Learning in the workplace
Workplace-based learning ranges from learner-centred, informal, unstructured learning in the absence of a trainer to formal learning in the presence of a mentor or supervisor. 8nformal WBL lacks the defined curriculum, timetables and linear teaching that characterise formal teaching at the university level, resulting in criticisms about WBL being disorganised and opportunistic. 8Despite a perceived lack of educational rigour, process and structure, WBL is becoming the preferred strategy.
Workplace-based learning draws on several learning theories, including cognitive, socio-cognitive and social constructivist theories. 8The cognitive theories emphasise individuals learning from their own experiences and selfreflective processes, while the socio-cognitive and socialconstructivist theories focus on learning from social interactions. 8In cognitive theories, learning is seen to happen during the interplay between existing and new knowledge. 9n socio-constructive theories, learning is visualised as part of everyday practice and occurs by engaging with peers, supervisors and other healthcare team members.Informal learning at the workplace involves implicit, reactive and deliberative learning. 10Implicit learning refers to knowledge acquisition that happens with no conscious attempt to learn; reactive learning is about spontaneous intentional learning that occurs while performing an action, and deliberative learning involves planning towards defined learning goals. 10lf-regulated and self-directed learning are integral components of WBL.In self-regulated learning, trainers regulate learning primarily in academic environments.In contrast, self-directed learning involves trainees developing their learning goals, identifying activities and resources and seeking external feedback at their workplaces. 11While self-directed learning is moderated by internal processes such as self-reflection, it is also modified by external educational interventions, such as feedback, assessments and learning portfolios. 12Self-directed learning consists of short loops that are triggered during consultations with a lack of knowledge during minor learning activities at the work place. 12Long loops include more extended learning periods based on complex problems, such as difficulties in communication or handling cases such as child abuse. 12

Factors affecting workplace-based learning
Several personal, interpersonal and contextual factors affect WBL.Personal factors that enhance learning include learner behaviours such as active involvement, accountability, professionalism, conscientiousness and acceptance of criticism. 13Prior experience, knowledge, motivation, attitude and confidence also affect trainees' learning. 14,15,16A lack of concentration, difficulties in dealing with negative feedback and managing work and private life, and being passive learners with a lack of motivation act as barriers. 12,13,17rainees' WBL is enhanced when using workplace artefacts such as case reports, reflective logs or portfolios. 16terpersonal factors affecting WBL arise from trainees' interactions with peers, supervisors, other professionals and patients. 16,18,19,20More WBL takes place during interactions with peers and supervisors 18 and as part of informal discussion or handovers than when trainees work alone. 19eer interactions motivated trainees to put in extra effort to reach their peers' standards. 19Interaction with a supervisor affirms trainees' behaviours, actions and decision-making during patient consultations and helps them apply their knowledge. 21Other motivational factors are supervisor, mentor or peer engagement and feedback offered in a safe environment, 15 good supervisor-trainee relationships, supervisory commitment, opportunities for supervision, and mutual observation and dialogue. 13In contrast, poor supervision and supervisory relationships hinder learning. 12ter-professional or intra-professional interactions promote informal and formal learning 16,18,22 through reflective practice, spontaneous or triggered, implicit through participation, increasing self-awareness and developing coping mechanisms. 16earning is enhanced by social integration, various task allocations, and successes and mistakes. 23Trainees also learn from engagements with patients 20 and their families, through reflecting on patient encounters, especially with difficult or peculiar patients, 22 critical incidents and patient communication. 24During patient interactions, trainees identify deficits in medical competencies but not always their lack of general competencies, such as communication skills and ethics. 19rkplace-based learning needs a supportive clinical environment, as it directly influences patient-care practices. 25 conducive WBL environment requires material resources, an appropriate patient mix and clinician-to-patient ratio, sufficient clinical trainers, a manageable workload, protected time for learning and support from relevant stakeholders. 26,27reating a positive learning environment with sufficient trainee support and knowledgeable, passionate and skilled trainers is vital. 26,27,28A collaborative learning climate encourages reflection, feedback, debriefing, supervision and guidance from supervisors. 14Organisational factors, such as the layout of the work environment, interpersonal dynamics among team members and the availability of complex patients 14 augment learning. 16Health system factors such as high supervisor and trainee workloads and high patient volumes prevent trainee observations. 13

