Research | Open | Open Peer Review | Published:
A scoping review of the association between rural medical education and rural practice location
Human Resources for Healthvolume 13, Article number: 27 (2015)
Inequitable distribution of the medical workforce is an international problem that undermines universal access to healthcare. Governments in many countries have invested in rural-focused medical education programs to increase the supply of rural doctors.
Using a structured five-step approach, a scoping review was conducted to map the existing evidence on the relationship between professional entry-level, pre-vocational medical education delivered in rural settings and rural workforce outcomes. Key search terms were developed, with database searches yielding 37 relevant articles. During data charting, a set of types of studies emerged, and we developed a typology to assist with article sorting and information structuring.
Medical students attending a rural campus or spending time in a rural area are more likely to practise in non-metropolitan areas upon graduation than students studying at a city campus. In many cases, these positive findings could be confounded by students having a rural origin or being predisposed to want rural work. There is some evidence to suggest that the longer a person spends time as a medical student in a rural area, the more likely they are to work rurally following graduation. Overall, the articles located had limitations related to small sample size, inconsistent definition of rurality and lack of attention to controlling for variables that might influence rural practice decision, for example, rural background. Comparative data were lacking, and most studies were conducted by staff from the medical schools that were the focus of the research. There was no consideration given in any study found to the cost-effectiveness of entry-level medical education delivered in rural settings versus other ways of producing rural practitioners.
Given limitations, available evidence suggests that medical education in a rural location does increase the number of medical graduates that will work in a rural place. There are indications of a gradient effect where increased rural practice exposure during medical education leads to more rurally located graduates; however, robust studies are needed to verify this finding. Given the significant funding being directed to universities to increase graduates that will work rurally, appropriate future research is recommended.
In this article, we review what is known about an association between professional entry-level, pre-vocational medical education in rural areas and rural medical practice location following graduation. We address two research questions: does medical education in a rural place produce rural doctors? And, is there a gradient relationship between time spent in rural-based medical education and likelihood of working in rural practice? For the purposes of this review, our interest was rural practice following graduation from a medical school, so we did not include the impact of postgraduate study. We acknowledge that there exists evidence indicating diverse factors associated with practising rurally. In this article, our specific interest is in studies that explored associations between rural medical education and rural practice location of doctors.
Internationally, there is significant funding invested in initiatives to increase the numbers of medical graduates choosing rural practice. While there is a body of evidence that suggests rural upbringing as a predictor of rural medical practice [1-3], less attention has been given to specifically considering the effects of place of education.
Especially in a constrained financial environment, funding decisions about medical student programs should be based on sound evidence. The purpose of our review is to map the evidence and summarise what is known about relationships between entry-level, pre-vocational medical education delivered in rural places and rural work outcomes. Rural as a spatial categorisation is a contested concept and, internationally, is subject to multiple definitions . For the purposes of our work, we have included studies where the authors state their interest is rural and define or describe it. We allowed the use of synonyms—remote and regional.
Inequitable distribution of the medical workforce is an international problem that undermines universal access to health services [5,6]. Although direct evidence of causation between medical professional presence and better health is illusive, studies have associated higher physician–population ratios with higher overall state health rankings in the USA  and better self-reported general health in Canada . Places with poorer health and few medical practitioners tend to coincide. Given that service accessibility is an established social determinant of health , addressing medical workforce maldistribution is important.
Evidence shows that physician assistants, nurse practitioners and community health workers can substitute for high proportions of medical practitioner competencies , but medical doctors have the widest scope of practice  and are integral to healthcare provision in any context. Access to sufficient medical care, as a basic human right to achieve adequate health and well-being, has been enshrined in United Nations policy since 1948 .
Inequities in doctor distribution are most significant when rural–urban comparisons are made . Per 100 000 people, Canada has approximately 100 physicians in rural areas, compared with 150 for urban . Australian figures are 98.5 per 100 000 in very remote areas, compared with 266 for state capitals . A Japanese study showed 185 doctors per 100 000 in rural, compared with 272.5 for urban prefectures .
The rural general medical practitioner (GP) is skilled to deal with diverse encounters and contexts, clinical procedures, after-hours work, emergencies, complex care, public health issues and Indigenous peoples’ needs . Practising medicine in rural hospitals involves the following:
…[a] broad generalist set of skills, knowledge and attitudes…practiced at a distance from comprehensive specialist medical and surgical services and investigations. 
Rural medical staff
Historically, attracting rural medical staff to improve health and social conditions is a government concern. For example, the 1912 Dewar Report established the Scottish Highlands and Islands Medical Service, precursor to the UK National Health Service (NHS), to address poor rural health outcomes and a dearth of rural doctors .
Attracting medical graduates to work in rural areas is challenging, with recent United States (USA) research indicating only 3% of medical school entrants planned to practise in rural areas , compared with 19.3% of the USA population living rurally . The inability of governments to ensure sufficient entrants to rural medical practice has resulted in numerous initiatives aimed at increasing rural medical graduate supply—including bonded student places, clinical schools, workforce agencies, work incentives and medical undergraduate degree programs or pathways focused specifically on rural practice . In many countries, there has been major investment in such strategies, but the World Health Organisation (WHO)  notes a lack of robust evaluations and limited evidence for their effectiveness in increasing the overall supply of rural doctors.
