Tracking university graduates in the workforce: Information to improve education and health systems in Tanzania

Transcription

1 Viewpoint Tracking university graduates in the workforce: Information to improve education and health systems in Tanzania Senga Pemba a, *, Sarah B. Macfarlane b,rosempembeni c, Alex J. Goodell d, and Ephata E. Kaaya e a Tanzanian Training Centre for International Health, Mlabani Passage, P.O. Box 39, Ifakara, Tanzania. pemba@healthtrainingifakara.org b School of Medicine, and Global Health Sciences, University of California San Francisco (UCSF), San Francisco, USA. c School of Public Health, Dar es Salaam, Tanzania. d Phillip R. Lee Institute of Health Policy Studies, UCSF. e School of Medicine, MUHAS. *Corresponding author. Abstract With a severe shortage of highly trained health professionals, Tanzania must make the best possible use of available human resources and support training institutions to educate more graduates. We highlight the overlooked but significant role of universities in collecting, managing, and using human resources data in Tanzania and in other countries struggling to build their health workforces. Although universities, professional councils, ministries of health, education, and finance, and non-governmental organizations in Tanzania all maintain databases that include details of health professionals education, registration, and employment, they do not make the information easily accessible to one another. Using as an example Muhimbili University of Health and Allied Sciences the leading public institution for health professions education in Tanzania we explore how training institutions can gather and use data to target and improve the quality of education for increasing numbers of graduates. We specifically examine the substantial challenge universities face in locating more members of each graduating class and conclude with recommendations about how the situation can be improved. Journal of Public Health Policy (2012) 33, S202 S215. doi: /jphp Keywords: health professions education; workforce; tracer study; graduate survey; human resources information systems; Tanzania r 2012 Macmillan Publishers Ltd Journal of Public Health Policy Vol. 33, S1, S202 S215

2 Tracking university graduates in the workforce Introduction Shortages of well-educated health professionals, especially in many low- and middle-income countries, impact health outcomes. 1,2 The World Health Organization (WHO), for instance, estimates that countries with fewer than 23 physicians, nurses, and midwives per inhabitants generally fail to achieve adequate coverage rates for selected primary health-care interventions such as provision of immunizations and family planning. These countries also fail to prevent, for example, maternal and neonatal deaths as prioritized by the United Nations Millennium Development Goals. 3 Tanzania, with an estimated 2010 population of 44.8 million, 4 has just 4.3 physicians, nurses, and midwives per inhabitants. 5 To address this severe shortage of health professionals, the government of Tanzania is urging universities to increase admissions of health professions students, and is supporting the growth of health professional schools in public universities; 6 private universities are opening additional professional schools. 7 The government has requested that all universities move to competency-based curricula, creating a new national context for teaching and assessing students and programs. To maintain educational standards as student numbers increase, and to assess the relevance and effectiveness of educational innovation, universities need to locate and follow up on their students after employment. Responding to the government s call for improving the relevance of higher education, Muhimbili University of Health and Allied Sciences (MUHAS) in Tanzania introduced competency-based curricula for all professional students entering programs in October To align the new curricula to national workforce needs, MUHAS attempted to assess the on-the-job performance of recent graduates, but encountered difficulties in locating them despite many potential sources of information. We describe the training and subsequent employment of health professionals in Tanzania, and why it is so important but has been so difficult for MUHAS to trace them. We then propose ways in which universities could collaborate with one another, and with other organizations maintaining records, to better contribute to the national goal of increasing the pool of well-prepared health professionals. Training and Employment of Health Professionals in Tanzania Nine Tanzanian universities educate dentists, doctors, environmental health scientists, nurses, and pharmacists. MUHAS, established in r 2012 Macmillan Publishers Ltd Journal of Public Health Policy Vol. 33, S1, S202 S215 S203