Workplace-based learning in South Africa
Two SA studies identified several factors that promoted and hindered WBL experiences.One study of postgraduate registrars (SA trainees) in a laboratory setting categorised the origin of these factors as the university, workplace, home circumstances and personal. 29Academic role models and supportive trainers who enjoyed teaching promoted learning, while unstructured academic activities, feelings of demotivation, conflicting family and work responsibilities and negative supervisor feedback hindered learning. 29 family medicine (FM) study identified context, adequate utilisation of a learning portfolio, patient consultations, engagement with clinically relevant supervisors and providing sufficient feedback as promoting learning. 30gistrar learning in workplaces in post-graduate FM decentralised training at the University of the Witwatersrand (Wits University) in SA is primarily based on self-directed and self-regulated learning.The university curriculum and national learning outcomes, as determined by the SA Academy of Family Physicians (FPs), are readily accessible in the public domain. 6,31Registrars develop individualised learning plans aligned with the national learning outcomes, according to their learning goals in workplace settings.The activities they undertake to achieve their goals, including setting timelines and finding resources, are augmented by guidance from their supervisors. 31Workplace-based learning involves clinical and educational supervision by FPs or specialists in various disciplines during clinical rotations, the annual compilation of a learning portfolio and formative and summative assessments.
The study reported in this article forms part of a broader mixed methods case study, evaluating a postgraduate FM decentralised training programme by using a complex programme evaluation logic model.The previously published articles from the larger study reported on resource availability, postgraduate supervision and supervisory feedback evaluated as the inputs, processes and outputs of the logic model. 32,33,34his article reports on the FPs' and registrars' perceptions of types of the learning opportunities, registrars' learning behaviours and the learning environments of the decentralised postgraduate FM training.

Study design
Case studies explore, in-depth, the multiple perspectives of the complexity and uniqueness of a phenomenon in a 'real life context' of a 'bounded system', such as a programme or an event 35 by using various data sources and data-collection methods. 36An instrumental case study focuses on a particular case to gain in-depth insight into an issue or to redraw a generalisation. 35An instrumental case study investigating the various aspects of the 'phenomenon of interest', that is, the postgraduate FM decentralised registrar training programme, in the context of Wits University was conducted.

Study population and sampling
The target population comprised 20 FPs and 21 registrars.These individuals were purposively sampled 39 because they were perfectly positioned to provide in-depth insights into the WBL opportunities in the programme.Purposive sampling involves the deliberate selection of individuals or sites 39 to achieve representativeness across settings, capture adequate homogeneity of the study population by obtaining a range of variations and examining critical cases to compare or illuminate the differences between settings or individuals. 40The FPs and registrars had varying exposures to and experiences of WBL opportunities in the decentralised training programme depending on their roles and responsibilities in each district.We included FPs and registrars from all five training sites for broader geographical representation.
All second-and third-year registrars in the programme were invited to participate.First-year registrars were omitted as they did not have adequate training experience to contribute meaningfully.Registrars in their fourth year and beyond were excluded as they were completing their elective rotations and were not actively involved in the yearly training programme.Family physicians, who are jointly appointed by the university and the relevant provincial health departments, were invited to participate.Recently qualified FPs who had not registered as specialists and were not joint appointees were excluded.All participants who were invited from the FPs and registrars agreed to participate in the study.The final samples consisted of 11 FPs and 11 registrars.

Data collection
The principal author conducted 80-to-90-min semi-structured interviews between March and August 2020.Eight face-toface interviews were conducted with the remainder (n = 14) on Zoom or Microsoft Teams because of the coronavirus disease 2019 (COVID-19) lockdown.Table 1 represents the interview guide.The interview process continued until data saturation was reached. 41The interviews were audiorecorded, transcribed verbatim and checked for fidelity.The registrars and FPs transcripts were open-coded 42 (first level of coding into individual segments) separately.Codes were compared and contrasted across the transcripts for each participant group.No new codes were identified after nine interviews from each group, but two more were interviewed from each group to ensure data saturation.

Data analysis
Braun and Clark's six-step approach to thematic analysis was used to analyse the transcripts. 41The steps are: (1) familiarisation with the data; (2) generation of initial codes; (3) searching for themes; (4) reviewing themes; (5) defining and naming themes; and (6) producing a final report. 41MAXQDA 2020 (Verbi Software, Berlin, Germany) was used to manage the analysis.
The initial coding system was checked by the two co-authors until agreement was reached on the naming of codes.The final coding system was applied to transcripts through an iterative process of coding and checking until agreement was reached about grouping codes into categories and themes, and reviewing and naming the themes.