Australia as an example: the expenditure–evidence gap
To grow rural doctor supply from domestic students, a strong focus of Australian investment has been the funding of city-based medical schools to develop rural components in degree programs. In the Australian context, initial medical education typically lasts from 4 to 6 years. Students enter undergraduate programs direct from secondary school, or graduate entry programs, where students are admitted on the basis of completing an earlier undergraduate degree. The rural component of Australian programs tends to be delivered largely through rural clinical schools. The University Departments of Rural Health and Rural Clinical School (UDRH/RCS) programme began in 1994 . The Australian Rural Clinical Training Scheme (RCTS) targets include the following: 25% of all medical students to have at least 1 year of rural clinical training by graduation, 25% of government-supported medical students to be recruited from a rural background, and all government-supported medical students to have at least 4 weeks of rural placement . Supporting a ‘trickle-down’ strategy , government-supported medical places have substantially increased, and from 2014, Australia will have tripled overall medical graduate numbers since 2001 (to around 3 800 a year) .
Noting proliferation of initiatives to encourage rural working, an Australian government review  identified the need for a better evidence base on rural medical education and systematic monitoring of rural career choices. The report is consistent with academic reviews that identified the need for ‘comprehensive, methodologically rigorous longitudinal studies’ of outcomes from rural medical investment .
New rural medical schools
In an era of fiscal constraint, Australian universities have attracted funding for new medical programmes built on a case of addressing rural disadvantage. ‘Rural’ medical schools have been developed in Geelong and Wollongong, although both are categorised as ‘major cities’ using the Australian Standard Geographical Classification . James Cook University medical school in Townsville, Queensland, and Flinders University Northern Territory Medical Program, based in Darwin, Northern Territory, are recently developed medical programmes located in ‘outer regional areas’.
Establishing medical programmes in rural places is not a purely Australian phenomenon. The Northern Ontario School of Medicine was established in 2005 as a ‘rural, distributed community-based medical school’ recruiting students from Canadian northern, rural, remote, Aboriginal or Francophone backgrounds . Exeter and Plymouth medical school (now two separate schools) was established in 2000 to provide medical education for the peripheral English southwest area.
Medical programmes in rural locations
An evidence base exists that identifies predictors for medical graduates choosing to practise rurally. Rural background and duration of time spent living in a rural place, prior to medical education, have been consistently described as strong predictors [26-28].
Considering associations between experience of living rurally and rural practice following graduation, Farmer et al.  use Bourdieu’s notion of habitus (expectations and understanding, based on experience of living in an environment with certain cultural features, that shape one’s sense of the social world) to explain why extended rural experience predicts comfort with rural work and living . Researchers have confirmed the importance of rural ‘acculturation’ in practice choices .
Building on evidence of associations between time residing rurally and eventual work outcomes, some argue that the key to increasing rural practitioners is to develop medical schools where all education is delivered in a rural place. The success of such programmes, for example, that of James Cook University, where 59% of 2005–2011 graduates primarily worked outside of major cities postgraduation , has fuelled interest in wholly rural medical education.
A polarised debate
There is vigorous debate about the need for new medical programmes located wholly in rural locations. Proponents argue rural location is a key feature for addressing workforce disparity, while metropolitan-based medical schools argue their programmes with rural components are effective and increasing medical student places is unnecessary . The issue is complex. Overall increases in medical student numbers, especially if exposed to rural experience, could lead to more rural doctors. However, actual effects of increasing graduate output on workforce distribution are unknown, partly due to the novelty of many initiatives.
With increasing higher education marketisation, the debate intensifies. Medical schools are attractive for universities in any country . Universities are attracted by high payments for training medical students (for example, in 2014, the Australian government paid universities $21,273 per medical student place, compared with $13,163 for nursing ). Internationally, universities are attracted to medical schools as their biomedical research produces high-research income, and international university rankings tend to be weighted in favour of biomedical research outputs . Across all countries, there is high demand for medical student places from a growing middle class .
In an environment characterised by these debates, having good, independent, evidence to inform funding decisions about medical school places would be useful. While the weight of evidence shows an association between rural upbringing and practising rurally [26-28], the impact of professional entry medical education in rural places on increased rural workforce capacity appears less clear. It is this opaqueness that prompted our review.
A scoping review was undertaken to identify what was known about the impact of professional entry medical education in rural places on rural medical practice location following graduation. Scoping reviews are an effective, robust approach to map, collate and summarise literature . This type of review is useful to map research areas, as unlike systematic reviews, all study designs can be included. The focus is on what knowledge exists, rather than on research quality . We used Arksey and O’Malley’s  five-stage framework for scoping reviews: identify the research question; identify relevant studies; study selection; chart the data; and collate, summarise and report results.
Defining the research question
Consistent with Arksey and O’Malley’s  recommendations, a broad question and key terms were used to ‘generate breadth of coverage’. Our aim was to maximise the range of literature captured, so the question, ‘What is known about the impact of professional entry medical education in rural places on rural medical practice location following graduation?’ guided the search strategy. Acknowledging graduate entry programmes, for students who already hold a degree, the term professional entry medical education was used to denote the first medical degree a person completes for entry to practise as a doctor.