3 Pemba et al 1963, 9 the first publicly funded health sciences university, trains dentists (who earn a DDS after 5 years of study), environmental health scientists (BSc EHS; 3 years), medical doctors (MD; 5 years), nurses (BSc N; 4 years), and pharmacists (BPharm; 4 years). Each of the other eight universities (one public and seven private), established between 1995 and 2010, trains some of these health professionals. 10 After obtaining their professional degrees, graduates work for the Ministry of Health and Social Welfare (MOHSW) as interns at employment sites (mainly hospitals) certified to supervise them. After internship, the graduates register with their respective professional councils to obtain their licenses to practice. On successful completion of their internships, health professionals work either for the MOHSW, local government, faith-based or nongovernmental organizations, or the private sector. Graduate health professionals who practice at health facilities in Tanzania mainly work in referral/specialized hospitals, regional hospitals, or district hospitals. About 42 per cent of 227 hospitals in the country are run by the government, the remainder is run by faith-based or nongovernmental organizations, or by the private sector. 11 According to the MOHSW Human Resources for Health Strategic Plan , public health facilities operate with a shortage of 65 per cent of required health workers, whereas private health facilities operate with a shortage of 86 per cent 6 indicating that Tanzania s health system faces a significant human resources crisis that requires urgent measures. Figure 1 shows the distribution of skilled health workers in health facilities across 129 districts in Tanzania in 2006, demonstrating that they are heavily concentrated in urban areas. 11 Environmental health scientists and other graduates also take posts with the MOHSW, or in regional or district health management teams, serve as tutors in health training institutions, work for nongovernmental organizations, or work in mining and other industries. Pharmacists work in regional or district hospitals, for pharmaceutical manufacturers, or in private pharmacies. In order to specialize in public health or clinical fields, or to become researchers, professionals undertake further training in the country or abroad. Some leave their professions completely, and some leave the country to work abroad. The proportion of graduates who take any of these career paths is not known. S204 r 2012 Macmillan Publishers Ltd Journal of Public Health Policy Vol. 33, S1, S202 S215

4 Tracking university graduates in the workforce Figure 1: Distribution of skilled health workers in health facilities in Tanzania 2005/2006. Note: 1 dot is equivalent to 15 workers; dots are randomly distributed in a district and proportional to the number of workers in the district. In 2005/2006, there were the following skilled health workers in health facilities: 1346 medical doctors, or 3.5 per population. 104 dentists, or 0.3 per population. 292 pharmacists, or 0.8 per population nurses and midwives (very few who were graduates), or 38.9 per population assistant medical officers (5 years clinical training after secondary school graduation), or 3.8 per population clinical officers (3 years training after secondary school graduation), or 18.1 per population. Information Sources for Locating Health Professionals Working in Tanzania Although much is written about the shortage of health professionals and other health workers in Tanzania, 5,12,13 there is considerable uncertainty about how many there are and where they work, despite a number of potential information sources: Databases with information about the education and licensing of health professionals Universities maintain records of all of their graduates and submit this information to the Ministry of Education through the Tanzanian r 2012 Macmillan Publishers Ltd Journal of Public Health Policy Vol. 33, S1, S202 S215 S205

5 Pemba et al Commission for Universities (TCU), which maintains a database of graduates from all universities. Any contact details kept by the universities or the TCU rapidly become out of date, but the MOHSW, which oversees the internship period, maintains a register of all employed interns. As an indication of the availability of accurate data, the authors were unable to obtain from any single source the annual admission or graduation numbers for all health professional programs in Tanzania. The Medical Council of Tanganyika maintains a database of all dentists and medical doctors, and the councils for Environmental Health, Nursing, and Pharmacists maintain registers of their respective professionals, who qualified either within or outside Tanzania and are allowed to practice in Tanzania. For all professionals except pharmacists, whose council requires them to reregister once a year (to maintain their licenses), registration is a one-time event; none of the councils maintains up-to-date contact details of practitioners they have licensed to practice. All of the councils are housed in the MOHSW and are not autonomous, limiting the capacities of the councils to make independent decisions in crucial matters related to human resources for health. Databases with information about the employment of qualified health professionals in Tanzania The MOHSW is responsible for posting and keeping records of newly registered or licensed health workers who enter public service, and shares employment data with the President s Office (Public Service Commission), which maintains a database of all public service employees. Information about staff movement and salaries is also available at local government offices in each district through the Ministry of Finance. The Christian Social Services Commission (CSSC) keeps a database of employees of faith-based, non-governmental, and private organizations. 14 Information kept by these institutions varies in quality and is difficult for other institutions to access. Recognizing the inadequacy of the current situation, the MOHSW is working with the CSSC and partners to develop and implement a health worker tracking system. 15 In cases where graduates work elsewhere in Tanzania or outside Tanzania, there are no formal channels through which to locate them. S206 r 2012 Macmillan Publishers Ltd Journal of Public Health Policy Vol. 33, S1, S202 S215