Trustworthiness
Frequent discussions between the authors improved the intercoder reliability and the credibility of the findings. 43A codebook was developed and constantly revised until the coding system was finalised, further improving credibility. 44etailed descriptions of the study methods, including setting, inclusion and exclusion criteria, and data collection, augmented the transferability. 45Data triangulation from the two groups of participants also improved credibility and dependability.Dependability of the findings were further improved by recording decisions made from the emergent data at a particular time and their rationale, and how various codes and categories were grouped, themes developed, reviewed 46 and renamed during analysis.
Reflexivity relates to the degree of bias that the researcher intentionally or unintentionally introduces into the research. 47he primary author was cognisant that her position as a supervisor and colleague to some participants may have influenced their responses.She mitigated these by reassuring them about the anonymity and confidentiality of the findings.She also self-critiqued and constantly reflected on her personal biases and assumptions in a journal to improve the confirmability.The co-authors were not involved in the postgraduate FM training being evaluated.Their reflexivity centred around their roles as medical educators -they reflected, examined and explored their interpretations of the findings based on their involvement and experience in postgraduate FM training.

Ethical considerations
Ethical clearance to conduct this study was obtained from the University of the Witwatersrand Human Research Ethics Committee (Medical) (No.M191140).Permission to conduct the study was obtained from five districts through the National Health Research Database (GP_201910_050).Informed written consent was voluntarily obtained from study participants.

Results
The age of the registrar participants' (RG1-RG11) ranged from less than 30 to 60 years.Six registrars were in their third year and five were in their second year of study.Registrars had PHC experience before joining as community service doctors, medical officers or in private practice.The FP participants' (FP1-FP11) ages ranged from 41 to 60 years and most had more than 5 years of training experience (see Table 2).
The four themes identified were classified according to the learning opportunities from people, activities, self and the environment.Figure 1 shows the sub-themes within each theme.

Learning from interpersonal interactions
Registrars perceived that the learning opportunities were primarily from patient interactions, other professionals and FPs.The FPs agreed with registrars but thought there were peer learning opportunities for registrars during training.

Patients
Registrars and FPs felt that the registrars had adequate opportunities to learn from undifferentiated and complex patients with various conditions.These challenges broadened their thinking processes: 'You get the opportunity to be exposed to so many available, let me say, good patients that will come with many challenges at once.So, a complex patient.You'll have this very often that challenges you as a clinician to broaden your thinking and to do.' (RG 6) Most registrars recognised the importance of taking each patient encounter as a learning opportunity.Each patient motivated them to apply FM principles, read the literature and critique their approach to patient management:

Family physicians
Registrars mentioned discussing patients with FPs at their workplaces as an excellent learning opportunity.During patient consultations, the presence of FPs provided opportunities for registrars to identify their strengths and challenges, which motivated them to read more about the patient: Family physicians agreed with registrars that they had more opportunities to practice FM principles in the presence of their supervisors: 'Where the registrar is placed, and he or she can make use of all those tools adequately and necessary, with supervision or with help, is important.'(FP 10) In contrast, many registrars felt they did not have adequate opportunities to observe role modelling, either because of the lack of availability of FPs but more often because of the small number of FPs or their multiple roles.Registrars wanted to observe more of their consultant's approach to patients and their families, which was sometimes challenging: Family physicians raised concerns that despite various opportunities to learn from other professions, registrars did not maximally utilise them, even at the clinic level: 'I tell this registrar, you've got an optometrist who's got a big machine that we've just bought for her, why are you struggling to know how to do fundoscopy … You can make a time and go to her and say, I'm going to come to you every Friday for one hour for the next two months.So, the opportunities are there, but I'm not so sure whether our registrars are using it on their own.' (FP 9) Some registrars agreed with FPs that they could have made better use of interprofessional learning opportunities: 'I think our learning is quite fragmented.You mostly learn from the doctors, from the family physicians.We do sometimes get involved in student teaching.But in terms of your multidisciplinary level of teaching involving nurses, physiotherapists, OT [occupational therapists] … it's from person-to-person interest, but there's not much integrated learning.'(RG 1)

Learning from district activities
According to FPs and registrars, registrars' active involvement in training medical and other students and staff is an essential component of WBL learning.Most registrars and FPs concurred that registrars had opportunities to participate in district CME (Continuing Medical Education) activities, community-oriented primary care with home visits and mortality and morbidity meetings, which enhanced their learning.