Identifying relevant studies
In designing scoping reviews, there is a need to establish key search terms to achieve a balance between a rigorous review and practicalities such as time and cost . Here, we developed a Boolean string from key search terms and used truncated words and wildcards (in this instance *) to broaden the search and ensure all terms with the same root word were included. Search terms are in Table 1.
A preliminary search of Google Scholar showed the likely extent of findings and relevance of terms selected and helped to refine inclusion and exclusion criteria. The search results are not included due to the inability to accurately replicate Google Scholar searches . Inclusion and exclusion criteria, consistent with our review purpose, were developed (see Table 2). The search concluded in September 2014.
A search of the Cochrane Library identified no systematic reviews that were directly relevant to our research question. Databases searched included Medline, CINAHL, Proquest, Expanded Academic and Informit.
Using the developed search terms, 274 articles were identified, and after removal of duplicates, 202 remained. The title and keywords of the articles were compared with the inclusion and exclusion criteria, with the research team agreeing and confirming the elimination of irrelevant studies. Of the 157 articles that remained, the full-text versions of the articles were read by a minimum of three researchers. Through a process of matching against inclusion and exclusion criteria, 127 references were excluded. Primary reasons for exclusion were the following: literature review or commentary, the dependent variable was not rural practice, the authors did not quantify the time spent in rural areas or the authors considered intention rather than actual practice. The 30 identified articles that matched our inclusion and exclusion criteria were deemed directly relevant to our research question. We hand-searched reference lists from these articles and checked citations and identified seven further relevant articles. A total of 37 articles were included in our final review. To ensure that we had captured all articles, we completed an additional hand search of the reference lists of all articles that had been extracted for full-text review. No further articles were identified. The process of article selection is outlined in Figure 1.
Data charting and collation
Consistent with the fourth stage of Arksey and O’Malley’s framework, the following is reported for each article: author, publication year, location, sample, method, major findings, discussion points and limitations. During data charting, a set of types of studies emerged, and we developed a typology (see Table 3) to assist with article sorting and structuring information for the reader.
The results presented in each typology category are detailed in Tables 4, 5, 6, 7 and 8. In charting the data, we were focused on consistent reporting. All available details are included in Tables 4, 5, 6, 7 and 8. Where detail is missing, the issue was not addressed in the study reviewed.
Consistent with Arksey and O’Malley’s  framework, findings and reporting provide an overview of the articles selected for review. As the purpose of the scoping review was to identify what is known about an association between professional entry medical education in rural places and rural medical practice location following graduation, reporting focuses on the types (often related to durations) of rural medical education, the effect of this education on employment location following graduation, any key messages and limitations about the link between rural medical education and rural employment, as highlighted by study authors. We were interested in whether studies included discussion of cost-effectiveness, but following a review of all, we found none that considered this issue.
Typology category 1: rural versus urban campuses of the same university
In Table 4, we summarise studies that compare rural practice location outcomes for rural versus urban campuses of the same university.
Findings from the two studies reviewed [41,42] indicated that those who attended rural campuses are more likely to practise in non-metropolitan areas than those who attend the city campus of the same university. There are limitations to these findings, associated with student self-selection, with the hypothesis that students who choose to study in rural campuses have a predisposition to work outside main cities; for example, in the article by Crump et al. , 68% of students who attended the Trover campus came from a rural hometown.
Typology category 2: graduates of rural medical schools—medical education provided completely in a rural place
These articles demonstrate the diversity of ways that rural is depicted. Jichi Medical School  was described as located in a country town with a population of less than 20 000 approximately 100 km north of Tokyo. The Tromso medical school in Northern Norway  was described as 70° north of the Arctic Circle. In the article by Stratton et al. , North Dakota, USA, is described as ‘a rural state’.
Collectively, study findings indicate that, variously, around 30%–56% of graduates at the time of the studies were working in rural places: Jichi 42% , Tromso 56.1%  and North Dakota 29.5% . Sen Gupta et al.’s  study showed that 59% of James Cook University medical graduates had worked in non-city areas at some time during the first 7 years postgraduation. The researchers in each article note that the medical schools were developed specifically to address rural workforce challenges, to address the workforce needs of local ‘underserved populations’  and to ensure graduates for rural areas [43-45].
For Jichi  and Tromso , findings were not compared with rural doctor outcomes for other medical schools in the same country due to unavailable data. As with category 1 of our typology, success in delivering rural doctors could be attributable, to some extent, to students’ predisposition to work in rural areas. Differences in the percentage of graduates working in rural areas between Tromso at 56.1%  compared with North Dakota at 29.5%  could be linked to the measurement of different outcomes. For North Dakota, the statistics relate to working in communities with up to 25 000 residents, while for Tromso , it is Northern (less populated) as opposed to Southern (more populous) Norway. There are differences in programme focus. As an example, Stratton et al.  described the focus of their study as retention of North Dakotans in North Dakota, while Inoue et al.  were concerned with doctors working in rural prefectures.
Typology category 3: studies of the impact of partial rural medical education on rural medical recruitment
In Table 6, we summarise studies that consider the impact of medical education partially in a rural area on producing rural doctors.