6 Tracking university graduates in the workforce Surveys and censuses describing the distribution of health workers For planning purposes, the MOHSW relies on figures from a national census of health workers it undertook in and a facility survey it undertook in The facility survey provides valuable crosssectional information about the distribution of human resources across districts in Tanzania (Figure 1). Nationwide surveys and censuses cannot be repeated frequently because they are logistically difficult and expensive, but they can be supplemented by data from smaller ad hoc research surveys and from routine sources. Tracer studies undertaken by universities to follow up on graduates MUHAS has twice undertaken tracer studies to physically track its graduates to their places of employment to inform curricular revision. The tracer study approach, originating from the University of Kassel in Germany, 17 (Box 1) was first used in Tanzania by the University of Dar es Salaam (UDSM) Faculty of Engineering in The approach continued to be used by the Faculty of Commerce and Management in and by the Faculty of Education in In , UDSM undertook a university-wide tracer study in which MUHAS (then a constituent college of UDSM) participated. 19 MUHAS followed this with a second tracer study in 2009 which, as we describe in the following section, demonstrated the inadequacies of existing information sources for locating graduates. Box 1: Trace r studies to inform university education Universities use tracer (sometimes called alumni) studies 17 to assess the need for curricular change. By following up on their graduates by mail or in person at their place of employment tracer studies can assist universities to: (i) Understand graduates current employment, for example, by asking graduates about their perception of their preparedness for employment, job satisfaction, and job duties and other characteristics of their employment; or by asking graduates and their employers about whether graduates have the required competencies for their employment; and (ii) Understand the career choices that graduates make, including whether to work in a urban or rural setting; public or private setting; leave the profession or migrate out of the country; or how they decide upon a field of specialization. r 2012 Macmillan Publishers Ltd Journal of Public Health Policy Vol. 33, S1, S202 S215 S207

7 Pemba et al MUHAS 2009 Tracer Study: Challenges in Locating enough Graduates In 2009, MUHAS set out to assess the extent to which its education had equipped graduates for their current employment, in particular: (i) how well graduates thought they had been prepared for their work responsibilities; (ii) whether their supervisors felt that the graduates were able to perform as expected; and (iii) the graduates opinions about the curricula under review and the MUHAS educational environment. For this study, MUHAS identified from its records the names of students who graduated from its programs between 2007 and 2009 and attempted to locate them. But MUHAS could not immediately locate all graduates 2 years after they had graduated because: (i) MUHAS kept no alumni register with up-to-date contact details (although some schools maintained contact with some students); (ii) professional organizations could not provide up-to-date lists of health professionals registered with them (even the details from the Pharmacy Council, with its annual re-registration, were not complete); and (iii) the lists of employed health personnel that MUHAS obtained from the MOHSW did not contain accurate contact addresses for many of its graduates. As it was not possible to locate sufficient contact details of graduates working in Tanzania to form a sampling frame for the study, MUHAS faculty and student survey teams visited nine sites in regions with referral hospitals where they expected most graduates to be employed. Once at a location, the teams generated a snowball sample, 20 an approach used by other tracer studies by asking key district and regional employers for the names of graduates to interview, and then asking the graduates themselves for colleagues to interview. Despite these efforts, the process was far from adequate, resulting in the teams interviewing only a proportion of each type of graduate (for example, 50 per cent of the dentists and 30 per cent of the doctors), with the selection process likely introducing biases to the responses. If MUHAS s tracer study had been a dedicated research project, its investigators could have spent more time trying to trace graduates physically and through social networks; 24 but the study had to be undertaken while faculty were on vacation and before imminent curricular revision. Availability of routine records of graduates would have raised the potential survey response rate. In Uganda, for instance, a tracking study conducted to identify bottlenecks and constraints in the S208 r 2012 Macmillan Publishers Ltd Journal of Public Health Policy Vol. 33, S1, S202 S215