Attend district, mortality and morbidity and management activities
Family physicians mentioned that participating in district community-oriented primary care and quality improvement projects facilitated registrar learning at the workplace: A significant concern raised by a few FPs was that registrars were insufficiently exposed to leadership roles to prepare them to take on these roles once qualified as specialists: 'Especially maybe our senior registrars, looking back in terms of even when we were registrars, say, for example, we never attended any senior meeting.Now suddenly you are qualified, and suddenly you must attend senior meetings, you must make decisions, but you've never been technically trained in terms of…' (FP 1)

Self-directed learning
Family physicians and registrars thought that learning occurred during registrars' self-directed learning activities such as self-learning, reflection, feedback and compiling their learning portfolio.Both groups of participants varied in their opinions on registrar learning behaviours, approach to feedback, reflection and utilisation of a learning portfolio.

Self-learning
Some FPs felt that some registrars did well on self-directed learning, with exceptional efforts to keep up to date with current knowledge.In contrast, other registrars failed to understand the self-directed learning process and struggled to manage it: One FP commented that registrars believed they knew what adult education was when they did not: In contrast to this, others thought that registrars were committed and improved their behaviour: 'It's exceptional.They're committed to learning.I'm impressed with their learning behaviour, and I'm impressed with their dedication.They have really improved tremendously, and they are willing to learn as well.' (FP 4) The FPs in some districts commented on the lack of professionalism and work ethic, but opinions varied among FPs:

Learning portfolios
All registrars and FPs agreed that a learning portfolio is an exceptional tool to support self-directed learning, providing ample guidance on learning objectives and how to achieve them:

Reflection
Family physicians believed that more reflection on their patients and the feedback they received could provide better opportunities for self-directed learning, instead of the variable reflection they thought was being undertaken: 'The reflection is something that is also a little bit neither here nor there.Because if you see…you give a registrar feedback and then when he comes next time and the same problem crops up, then you begin to realise actually this registrar did not reflect on the feedback that you give.' (FP11) In contrast, registrars believed that they reflected on supervisor feedback, especially on the strengths and challenges addressed, and they attempted to improve themselves in future patient encounters:

Contextual influences on learning opportunities
Registrars and FPs agreed that the learning environment was conducive to learning, with adequate patient numbers and various levels of hospitals, but the resources were variable.Contextual challenges included excessive workload, a lack of adequate support from district management and the university, and a lack of standardisation, negatively influencing registrar learning.

Training across different facilities promotes learning
Both participant groups agreed that training across various district health facilities, including PHC clinics and district and regional hospitals, allowed access to patients with different sociodemographic characteristics and high burden of disease, which added to the richness of the learning environment: 'We have the district hospital and then you have the PHCs, and then we have the regional hospital.So, when we are at [sic] the district we're able at least as registrars in family medicine, we're able to see cases that are sometimes then referred from the regional hospital, and it gives you an idea what is for the district hospital, which cases are supposed to go to.' (RG4) 'We're in a district where more than 45% of adults are unemployed.So, the social determinants of health, they are [visible] here; if there's any place where they impact, it's here.So, the burden of disease, the environment, is good for learning.'(FP 9) One registrar commented that she got better learning opportunities in smaller hospitals with a stronger primarycare focus, compared with bigger hospitals: 'I wouldn't maybe learnt as much as I've learnt in these three years, by not given the opportunity and being put into the position, in such small facilities where you have to learn to, number one, make do with what you have, and you have to teach yourself and you need to make sure that you are up to date.' (RG 8)

Variable resources impede learning
Family physicians agreed that FP numbers had improved during the years in some districts, which was encouraging, but there was still a challenge in other districts.Both participant groups commented on the variability of available human and material resources across the districts: 'I think in terms of the skills, like clinical, especially like OSCE skills, that sometimes is a little bit challenging because there's sometimes a lack of resources.You know, for example, there's no like speculums to do like a pap smear, or things like that.' (RG10) 'I think has improved over the years with regards … They have the materials, the mannequins, the books, the opportunities to learn with tools around in the clinics … everything is actually provided for them; it's just a question of the trainers.'(FP 3)