These studies [19,46-61] consist of two main types: evaluations of the effects of special rural programmes (all from the USA) and evaluations of varying lengths of location in a rural area for GP attachments or attendance at ‘rural clinical school’. Fourteen of the 17 studies support positive effects of partially rural medical education on the production of rural doctors. Three studies (two Australian, one Canadian) indicate incremental effects of increasing time spent in rural places for medical education [46,47,61], although Orzanco et al.  only found this for one of the two universities included; for the other, findings were inconclusive. In New Zealand, Williamson et al.  did not find an association between rural placement and rural working. For the 11 studies of USA rural programmes, percentages of graduates who proceeded to work in rural places range from 26%  to 67% .
Some of the USA studies compare the effects of rural programme graduates versus non-programme graduates. Examples include Smucny et al.  (26% of graduates from a rural programme working in rural areas compared to 7% non-programme). Rabinowitz et al.  found that, after 11–16 years, 68% of rural programme graduates worked in rural areas, compared to 46% for non-programme graduates. Most USA rural programme studies compare the impact of rural programmes with a figure for aggregated national medical schools, variously stated as producing between 9% and 11% of all USA medical school graduates that work in rural areas.
Many of the USA programme studies described small numbers of graduates produced; for example, Lang et al.  report 134 graduates over 18 years, with Rabinowitz  reporting on a total dataset of 148 graduates. Several of the USA studies are of the same programme over time. In total, 16 of the 17 studies consider the output of one university only. The exception is Orzanco et al.  who include findings from two Canadian universities.
As previously noted, positive findings about graduates working rurally may be confounded by students’ rural origin or choice to undertake rural programmes, attachments or years at a rural clinical school. Glasser et al. , for example, note that their recruiters visit rural secondary schools to identify those interested in rural medicine. Anomalously, Strasser et al.  conclude that rural/urban background had a significant interaction with all their main outcome measures except place of current practice.
Typology category 4: studies of impact of various durations of rural medical education and rural postgraduate internship
We found nine studies [27,62-69] where the impact of various durations of rural medical education on rural employment immediately following graduation (i.e. for postgraduate internship) was considered. These are outlined in Table 7:
In two studies, small numbers of annual graduates from the University of Queensland were considered [63,64]. Those undertaking a rural placement, compared with no rural placement, or a long rural placement compared to a short rural placement, were more likely to undertake a rural internship [27,62,65,69]. Some studies identified whether students were from a rural place or intended to work rurally and noted that this had a positive impact on eventual rural internship . Clark et al.  found that extended rural placement had a stronger association with rural internship than coming from a rural place. A high rate of rural internship outcome was shown for James Cook University (rural Queensland, Australia), with 67% of graduates undertaking their internship outside a metropolitan centre. This compares with 17% from other Australian universities . Diverse ways of defining rural were again evident; perhaps the most challenging being Playford and Cheong’s  designation of rural internship as having at least 4 weeks in 1 year in a rural place.
Typology category 5: studies that identify rural practitioners and investigate an association with rural medical education
The final category in our typology is presented in Table 8.
We found four studies in this category [70-73]. Rourke et al.  and Pathman et al.  examined associations between working in rural areas and amount of time spent in rural rotations as students, showing some associations. They did not consider gradient effects. Rolfe et al.  found a positive association between rural working and having a rural general practice placement attachment. Findings of Rosenblatt et al.  indicated that a small number of USA medical schools have much higher production of rural doctors, but they did not explore for rural location of the medical schools or duration at a rural location during medical education.
The purpose of this review was to scope the state of evidence in relation to two research questions: Does medical education in a rural place produce rural doctors? And, is there a gradient of time spent in rural-based medical education that contributes to the increased likelihood of a graduate working in rural medical practice? We found few studies that specifically sought to identify the benefits of having medical education wholly in a rural place, or of gradient effects of increasing time spent in rural medical education, on producing rural doctors. Studies found were limited by considering outputs of single medical schools, lack of comparative data and/or being conducted by researchers from the medical school being studied. We found no studies considering the cost-effectiveness of different models of producing doctors for rural places. While most of the papers reviewed showed evidence that rural exposure during medical education increased the likelihood of practising rurally following graduation, insufficient detail, differing definitions, small numbers, inclusion of outputs from only one university, confusing reporting and lack of comparative data make drawing clear conclusions about both of our research questions unwise and impractical. The best we can say is indications are positive, but better quality evidence is needed. Our reporting is crude because we endeavoured to unite findings to obtain a sense of the overall state of the evidence.
Category 1 of our typology (two studies found) compared rural campuses with the main campus of the same university and appeared to show that rural campuses produce higher percentages of rural workers. Typology category 2 examined studies of the impact of medical schools providing education wholly in rural places. From the studies, we conclude that percentages working rurally appear high when compared with available national data. Typology category 3 would ideally have provided evidence about gradient effects as it unites 17 studies each defining lengths of time medical students spent in rural places during their undergraduate degrees. Three studies considered different lengths of time spent in rural education, concluding an association between longer time in a rural clinical school or GP attachment and greater likelihood of rural working. Most included studies showed associations between time spent in rural programmes, clinical schools or attachments and rural work outcomes. Typology category 4 considered first years of employment in internships. Again, evidence indicates a relationship between experiences of medical education in a rural place and rural working. The most striking finding is from Sen Gupta et al.  of around two thirds of students proceeding to a rural internship following wholly rural education. Typology category 5 considered studies involving rural doctors and identifying factors associated with rural working. Again, there were indications, in most included studies, of links between time spent in rural medical education and rural work outcomes.