8 Tracking university graduates in the workforce production, recruitment, and management of health workers and funds for the wage bill in the public health services found that out of 931 graduates who came from health training institutions between 2001 and 2003, only 269 (29 per cent) could be tracked. 25 From this study, it was concluded that identifying graduates largely depended on the comprehensiveness of the records left behind and employment opportunities for each category. The more outlets there were, the more difficult it was to locate an individual. These results, which are similar to the MUHAS tracer study findings, clearly show the need to plan tracer studies properly and to collaborate closely with other stakeholders when executing such studies. Need for National Collaboration to Trace Health Professionals in Tanzania Using figures provided by the TCU (to Author RM), we estimate that in 2010 nine Tanzanian universities accepted about 20 dental students, 40 environmental health science students, 940 medical students, 400 nursing undergraduates, and 90 pharmacy students, and these numbers are rising. Unless better mechanisms to trace them than were available to MUHAS in 2009, it seems unlikely that universities will know where these students careers take them once they graduate (in 2014 or 2015). But three significant developments show promise for enabling universities, professional councils, and employers to trace health professionals during the coming years: (1) Universities have identified the need to trace graduates to evaluate curricula. The government intends that universities will continue to train more health professionals and introduce competency-based education. In 2015, cohorts of more than 300 students will begin to graduate from the first such program of major educational innovation at MUHAS. To evaluate the effectiveness of these curricula, MUHAS intends to continue developing the practice of tracing its graduates routinely, but with a reliable way of locating as many of them as possible. Other Tanzanian universities are also revising their curricula; so it would be efficient to create a single system for locating graduates wherever they trained. (2) Universities and employers recognize the urgency for professional councils to license health professionals for limited terms, with r 2012 Macmillan Publishers Ltd Journal of Public Health Policy Vol. 33, S1, S202 S215 S209

9 Pemba et al renewal dependent on evidence of professional development. For several years, universities and employers have expressed concerns on how professional competence could be maintained without adequate licensing procedures. The open source software ihris Qualify, developed by the Capacity Project (dedicated to strengthening human resources in countries where there are shortages) 26 would make it straightforward for the medical council to set up such a database. 27 Other councils could build similar databases or collaborate in setting up a database for several professions. (3) The MOHSW is rolling out the Tanzania Human Resources for Health Information System (THRHIS) in collaboration with the Prime Minister s Office, the Ministry for Regional Administration and Local Government, the President s Office (Public Service Commission), the CSSC, non-governmental and donor partners, the UDSM, and MUHAS. 14,15 THRHIS, a modified version of ihris, 27 has been piloted in a number of districts. With time, it should be possible to locate health professionals who practice in any participating health facility in Tanzania. If the universities, the MOHSW, other employers, and the professional councils collaborate to ensure they can trace health professionals from university to licensing and employment, they will be able to follow up on graduate cohorts in order to: (i) assess how universities might enhance their education to improve the quality of the public health interventions and clinical services their graduates provide; (ii) review the extent to which and how the MOHSWand other employers could improve the circumstances under which the graduates are expected to perform; (iii) identify priority topics that professional councils need to require more learning of practitioners to maintain their licenses; and (iv) develop joint plans for training and employing well-qualified health professionals including providing continuing professional development. Weak Health Resource Information Systems Undermine Health Systems in Africa Human resources for health (HRH) data for policy and practice including strategies for the training, deployment, and retention of health S210 r 2012 Macmillan Publishers Ltd Journal of Public Health Policy Vol. 33, S1, S202 S215