Excessive workload
Registrars and FPs expressed challenges of heavy workloads as part of service delivery, negatively impacting learning and requesting service delivery support from their workplace: http://www.phcfm.orgOpen Access [clinics].So, there's been this on-going rift that the district is paying the registrars but they're working in the hospital.And they tend not to understand that this is a training programme.'(FP 4) Family physicians requested university engagement to create awareness about FM training in the district, which could augment learning opportunities: 'I think Wits [University] needs to come in here and assist us so that if they can talk at a higher level and say, look, the registrars are at training, much as they are part of the workforce but they need to be … people need to understand that their quality of work, or whatever they are doing, needs to be linked with their objective, which is learning.'(FP 11) Meanwhile, registrars appreciated accessibility to university resources such as the library: 'I think that we have access to the resources and references via the Wits Health Sciences library, and so whatever I do, I can always go back and review the best standard practice.'(RG7)

District learning opportunities need to be standardised
Some registrars and FPs emphasised that registrars' learning in the districts was variable, and there was a need to standardise learning opportunities:

Discussion
This study identified multiple types of learning opportunities for registrars, including interactions with various groups of people, participation in district activities and self-learning strategies.However, the WBL opportunities available were used variably both within and across districts.Compared with previous studies from high-income countries, this study in a middle-income country contributes a different perspective on what is needed to enhance WBL in authentic clinical settings of PHC.
Regular interactions with complex and undifferentiated patients in authentic clinical settings provided WBL opportunities.Patient engagements and reflections from those interactions are considered among the best WBL strategies for trainees. 12,16,19,23,27Registrars learned by integrating the FM principles, reading about patients' conditions and developing self-directed learning behaviours.Patient interactions also helped registrars to practice more soft skills, including communication skills, clinical reasoning and professional behaviour.
Registrars had adequate opportunities to learn from peers and FPs in most districts.Peer learning interactions and sharing views and experiences among trainees in different years are recognised WBL strategies. 27Interestingly, peer learning opportunities identified only by the supervisors underscored the importance of intergenerational learning, which was not recognised by registrars.Family physicians thought registrars were more pressurised to learn when they identified knowledge gaps in the presence of peers, as found elsewhere. 19Previous studies perceived that learning from peers was more acceptable for registrars than from supervisors, as supervisor presence could interfere with a safe learning environment. 19In contrast, interactions with FPs were identified as excellent opportunities for registrar reflections on their strengths and weaknesses.Family physician accessibility during and after work hours encouraged registrars to integrate the FM principles or tools, positively influencing WBL.Interactions with peers and supervisors provide excellent learning opportunities 12,18,19,48 and constitute examples of situated learning. 49Role modelling could potentially demonstrate excellent clinical teacher, human and professional behaviours 50 and has a greater impact on trainees to internalise those behaviours than any other teaching method. 51A lack of role-modelling opportunities hindered WBL in this study, as identified elsewhere. 20Peer teaching 48 and supporting trainers are also essential to optimise WBL opportunities in clinical settings. 3rticipants identified learning from other types of professionals as essential for WBL.Previous studies in highincome countries showed that learning occurs between health professionals during workplace interactions. 18,19,20,21,22,23nterprofessional learning between various health professionals working in collaboration is a WHO recommended strategy to address health systems challenges worldwide. 52According to most participants, although interprofessional learning was a good learning opportunity, it was insufficiently utilised.
Trainees were often reluctant to spend time learning from other professions, perhaps because they felt it was not part of their assessments.Other reasons could be the non-recognition of others' expertise, professional stereotypes and hierarchical challenges. 16,53Workplace-based learning opportunities from other professions varied across clinics and district hospitals, and registrars identified more opportunities at the district hospital.For example, dietary advice offered by dieticians could be observed and utilised by the registrars for managing a diabetic or hypertensive patient.Similarly, basic or advanced counselling skills learnt from a psychologist could be applied to patients with mental-health illnesses while practising holistic care.A recent SA study reiterated the importance of trainees' learning from other professions, as it improves collaboration and teamwork, professional satisfaction and patient care. 54e availability of a range of other learners has been found to enhance learning opportunities. 26,51The participants in this study reported that registrar training of students and staff enhanced learning by providing them with a platform to gain practical experience towards becoming an effective trainer, an expected FP role in SA. 55,56 In WBL, self-directed learning, reflection, and the ability to incorporate internal feedback at expected standards and external feedback from peers or supervisors 15 are vital for trainee personal and professional development. 16,57Reflection is essential to learning from experience 57 and it develops registrars as lifelong learners. 