Stakeholders’ Perceptions on Shortage of Healthcare Workers in Primary Healthcare in Botswana: Focus Group Discussions

Background An adequate health workforce force is central to universal health coverage and positive public health outcomes. However many African countries have critical shortages of healthcare workers, which are worse in primary healthcare. The aim of this study was to explore the perceptions of healthcare workers, policy makers and the community on the shortage of healthcare workers in Botswana. Method Fifteen focus group discussions were conducted with three groups of policy makers, six groups of healthcare workers and six groups of community members in rural, urban and remote rural health districts of Botswana. All the participants were 18 years and older. Recruitment was purposive and the framework method was used to inductively analyse the data. Results There was a perceived shortage of healthcare workers in primary healthcare, which was believed to result from an increased need for health services, inequitable distribution of healthcare workers, migration and too few such workers being trained. Migration was mainly the result of unfavourable personal and family factors, weak and ineffective healthcare and human resources management, low salaries and inadequate incentives for rural and remote area service. Conclusions Botswana has a perceived shortage of healthcare workers, which is worse in primary healthcare and rural areas, as a result of multiple complex factors. To address the scarcity the country should train adequate numbers of healthcare workers and distribute them equitably to sufficiently resourced healthcare facilities. They should be competently managed and adequately remunerated and the living conditions and rural infrastructure should also be improved.

through a network of 3 national referral hospitals, 7 district hospitals, 14 primary hospitals, 265 primary care clinics (101 with maternity beds), 343 health posts and 861 mobile clinic sites [9]. Management of the health services, including recruitment and deployment of HCWs, procurement and distribution of equipment and drugs, is centralised at the Ministry of Health headquarters. District health management teams (DHMTs) were created in 2010 to manage healthcare services in their respective districts. The DHMTs work closely with the tribal, political and executive leadership of their districts to enable inter-sectorial coordination of services.
Until 2007, training of doctors, dentists, laboratory scientists, pharmacists, physiotherapists and other allied health professionals was outsourced to universities in the region and overseas. The training of doctors and laboratory scientists was localised in 2009 through the creation of the University of Botswana Faculty of Health Sciences in Gaborone.

The study design
Fifteen focus group discussions were facilitated with policy makers, HCWs and members of the community from March to May 2012 in all three districts chosen for the study.

Sampling strategy
Separate meetings were held with policy makers, HCWs and members of the community in January and February 2012 in each of the three districts to recruit participants for the interviews. In Gaborone only one combined meeting was held with all groups. The following criteria were used to purposively select people to attend the meetings: 1. Policy makers were defined as administrative, political, or traditional leaders in the districts whose work had an impact on healthcare. District executive managers (district commissioners) and DHMTs were invited as administrative leadership in the district. The DHMTs included senior doctors, nurses, other HCWs and health administrators in the district. In Gaborone, the capital city, the national human resources manager and primary healthcare coordinator were also invited. Village health committee, village development committee and district council chairpersons were invited as political leaders and local tribal chiefs as traditional leaders. They were included as policy makers and not in the community groups as they are part of the district inter-sectorial leadership. All of the designated people above were invited to the meeting.
2. HCWs were defined as public employees involved in primary healthcare. These comprised doctors, dentists, nurses, community health workers, lay counsellors, pharmacy technicians, laboratory technicians, health education assistants and family welfare educators. Invitations were sent to all the primary healthcare facilities in the district.
3. Members of the community were defined as people that had lived in the district for at least 6 months. Invitations were extended to those in organised community groups and structures such as village health committees, home-based care volunteers, and village leadership (including village development committee members and village headmen). Invitations were sent to all villages in the district.
At the meeting the study was explained and all attendants were invited to participate. All attendants were asked to identify additional suitable people that could be invited should more be needed for the interviews or who could replace them should they be unwilling or unable to participate (Fig 1).

Data collection process
A series of 15 focus groups, comprising 5 to 12 participants each, were conducted in the three districts. Each district had one focus group with policy makers, two with HCWs and two with community members. We conducted the interviews using a semi-structured interview guide. Each focus group was held in a quiet and private space and lasted between 2 to 3 hours. The groups discussed the following questions: 1. In your opinion, is there a shortage of healthcare workers in primary healthcare in Botswana?
2. What do you believe are the causes of any shortage?
3. What do you think can be done to address any shortage of healthcare workers in primary healthcare?
The interview guide was piloted with members of the university community (HCWs and support staff) and revised based on their feedback. The discussions with community members were in Setswana (the national language of Botswana) and the other group discussions used both Setswana and English. ON (female, MD) facilitated 12 of the 15 interviews while NP (female: PhD, Nursing) and SS (female, PhD, Nursing) facilitated the remaining three. The discussions were audio-recorded and transcribed verbatim. All the members of the team were trained in qualitative interviewing and were bilingual (English and Setswana). The Setswana documents were translated into English before analysis.

Ethical considerations
Ethical clearance to conduct the research was obtained from the University of Botswana Institutional Review Board and the Ministry of Health Research and Development Unit: Reference No: PPME 13/18/1 V11 (368). Written informed consent was obtained from each study participant.