10 Tracking university graduates in the workforce workers 28 are essential in strengthening national health systems to deliver quality health services. Universities need to understand whether the skills, knowledge, and attitudes students acquire through education are appropriate for performing well in employment. Successful tracking of health professionals can provide valuable information but requires that: (i) universities, professional councils, ministries of health, and other employers maintain electronic data bases and make them accessible to each other; and that (ii) these institutions work together to commission researchers to use the records to answer questions of common concern about the generation and employment of health professionals throughout their careers. Most African countries are challenged by weak and inadequate human resources for health information systems (HRHIS). 28 WHO and the Global Health Workforce Alliance are promoting the development of HRHIS, for example, through health workforce observatories. 29 In some African countries, these observatories coordinate stakeholders to produce and share information and build capacity to improve workforce development. Tanzania has yet to establish such an observatory. In a global review of information systems for human resources, findings indicate that little is known about the development, capacity, and use of HRHIS. Riley et al report that the majority of countries experiencing crisis levels of HRH shortages (56 per cent) did not report data on health worker qualifications or professional credentialing as part of their HRHIS. 27 There are some success stories. Riley et al describe the development of a nursing database in Kenya through which it is possible to track nurses from graduation through registration and employment. 30 The Uganda Nurses and Midwives Council has developed a database and demonstrated its value for monitoring the training, licensing, and employment of nurses. 31 Other countries and institutions can adapt and replicate these efforts to improve their HRHIS. For universities to provide the best educational experience for changing circumstances, longitudinal observations of the performance of graduates in varied employment situations will be the next challenge. Health professionals can be traced after graduation in countries where professional councils require their members to keep their licenses up to date. In 1997, Dovlo et al, for example, traced 9 years of graduates from the University of Ghana Medical School by cross-referencing with records from the Ghana Medical and Dental Council, which is updated r 2012 Macmillan Publishers Ltd Journal of Public Health Policy Vol. 33, S1, S202 S215 S211

11 Pemba et al yearly. 23 In two South African studies, undertaken in the late 1980 s and early 2000 s, researchers contacted medical graduates directly after being given their addresses and telephone numbers by the South African Medical Association, which requires physicians to update their registrations annually. 32,33 In an article by Goldacre et al published in 2009, researchers traced 28 years of United Kingdom (UK) medical graduates by cross-checking university records with the UK General Medical Council records, which are also updated annually. 34 Conclusion In Tanzania, universities, the MOHSW and professional councils need information about numbers graduating, their career choices, and the proficiency of graduates as they enter the national workforce. Such information will boost the government s ability to deploy and support new entrants to optimize the benefit of the health system for improving the health of all Tanzanians. Health professional schools in high-income countries compete with each other to meet national institutional accreditation standards but do not necessarily feel obliged to demonstrate that their education impacts health outcomes. Instead, they assume that the quality of practice will be monitored by employers and through continuing professional board exams. The Lancet Commission report on the education of health professionals called on schools across the world to radically revise traditional methods of education, attuning the new approaches to the desired outcome of improved population health. 35 Thus, it is important that schools evaluate their programs, as an essential step in a complex chain. To reach the ultimate health goals, we need to learn about graduates, and link that to improving understanding of health system structure and operations, and health outcomes. About the Authors Senga Pemba, M.Med.Ed, PhD, is a Director of the Tanzanian Training Centre for International Health and an Associate Professor of Medical Education at Saint Francis University College of Health and Allied Sciences in Ifakara, Tanzania. Sarah Macfarlane, PhD, MSc, is a Professor in the Department of Epidemiology and Biostatistics in the UCSF School of Medicine. She was S212 r 2012 Macmillan Publishers Ltd Journal of Public Health Policy Vol. 33, S1, S202 S215