Overall, we found some studies confusingly written, particularly around identifying numbers of students or practitioners included in analyses relating to our research question. Considering the impact of place was often not the papers’ focus but was explored alongside other variables and relationships. Some studies had very small sample sizes. Some lacked transparency, for example, by not reporting original sample size or methods. Several studies had to be excluded from our review, as they did not quantify precise periods of rural exposure. Some studies combined (and in some cases conflated) intentions to work rurally with actual rural work. With studies considering programmes, GP attachments or time spent at rural clinical schools, it might be argued that what is being measured is not simply the effect of rural education location per se but, rather, the ‘package’ of experience provided, including ‘hard to measure’ factors like the enthusiasm of educators and mentors for rural working. None of the associations linking place of education with place of employment are therefore independent of other factors. Invariably, the studies of medical school students’ conversion to rural practice, following the range of rural education durations and types, were conducted by teams involving researchers from the medical schools concerned.
There was considerable variety in the ways rural was discussed and defined. This is a regular challenge when comparing studies in rural health. We included studies where authors specifically stated they were considering rural working as an outcome; that is, they specifically used the term rural (remote and regional were also included), and they specifically defined or described what was meant. For example, Inoue et al.  note there is no definition of rural in Japan and state: ‘In this study, rural areas were defined as areas that are subject to certain laws in local municipalities: the Special Promotion Act for Sparsely Populated areas, the Mountain Village Promotion Act, or the Special Act for Areas with Heavy Winter Snow’. Australian studies tended to use established categorisations, for example, the Rural, Remote and Metropolitan Areas Classification (RRAMA) . USA studies used various categorisations, including ‘rural–urban commuting area (RUCA) codes’ and Rural Urban Density Typology, categorising variously towns and counties on a continuum of rural to urban. Other studies applied population size of towns/cities or population size plus distance from a large centre as proxy for rural; for example, Rourke  defined rural communities as having populations of 1 000–30 000. Forster et al.  used participant-defined location—and was the only study reported here to do so.
Given the caveats outlined, there is evidence (lacking in robustness though it is) that does suggest that medical education in a rural place may positively affect production of rural doctors. Additionally, there is some evidence to suggest a gradient effect, with increasing duration of educational experiences in rural settings, although the number of studies is small and limited to evaluating single programmes.
A key challenge to having these findings acknowledged as interesting and worthy of further exploration is the regular reporting that the findings are likely confounded by students’ having a rural background or pre-existing propensity to choose rural work. This is often suggested to influence their initial choice of a rural-based medical programme, special programme or attendance at a rural clinical school.
While rural background is influential, its inclusion as a problematical confounding factor, detrimental to considering the value of rural medical education, raises questions. The success of rural medical schools such as Jichi, Tromso and James Cook University in producing high proportions of rural doctors prompts consideration of a counterfactual. Put simply, what happened to these rural-origin students before the rural programmes existed? Did the students attend city universities and become converted to urban work? Or, did they not attend medical school at all? Are rural medical programmes, therefore, simply efficiently finding those with a propensity to rural work and converting them into rural doctors? It is an intriguing, but inherently un-researchable, question to compare the situation now with the past.
Limitations and recommendations
Our interest in conducting this review was focused on exploring an association between rural medical education and rural working. We were specifically interested in rural practice following medical school, so we did not include the impact of postgraduate study. A review that considers studies reporting work location following the completion of all postgraduate education would have a different purpose to ours.
Confounding factors feature in discussions considering the relationship between rural medical education and rural practice location—most notably the effect of rural background. Due to varying reporting in the studies we reviewed, we made the early decision not to divide studies into those that stated they had controlled for rural background and those that did not state this. Put simply, we did not have the consistent information for every study about whether or not they had controlled for rural background, so it would be misleading of us to attempt to divide. Of the few studies that stated they had accounted for rural background, the strongest predictors for practising rurally, following graduation, were location of the medical school and time spent gaining experience in rural practices during medical education. There is a gap in robust studies which allow for distinction of the effects of different variables on rural practice outcomes.
We believe that the reporting typology that we developed provides a useful, if pragmatic, structure for reporting about a relatively large group of articles. Providing an overview of studies from different countries, and typologising them, given variable coverage in reporting is inevitably challenging. Consistent with scoping review methods, the purpose was to map the nature and extent of existing international published evidence.
The WHO  recommends medical education in a rural place as one of a basket of means to increase rural doctor supply, while—to date—suggesting that the evidence base can be merely indicative (due to study paucity and methods). Within the evidence gap, traditional city-based medical schools continue to provide small-scale self-evaluations of their rural-focused programmes. In the studies that we reviewed, there was a lack of description of independent, or external, scrutiny in evaluation and auditing of data. There is a need for studies incorporating these strategies.
For countries that lack access to well-constructed longitudinal medical workforce databases, there is a dearth of quality comparative statistics. International collaborations that collect comparative data would strengthen the evidence base. Findings from independent, international, robust, studies would be valuable because rural advocacy groups continue to state rural doctor deficits, while city-based medical schools continue to provide small-scale evidence that their programmes are working.