Data analysis
Data analysis was performed using the framework method [13][14][15]. ON developed a thematic framework, with supervision from RM, using two of the transcripts. ON then applied the index to code all the remaining 13 transcripts. The themes were derived inductively from the data. ATLAS.ti software (version 7.1.3) was used to manage the transcribed data and to expedite data analysis. No new themes were identified after analysing the first seven transcripts; nevertheless, all the transcripts were analysed. The number of times that a code, relating to particular theme, was used in the 3 groups was also tabulated to indicate the relative strength of the item in the respondent groups. The emerging themes were also presented to and discussed with the different participant groups for respondent validation [16].

Findings
Fifteen focus group discussions (Table 1) were conducted with a total of 133 participants: 46 HCWs, 27 policy makers (PMs) and 60 community members (CMs). The average length of service of HCWs at the districts was: Gaborone (G) 13 years, Mahalapye (MH) 10 years and Maun (M) 8 years. In the quotes below the voice of the participant can be identified by using the abbreviations above for the district and type of participant.
The themes that emerged from the data are presented in three sections below as they relate to the perceived shortage of HCWs in PHC, the reasons for shortages of HCWs and suggested strategies to rectify the situation.

Shortage of healthcare workers in primary healthcare
All the groups believed there was a shortage of HCWs, more especially of Setswana-speaking doctors, midwives, and specialist nurses ( Table 2)

Reasons for the scarcity of healthcare workers in primary care
Increased need for health services, inequitable distribution, migration, and training inadequate numbers were blamed for the deficiency of HCWs in PHC (Table 3). The participants argued that population growth, the HIV/AIDS epidemic, and increasing prevalence of non-communicable diseases led to the establishment of new clinics and services, without equivalent increases in the number of HCWs: Nonetheless, some HCWs alleged that allocation was reasonable, but there was higher attrition in rural areas: '. . . they are distributed equally, but . . . they leave the service in high numbers . . . in rural areas . . . compared to urban areas.' (GHCW1)[S7 Text] The problem of training inadequate numbers of HCWs was reported by all the groups, who also claimed that those trained in other countries did not return to work in Botswana:  Migration of HCWs from the public health system was held as a worrisome contributor to HCW scarcity in rural areas by all the groups ( Table 4). The migration was mainly to private health services, to urban areas, and to other countries: The study participants acknowledged that ineffective healthcare and human resources management, inadequate incentives, personal and family issues coupled with socioeconomic and contextual challenges contributed to the movement of HCWs from PHC.
Healthcare management was deemed weak and ineffective by all the groups. This resulted in demoralising working environments characterised by poor infrastructure and absence of critical resources such as patient transport, medicines and equipment: Human resources management practice was a matter of concern for all the groups, especially HCWs and PMs, who maintained that policies on the length of rural deployment, transfer, promotion and continuing professional development were unclear and not always informed by the need for health services:  These policies were also inconsistently and unfairly applied and often favoured those working in hospitals: '. . . you find that . . . more health workers working at the hospitals are promoted unlike in clinics . . .' (GHCW5)[S7 Text] Additionally, a lack of supportive supervision was of particular concern to HCWs who were generally demoralised by the lack of appreciation, recognition and rewards: The participants recognised that the socioeconomic obstacles to retention of HCWs could not be tackled by the Ministry of Health alone, but required a concerted multi-sectorial approach. However, they wanted appropriate accommodation to be provided to HCWs as a matter of urgency: '. . .sometimes you may find that an issue is beyond the scope of . . . the ministry . . . example of electricity maybe or telephone . . . water . . . so if these resources are not available in the clinic you may have the best nurse and all the drugs, but find a fridge does not work . . . Furthermore, they argued that differential incentives should be given to midwives and nurses with additional qualifications or who perform non-nursing duties: '. . . government should start giving incentives for some of these courses, like midwifery. Midwifery . . . is a high risk job . . .' (MHCW5).[S1 Text] The participants proposed that human resources management policies on promotion, transfer, length of service in rural areas and in-service training needed to be clarified and applied fairly, consistently and transparently. They also insisted that HCWs should not work in rural areas for a very long time. In addition, they recommended that the career structures for nurses be reformed: '. . . you need to make a clear policy of transfers, because if you send someone to a rural area he has to know and be prepared that I'm going . . . to provide service [for] two years . . .' (MHCW3)[S1 Text] HCWs also wanted their supervisors to visit them regularly to appraise their work accurately, understand their work challenges and reward them appropriately: '. . . how do we reward people for putting in long hours and working so hard . . .your employees when they've done a good job, do you congratulate them or do you always look at the bad things . . .' (GHCW6)[S7 Text] All the participants requested role clarification and adequate supervision, especially for junior and inexperienced staff.
The participants advocated for differential increases in the numbers of HCWs trained, based on the healthcare needs of the country. They also proposed that training should instil appropriate knowledge, skills and attitudes:

Discussion
To our knowledge this study is the first to explore the perceptions of policy makers, HCWs and the community on human resources for primary healthcare in Botswana. The number of HCWs in PHC were perceived to be inadequate, a perception that is confirmed by existing evidence [9]. That notwithstanding, Botswana, an importer of healthcare workers from other countries [3] has more healthcare workers per population than many low and middle income nations [8].
Like many other African countries, Botswana is perceived to be failing to produce enough health professionals for its health systems [17,18]. This situation has been exacerbated by failure to retain those that were trained in the public health system [9,19].
Botswana's situation has been further complicated by its policy of outsourcing the training of many health professionals to other countries, with the result that the majority have remained in the host countries after graduation [9]. There is hope, however, that the opening of a medical school and a faculty of health sciences for the training of doctors and other HCWs at the University of Botswana will address this challenge. Local production of HCWs in Botswana, as in many other countries in sub-Saharan Africa, is however hampered by the limited of capacity of training institutions. [17,18,20]. The policy makers and healthcare workers were of the opinion that the training of the healthcare workers was not informed by the needs of the health system.
Inequitable distribution, favouring hospitals and urban areas, and higher attrition rates in rural areas have resulted in fewer HCWs in PHC and rural areas [9]. This 'inverse care law' [21] besets many health systems, especially in low-and-middle-income countries [1,22,23]. Similar to other southern African countries, Botswana's need for HCWs has also grown, primarily because of the HIV/AIDS epidemic, which has placed increasing demands on the healthcare system [1,24].
Despite its being a destination of migrant health workers from the region [3,9], Botswana suffers from both internal and external migration of HCWs, a problem which plagues many low-and-middle-income countries [25,26]. The determinants of migration are multiple, but personal and family issues stand out as a priority. Living away from the family is significantly associated with attrition of HCWs from PHC and rural areas [22]. HCWs also migrate because of a dearth of opportunities for career progression and continuing professional development, especially in rural areas [19].
Weak and sometimes incompetent healthcare and human resources management, characterised by lack of essential resources for patient care, poor career structures, unclear policies, absence of supportive supervision as well as poor conditions of service, were identified as 'push factors' for HCW migration. This finding is in keeping with studies on migration of African HCWs [19,27,28].
HCWs are also reluctant to work in rural areas because of poor infrastructure such as roads, schools for children, accommodation and telecommunication, and limited opportunities for career and continuing professional development. Deficient development has also been blamed for the failure of many countries to staff rural healthcare facilities adequately [7,22,23].
Incentives and remuneration packages were deemed unattractive and non-competitive compared to those in Namibia and South Africa. Inadequate incentives and low salaries are known to be strong 'push' factors for migration of health professionals [7,23]. Conversely, the hope of 'greener pastures' is a 'pull' factor to high-income countries.
Our study suggested that in order to tackle staffing insufficiency in PHC, multiple barriers must be addressed simultaneously [28,29]. Few such strategies have been evaluated in Africa, but Ghana's monetary incentives and Zambia's more comprehensive strategies have had some success [23,25].
Interviewing the three groups from three different settings enabled the exploration of different perspectives, varying experiences of the healthcare system and triangulation of themes. The study was conducted in three of the 28 health districts only and it is possible that other viewpoints would have been obtained elsewhere. Given the consistency of responses and saturation of themes, as well as the exposure of respondents to other districts in previous postings, we believe the results are transferable and likely to be replicated throughout Botswana.

Implications and recommendations
Various recommendations can be made based on this research, as suggested below.
Capacity should be built in human resources for health, healthcare management and systemic monitoring and evaluation. Botswana should also institute strategies to train and retain more healthcare workers. The training should be informed by needs of the health system and also inculcate appropriate competencies [30]. The retention strategies should include better incentives including fiscal, accommodation, continuing professional development opportunities, and improvement of rural infrastructure. Moreover, the capacity of DHMTs to provide supportive supervision to HCWs in PHC should be strengthened.

Conclusion
Botswana, despite its relatively high expenditure per capita on health [8], has a shortage of healthcare workers, which is worse in primary healthcare and rural areas, as a result of multiple complex factors. The causes of the deficiency of health workers are similar to those of countries which spend much less on health, some of which are also sources of the Botswana health workforce [9,23]. To address the scarcity the country should train adequate numbers of competent HCWs and distribute them equitably to sufficiently resourced healthcare facilities. These HCWs should be competently managed and adequately remunerated and the living conditions and rural infrastructure should be improved. S9 Text. Gaborone community group 2_File.pdf. (PDF) S10 Text. Gaborone community group 1_File.pdf. (PDF) S11 Text. Mahalapye healthcare workers group 1_File.pdf. (PDF) S12 Text. Mahalapye community group 1_File.pdf. (PDF) S13 Text. Mahalapye healthcare workers group 2_File.pdf. (PDF) S14 Text. Maun healthcare workers group 1_File.pdf. (PDF) S15 Text. Interview guide for policy makers and healthcare workers_File.pdf. (PDF) S16 Text. Interview guide for community members_File.pdf. (PDF)