12 Tracking university graduates in the workforce the UCSF principal investigator of the Academic Learning Project and directs the MUHAS UCSF Partnership for Global Health Sciences. Rose Mpembeni, MSc, PhD, is a Lecturer in the Department of Epidemiology and Biostatistics in the School of Public Health and Social Sciences at MUHAS. Alex Goodell, BS, is an Analyst at the Institute for Health Policy Studies at the University of California San Francisco. Ephata Kaaya, MD, PhD, is the Director of Continuing Education and Professional Development and a Professor of pathology in the School of Medicine at MUHAS. He was the MUHAS Principal Investigator of the Academic Learning Project and coordinates the MUHAS UCSF Partnership for MUHAS. References 1. Anand, S. and Bärnighausen, T. (2007) Health workers and vaccination coverage in developing countries: An econometric analysis. The Lancet 369(9569): Speybroeck, N., Kinfu, Y., Dal Poz, M.R. and Evans, D.B. (2006) Reassessing the relationship between human resources for health, intervention coverage and health outcomes. Background paper prepared for The World Health Report, relationship.pdf, accessed 4 September World Health Organization. (2006) World health report 2006: Working together for health. Geneva: World Health Organization, accessed 4 September United Nations, Department of Economic and Social Affairs. Country Profile: United Republic of Tanzania, accessed 4 September Kurowski, C., Wyss, K., Abdulla, S. and Mills, A. (2007) Scaling up priority health interventions in Tanzania: The human resources challenge. Health Policy and Planning 22(3): United Republic of Tanzania Ministry of Health and Social Welfare. (2008) Human Resource for Health Strategic Plan , Sector_Documents/HRH_Documents/Human_Resources_for_Health_Strategic_Plan-_ _.pdf, accessed 4 September Devarajan, S., Monga, C. and Zongo, T. (2011) Making higher education finance work for Africa. Journal of African Economies 20(Supplement 3): iii133 iii Ngassapa, O.D. et al (2012) Curricular transformation of health professions education in Tanzania: The process at Muhimbili University of Health and Allied Sciences ( ). Journal of Public Health Policy 33(S1): S64 S Mkony, C.A. (2012) Emergence of a university of health sciences: Health professions education in Tanzania. Journal of Public Health Policy 33(S1): S45 S63. r 2012 Macmillan Publishers Ltd Journal of Public Health Policy Vol. 33, S1, S202 S215 S213