Given the considerable public finance invested in schemes to generate rural doctors, including bonded student places, clinical schools, workforce agencies and work incentives , independent studies that consider the relative cost-effectiveness of rural-based medical education versus other strategies would be useful for policymakers.
Studies to address the question—does medical education in a rural area most effectively produce rural doctors?—might adopt modelling or prospective or post hoc approaches. Ideally, a study would include the following: rural and urban medical schools; apply a practical, but meaningful, classification of rural; include data on students’ rural/urban background on entering medical school (to establish the effects of rural origin vis a vis rural medical education); apply a measure of rural work outcome (e.g. place of work after 5 years or durations of work in rural/urban places); and identify per capita costs of producing a medical graduate in a rural or urban area.
A study to establish whether there is a gradient in rural work outcomes in relation to time spent in rural places during medical education would ideally include the following: several rural- and urban-based medical schools; use data on students’ rural/urban background on entering medical school; identify number of weeks spent located in a rural place, as part of entry-level medical education for each student; apply a measure of rural work outcome; and identify costs associated with different numbers of weeks in rural places.
While such studies might be dismissed as unfeasibly complex or large and thus expensive, the ongoing cost, internationally, of medical workforce incentives and payments to universities to produce rural practitioners is considerable. The question is perhaps whether enough of the powerful stakeholders want to know the answers.
Technology should make education of all types more readily available in future, regardless of location, but there will always be a strong element of medical education that involves hands-on work with patients and learning from skilled role models. The location where these experiences take place may be important, and it seems only just that rural populations, national governments and international agencies should have the best evidence so they can fund the best methods of producing rural doctors. Some of the significant international spending applied to getting doctors to work rurally could be diverted to finding out what works. In the short term, an outlay of resources to identify what works best could lead to considerable long-term gains in healthcare accessibility and ultimately to a cost-effective contribution to improving rural health outcomes.
Jones M, Humphreys J, Prideaux D. Predicting medical students’ intentions to take up rural practice after graduation. Med Educ. 2009;43:1001–9. doi:1010.1111/j.1365-2923.2009.03506.x.
Royston PJ, Mathieson K, Leafman J, Ojan SO. Medical student characteristics predictive of intent for rural practice. Rural Remote Health. 2012;12:2107.
Walker JH, DeWitt DE, Pallant JF, Cunningham CE. Rural origin plus a rural clinical school placement is a significant predictor of medical students’ intentions to practice rurally: a multi-university study. Rural Remote Health. 2012;12:1908.
Wakerman J, Humphreys J. Rural and remote health: definitions, policy and priorities. In: Liaw S, Kilpatrick S, editors. Rural and remote health: definitions, policy and priorities. Canberra: Australian Rural Health Education Network; 2008. p. 13–30.
Buchan J, Couper ID, Tangcharoensathien V, Thepannya K, Jaskiewicz W, Perfilieva G, et al. Early implementation of WHO recommendations for the retention of health workers in remote and rural areas. Bull World Health Organ. 2013;91:834–40.
World Health Organisation. Increasing access to health workers in remote and rural areas through improved retention. Geneva: WHO; 2010.
Bigbee J. Relationships between nurse- and physician-to-population ratios and state health rankings. Public Health Nurs. 2008;25:244–52.
Sarma S, Peddigrew C. The relationship between family physician, density and health related outcomes: the Canadian evidence. Cah Sociol Demogr Med. 2008;48:61–105.
Smith KB, Humphreys JS, Wilson MG. Addressing the health disadvantage of rural populations: how does epidemiological evidence inform rural health policies and research? Aust J Rural Health. 2008;16:56–66.
Duckett S, Breadon P, Ginnivan L. Access all areas: new solutions for GP shortages in rural Australia. Melbourne: Grattan Institute; 2013.
Goldman L, Schafger A. Approach to medicine, the patient, and the medical profession. In: Goldman L, Schafger A, editors. Goldman’s Cecil medicine. 24th ed. Philadelphia: Saunders Elsevier; 2011.
United Nations. The Universal Declaration of Human Rights. New York: United Nations; 1948.
Pitblado R. Geographical distribution of rural health human resources. Vancouver: UBC Press; 2012.
Scott A, Witt J, Humphreys J, Joyce C, Kalb G, Jeon S, et al. Getting doctors into the bush: general practitioner’s preferences for rural location. In: Working paper series no 13/12. Melbourne: Melbourne University Institute; 2012.
Matsumoto M, Inoue K, Kashima S, Takeuchi K. Does the insufficient supply of physicians worsen their urban–rural distribution? A Hiroshima-Nagasaki comparison. Rural Remote Health. 2012;12:2085.
Duckett S, Wilcox S. The Australian healthcare system. 4th ed. Melbourne: Oxford University Press; 2011.
The Royal New Zealand College of General Practitioners. What is rural hospital medicine. 2015. [http://www.rnzcgp.org.nz/what-is-rural-hospital-medicine/]. Accessed 2 May 2015.
Dewar J. Report of the Highlands and Islands Medical Service Committee. 1912. [http://www.ambaile.org.uk/gd/download/files/Dewar%20Report.pdf]. Accessed 2 May 2015.
Quinn KJ, Kane KY, Stevermer JJ, Webb WD, Porter JL, Williamson Jr HA, et al. Influencing residency choice and practice location through a longitudinal rural pipeline program. Acad Med. 2011;86:1397–406.