58Incorporating feedback, reflection and self-directed learning skills were variable among the registrars in this study, as perceived by FPs, but registrars thought they integrated feedback sufficiently into their learning.The ownership for self-directed learning, looking actively for learning opportunities, reflection and feedback, improves during training because of trainees' increased awareness. 28The senior registrars in our study commented more than juniors on their reflection-on-action, on patient interactions after their daily work, looking into the supervisors' feedback and addressing knowledge gaps.Despite these processes being only learnt during training, FPs expected registrars to engage with feedback and reflection from the time they joined the programme, as found previously. 28Instead, supervisors could provide more guidance to registrars with various WBL strategies earlier in their training.The participants agreed that workplace artefacts such as learning portfolios, case logs and reflective logs were excellent WBL tools, as seen in other studies. 16,30he portfolio was only used as a last-minute paper-based exercise for yearly submission but was not efficiently engaged for deeper learning by reflection. 30Learning portfolios augment trainees' long-loop learning by engaging supervisors or peers when challenged by complex patients, 12 which was not evident in this study.
Compared with previous SA studies, 30 workplace-based learning opportunities occurred during patient interactions, integrating feedback and FP engagements.Additional enablers in this study were peer learning, student training, and attending district activities such as CME and mortality and morbidity meetings.Learning occurred when exposed to patient loads based on community needs, experience with undifferentiated and holistic care and adequate hands-on practice, as found before. 59Resource challenges and a lack of district and university support 29 also emerged as findings.Participation in district clinical activities was a major contributor to WBL, including community-oriented primary care, mortality and morbidity meetings and quality improvement projects.There were primary planned opportunities with FPs, but secondary opportunities occurred while engaging and immersing in the clinical workplace during patient interactions, as described in studies conducted in high-income countries. 60While FPs are expected to fulfil leadership and governance roles in SA, 6 this study showed that registrars were not provided sufficient opportunities to learn or practice leadership and governance roles during training.
Self-directed learning behaviours and professionalism were variable among registrars, although FPs and registrars differed in their opinions.Trainees' self-motivation 17 and positive approach to feedback augmented self-directed learning. 12Self-directed learning and professional behaviours are essential registrar competencies and part of the national programmatic learning outcomes. 6Professionalism and selfdirected learning are critical competencies to be attained by postgraduate FM trainees in many high-income countries (as prescribed in the Canadian competency FM framework 4,5 and Accreditation Council for Graduate Medical Education [ACGME] Program Requirements for Graduate Medical Education in FM 3 ).The availability of more explicit guidelines on programme requirements explaining how decentralised clinical training should be implemented would assist in a more standardised approach, as performed in high-income countries. 3,5More explicit guidelines on registrar professionalism, supervisor roles and characteristics, and supervisor-trainee relationships could be included.
A supportive learning environment requires good supervision opportunities and mutual observation, including modelling during practice, provision for narrative feedback, sufficient resources, additional organisational support, a manageable workload and time to reflect on patient interactions, 13,25 which were found challenging in these settings.To optimise WBL, the learning environment should have a manageable workload and should address trainees' well-being. 3,12,51The need for protected time is necessary, as work pressure impedes WBL. 26,27,51District management support is imperative for the growth of decentralised clinical training sites, 28 which were also challenging, according to participants.Inadequate FP trainers and material resources, such as essential equipment and mannequins, negatively impacted WBL.Adequate material resources, including training space with adequate lighting, less background noise, comfortable seating arrangements and equipment availability, are all prerequisites for a conducive learning environment, 15,26,27,51 the lack of which were found to be barriers.
The findings underscore the need for ongoing faculty development of supervisors focused on enhancing teaching and learning.Clinical educators typically have not undergone educational training in preparation for their supervisory roles and may thus have little understanding of relevant learning theories such as social learning theory 61,62 or how to promote self-directed learning.There needs to be greater awareness around and more opportunities for FPs to attend training that influence WBL and supervision.The courses are offered by the SA Academy on areas such as postgraduate supervision, supervisory feedback, WBL and WPBAs.Participation in fellowship or masters' programmes offered by the university could also improve clinical trainer skills as medical educators.Peer mentoring for personal and professional growth among trainers and trainees while working as a team in their workplace contexts should be encouraged.Faculty development should involve both 'local faculty' and 'extended faculty' 63 such as medical practitioners, nurses and multidisciplinary team members who play beneficial roles in registrar WBL enhancing interprofessional learning.For registrars, how to best utilise WBL opportunities by applying adult learning principles, feedback engagement and more reflective behaviours should be introduced and engaged earlier in their training.