13 Pemba et al 10. Kaaya, E.E. et al (2012) Educating enough competent health professionals: Advancing educational innovation at Muhimbili University of Health and Allied Sciences, Tanzania. PLoS Medicine 9(8): e Ministry of Health and Social Welfare [Tanzania and Zanzibar] and World Health Organization. (2007) Tanzania Service Availability Mapping Geneva, Switzerland: World Health Organization, Monitoring Evaluation/SAM_CountryReport_Tanzania_Final.pdf, accessed 4 September Kurowski, C., Wyss, K., Abdulla, S., Yémadji, N. and Mills, A. (2004) Human resources for health: Requirements and availability in the context of scaling-up priority interventions in low-income countries Case studies from Tanzania and Chad?, cealliance/knowledge/resources/priority_interventions_tanzaniachad/en/index.html, accessed 6 September Kolstad, J.R. (2011) How to make rural jobs more attractive to health workers: Findings from a discrete choice experiment in Tanzania. Journal of Health Economics 20(2): US Agency for International Development. (2009) Human Resources-Geographical Information Systems Data Development and Systems Implementation for the Christian Social Services Commission of Tanzania: Final report. US Agency for International Development, accessed 4 September Japan International Cooperation Agency (JICA). (2011) Strengthening leadership and ownership of HR information management. United Republic of Tanzania, accessed 4 September Ministry of Health Tanzania. (2004) Results of the 2002 Health Worker Census. Dar es Salaam, Tanzania: Ministry of Health. 17. Schomburg, H. (2003) Handbook for Graduate Tracer Studies, proj/edwork/handbook.ghk, accessed 4 September Kaijage, E.S. (2001) Knowledge and skills of B.com graduates of the Faculty of Commerce and Management, University of Dar es Salaam in the job market, accessed 4 September Ishumi, A.G.M. and Mkude, D.J. (2004) Tracer Studies in a Quest for Academic Improvement: The Process and Results of a University-Wide Tracer Study Project Conducted in Dar es Salaam, Tanzania: University Press. 20. Robson, C. (2002) Real World Research: A Resource for Social Scientists and Practitioner- Researchers: Hoboken. New Jersey: Wiley-Blackwell. 21. Shongwe, M. and Ocholla, D. (2011) A tracer study of LIS graduates at the University of Zululand, Paper presented at the 6th Biennial ProLISSA Conference; 9 11 March, Pretoria, South Africa, University of Zululand %20and%20Ocholla%20PROLISSA%20paper%20revised%20March%2029.pdf, accessed 4 September Msuya, J. (2005) The information studies program of the University of Dar es Salaam: Perceptions of stakeholders. Information Development 21(3): Dovlo, D. and Nyonator, F. (1999) Migration by graduates of the University of Ghana Medical School: A preliminary rapid appraisal. Human Resources for Health Development Journal 3(1): Al-Samarrai, S. and Bennell, P. (2007) Where has all the education gone in sub-saharan Africa? Employment and other outcomes among secondary school and university leavers. The Journal of Development Studies 43(7): Ministry of Health (Uganda). (2004) Tracking Human Resources and Wage Bill Management in the Health Sector: A Study to Identify Bottlenecks and Constraints in the Production, S214 r 2012 Macmillan Publishers Ltd Journal of Public Health Policy Vol. 33, S1, S202 S215

14 Tracking university graduates in the workforce Recruitment and Management of Health Workers and Funds for the Wage Bill in the Public Health Services. Kampala, Uganda: African Medical and Research Foundation. 26. US Agency for International Development TCP. (2009) Planning, Developing and Supporting the Health Workforce: Results and Lessons Learned from the Capacity Project, Chapel Hill, NC: IntraHealth International, option=com_content&task=view&id=16&itemid=33, accessed 4 September Capacity Project. Free and open health workforce information software, accessed 4 September Riley, P.L. et al (2012) Information systems on human resources for health: A global review. Human Resources for Health 10(1): Regional Office for Africa WHO. (2008) Concept of the Africa Health Workforce Observatory. Brazzaville, Republic of the Congo 2008, Document_Centre/concept_ahwo_eng2.pdf, accessed 4 September Riley, P.L. et al (2007) Developing a nursing database system in Kenya. Health Services Research 42(3p2): Spero, J.C., McQuide, P.A. and Matte, R. (2011) Tracking and monitoring the health workforce: A new human resources information system (HRIS) in Uganda. Human Resources for Health 9(1): Price, M. and Weiner, R. (2005) Where have all the doctors gone? Career choices of Wits medical graduates. South African Medical Journal 95(6): Zwarenstein, M., Youngleson, M. and Botha, J. (1989) Validity of the register of medical practitioners for manpower planning. Results and recommendations from a postal sample survey. South African Medical Journal 76(3): Goldacre, M.J., Davidson, J.M. and Lambert, T.W. (2009) Retention in the British National Health Service of medical graduates trained in Britain: Cohort studies. British Medical Journal 338, doi: /bmj.b Frenk, J. et al (2010) Health professionals for a new century: Transforming education to strengthen health systems in an interdependent world. Lancet 376(2010): r 2012 Macmillan Publishers Ltd Journal of Public Health Policy Vol. 33, S1, S202 S215 S215