United States Census 2010. [http://www.census.gov/2010census/] Accessed 2 May 2015.
Mason J. Review of Australian government health workforce programs. Canberra: Australian Government; 2013.
Strasser R, Neusy AJ. Context counts: training health workers in and for rural and remote areas. Bull World Health Organ. 2010;88:777–82.
Ranmuthugala G, Humphreys J, Solarsh B, Walters L, Worley P, Wakerman J, et al. Where is the evidence that rural exposure increases uptake of rural medical practice? Aust J Rural Health. 2007;15:285–8.
Australian Standard Geographical Classification (ASGC) [http://www.abs.gov.au/websitedbs/d3310114.nsf/home/australian+standard+geographical+classification+(asgc)]
Strasser RP, Lanphear JH, McCready WG, Topps MH, Hunt DD, Matte MC. Canada’s new medical school: the Northern Ontario School of Medicine: social accountability through distributed community engaged learning. Acad Med. 2009;84:1459–64.
Eley DS, Synnott R, Baker PG, Chater AB. A decade of Australian Rural Clinical School graduates–where are they and why? Rural Remote Health. 2012;12:1937.
Playford DE, Cheong E. Rural undergraduate support and coordination, rural clinical school, and rural Australian medical undergraduate scholarship: rural undergraduate initiatives and subsequent rural medical workforce. Aust Health Rev. 2012;36:301–7.
Rabinowitz HK, Diamond JJ, Markham FW, Santana AJ. The relationship between matriculating medical students’ planned specialties and eventual rural practice outcomes. Acad Med. 2012;87:1086–90.
Farmer J, Munoz SA, Daly C. Being rural in rural health research. Health Place. 2012;18:1206–8.
Bourdieu P. In other words: essays towards a reflexive sociology. California: Stanford University Press; 1990.
Wilkinson D, Laven G, Pratt N, Beilby J. Impact of undergraduate and postgraduate rural training, and medical school entry criteria on rural practice among Australian general practitioners: national study of 2414 doctors. Med Educ. 2003;37:809–14.
Sen Gupta T, Woolley T, Murray R, Hays R, McCloskey T. Positive impacts on rural and regional workforce from the first seven cohorts of James Cook University medical graduates. Rural Remote Health. 2014;14:2657.
Trounson A. University of Sydney proposes rural shift in medical training places. In The Australian Higher Education, The Australian. Sydney: News Corp Australia; 2013.
Whitcomb ME. New and developing medical schools: motivating factors, major challenges, planning strategies. Washington: George Washington University; 2009.
Higher education [http://australia.gov.au/topics/education-and-training/higher-education] Accessed 2 May 2015.
Marginson S. University rankings, government and social order: managing the field of higher education according to the logic of the performative present-as-future. In: Simons M, Olssen M, Peters M, editors. Re-reading education policies: studying the policy agenda of the 21st century. Rotterdam: Sense Publishers; 2007.
Kharas H. The emerging middle class in developing countries. Paris: OECD; 2010.
Kenny A, Hyett N, Sawtell J, Dickson-Swift V, Farmer J, O’Meara P. Community participation in rural health: a scoping review. BMC Health Serv Res. 2013;13:20130218.
Arksey H, O’Malley L. Scoping studies: towards a methodological framework. Int J Soc Res Methodol. 2005;8:19–32.
Giustini D, Boulos M. Google Scholar is not enough to be used alone for systematic reviews. Online J Public Health Inform. 2013;5:214.
Crump WJ, Fricker RS, Ziegler C, Wiegman DL, Rowland ML. Rural track training based at a small regional campus: equivalency of training, residency choice, and practice location of graduates. Acad Med. 2013;88:1122–8.
Brokaw JJ, Mandzuk CA, Wade ME, Deal DW, Johnson MT, White GW, et al. The influence of regional basic science campuses on medical students’ choice of specialty and practice location: a historical cohort study. BMC Med Educ. 2009;9:29.
Inoue K, Hirayama Y, Igarashi M. A medical school for rural areas. Med Educ. 1997;31:430–4.
Magnus JH, Tollan A. Rural doctor recruitment: does medical education in rural districts recruit doctors to rural areas? Med Educ. 1993;27:250–3.
Stratton TD, Geller JM, Ludtke RL, Fickenscher KM. Effects of an expanded medical curriculum on the number of graduates practicing in a rural state. Acad Med. 1991;66:101–5.
Orzanco MG, Lovato C, Bates J, Slade S, Grand’Maison P, Vanasse A. Nature and nurture in the family physician’s choice of practice location. Rural Remote Health. 2011;11:1849.
Strasser R, Hogenbirk JC, Lewenberg M, Story M, Kevat A. Starting rural, staying rural: how can we strengthen the pathway from rural upbringing to rural practice? Aust J Rural Health. 2010;18:242–8.
Williamson MI, Wilson R, McKechnie R, Ross J. Does the positive influence of an undergraduate rural placement persist into postgraduate years? Rural Remote Health. 2012;12:2011.
Kane KY, Quinn KJ, Stevermer JJ, Porter JL, Webb WD, Williamson Jr HA, et al. Summer in the country: changes in medical students’ perceptions following an innovative rural community experience. Acad Med. 2013;88:1157–63.