Study limitations
Exploring the WBL opportunities in decentralised sites of one university may have affected the transferability of the results to other similar contexts, although it is possible to some extent in case studies. 36Previous experience with the study contexts for the primary researcher may have influenced the data collection and analysis, but measures were taken to minimise this.

Recommendations
Based on this study, we recommend maximising WBL opportunities and addressing contextual challenges for registrars.Peer learning and more supervisory engagement with mutual observation during trainees' clinical practice need to occur.Adequate usage of the learning portfolio as a reflective tool, reflection on the supervisor's feedback, and utilising interprofessional learning is encouraged.Participation in multiple FP roles and responsibilities, such as leadership and governance and staff and student training by registrars, can be more frequent.We recommend sufficient organisational support from district management and the university and improved resource availability to enhance WBL opportunities across individual sites.More explicit guidelines or policies on decentralised training programme implementation nationally could assist in optimising and even standardising WBL opportunities.
The in-depth understanding of WBL opportunities derived from this study will be integrated as a 'process' in the evaluation of the FM training programme using logic model.These findings, together, with those from other parts of the larger study (resource availability, postgraduate supervision and supervisory feedback) 32,33,34 will be evaluated as the inputs, processes and outputs of the logic model.The overall logic model improved understanding of these factors and their relationships, which will be utilised for improving the programme as a whole.

Conclusion
This
What do you think about the feedback given to you by the family physician?What do you think about the feedback given to your registrar by you?How do you reflect on the feedback given to you by the family physician?What do you think about the registrar reflection on the feedback given to them by you?What do you think are the challenges for your learning in the district?What do you think are the challenges for registrar learning at your district?FP, family physicians.
access to my supervisor, even outside our formal working time.So, I'm free to discuss cases that I encounter with him, so this gives me an opportunity to learn.I can go back and search in terms of resources on the case that I encountered … I did discuss it with him, and then with my supervisor, who will also give his input, so, from this is [sic] a very good learning opportunity.'(RG5) CME, continuing medical education; M and M, mortality and morbidity.FIGURE 1: Themes and subthemes.http://www.phcfm.orgOpen Access 'I have Registrars and FPs felt that registrars training junior doctors, supervising interns or students, and conducting student assessments provided registrars good learning opportunities: 'By giving an opportunity to [the] registrar to train other, for example, giving in-service training, training junior doctors, training interns, by giving them [the] responsibility to supervise, also that's going to help them.'(FP 2) 'There is something to learn in terms of teaching styles, how to conduct a teaching or training session, how to gauge the level of knowledge in terms of how you should focus your training for the different levels of students.Firstly, you won't treat the final year the same as a third-year or a clinical associate, first-year as a fifth-year medical student.So, you need to be able to gauge the content need(s) of the students.'(RG 1) study was an in-depth exploration of perceptions of postgraduate learning opportunities for FM training in the clinical workplace in SA.Self-directed learning, peer learning, student training and participation in district activities were identified as strengths.Interaction with supervisors, peers and other professionals could augment WBL opportunities.More reflection on supervisory feedback, registrar professionalism, learning portfolio utilisation and interprofessional learning is needed.Well-resourced facilities and exposure to various complex patients promote WBL, while excessive workloads, inadequate resources, and insufficient district management support impede learning.Strengthening the utilisation of available opportunities while addressing the challenges can maximise WBL in decentralised sites.Optimising learning opportunities in clinical environments provides superior learning experiences at sites, translating to better patient healthcare within communities.This study provides several areas for future research.Not only does it underscore the need for ongoing training programme evaluation, it creates possibilities for exploring the influence of respondents' background on WBL and cross-case analyses of WBL across different training districts and sites, both nationally and internationally.