Lang F, Ferguson KP, Bennard B, Zahorik P, Sliger C. The Appalachian Preceptorship: over two decades of an integrated clinical-classroom experience of rural medicine and Appalachian culture. Acad Med. 2005;80:717–23.
Glasser M, Hunsaker M, Sweet K, MacDowell M, Meurer M. A comprehensive medical education program response to rural primary care needs. Acad Med. 2008;83:952–61.
MacDowell M, Glasser M, Hunsaker M. A decade of rural physician workforce outcomes for the Rockford Rural Medical Education (RMED) Program, University of Illinois. Acad Med. 2013;88:1941–7.
Halaas GW. The Rural Physician Associate Program: successful outcomes in primary care and rural practice. Rural Remote Health. 2005;5:453.
Halaas GW, Zink T, Finstad D, Bolin K, Center B. Recruitment and retention of rural physicians: outcomes from the rural physician associate program of Minnesota. J Rural Health. 2008;24:345–52.
Rabinowitz HK. Recruitment, retention, and follow-up of graduates of a program to increase the number of family physicians in rural and underserved areas. N Engl J Med. 1993;328:934–9.
Rabinowitz HK, Diamond JJ, Markham FW, Hazelwood CE. A program to increase the number of family physicians in rural and underserved areas: impact after 22 years. JAMA. 1999;281:255–60.
Rabinowitz HK, Diamond JJ, Markham FW, Rabinowitz C. Long-term retention of graduates from a program to increase the supply of rural family physicians. Acad Med. 2005;80:728–32.
Smucny J, Beatty P, Grant W, Dennison T, Wolff LT. An evaluation of the Rural Medical Education Program of the State University of New York Upstate Medical University, 1990–2003. Acad Med. 2005;80:733–8.
Jamar E, Newbury J, Mills D. Early career location of University of Adelaide rural cohort medical students. Rural Remote Health. 2014;14:2592.
Playford D, Evans S, Atkinson D, Auret K, Riley G. Impact of the Rural Clinical School of Western Australia on work location of medical graduates. Med J Aust. 2014;200:204–7.
Forster L, Assareh H, Watts LD, McLachlan CS. Additional years of Australian rural clinical school undergraduate training is associated with rural practice. BMC Med Educ. 2013;13:37.
Clark TR, Freedman SB, Croft AJ, Dalton HE, Luscombe GM, Brown AM, et al. Medical graduates becoming rural doctors: rural background versus extended rural placement. Med J Aust. 2013;199:779–82.
Eley D, Baker P. Does recruitment lead to retention? Rural clinical school training experiences and subsequent intern choices. Rural Remote Health. 2006;6:511.
Eley DS, Baker PG. Will Australian rural clinical schools be an effective workforce strategy? Early indications of their positive effect on intern choice and rural career interest. Med J Aust. 2007;187:166–7.
Eley DS, Zhang J, Wilkinson D. Self-sufficiency in intern supply: the impact of expanded medical schools, medical places and rural clinical schools in Queensland. Aust Health Rev. 2009;33:472–7.
Sen Gupta T, Murray R, Hays R, Woolley T. James Cook University MBBS graduate intentions and intern destinations: a comparative study with other Queensland and Australian medical schools. Rural Remote Health. 2013;13:2313.
McDonnel Smedts A, Lowe MP. Clinical training in the top end: impact of the Northern Territory Clinical School, Australia, on the Territory’s health workforce. Rural Remote Health. 2007;7:723.
McDonnel Smedts A, Lowe MP. Efficiency of clinical training at the Northern Territory Clinical School: placement length and rate of return for internship. Med J Aust. 2008;189:166–8.
Eley D, Baker P, Chater B. The Rural Clinical School Tracking Project: more IS better–confirming factors that influence early career entry into the rural medical workforce. Med Teacher. 2009;31:e454–9.
Rourke JTB, Incitti F, Rourke LL, Kennard M. Relationship between practice location of Ontario family physicians and their rural background or amount of rural medical education experience. Can J Rural Med. 2005;10:231–40.
Pathman DE, Steiner BD, Jones BD, Konrad TR. Preparing and retaining rural physicians through medical education. Acad Med. 1999;74:810–20.
Rosenblatt RA, Whitcomb ME, Cullen TJ, Lishner DM, Hart LG. Which medical schools produce rural physicians? JAMA. 1992;268:1559–65.
Rolfe IE, Pearson SA, O’Connell DL, Dickinson JA. Finding solutions to the rural doctor shortage: the roles of selection versus undergraduate medical education at Newcastle. Aust New Zealand J Med. 1995;25:512–7.
Department of Primary Industries & Energy, Department of Human Services & Health. Rural, remote and metropolitan areas classification. Canberra: Australian Institute of health and Welfare; 1994.
Rabinowitz HK, Paynter NP. The role of the medical school in rural graduate medical education: pipeline or control valve? J Rural Health. 2000;16:249–53.
No funding was obtained for this study and all contributors are authors.
The authors declare that they have no competing interests.
JF, AK, CM, RH—conception and design, acquisition of data and interpretation, drafting of manuscript, intellectual contribution to revisions, approval of final manuscript, accountable for all aspects of the work.