Aust. J. Rural Health (2002) 10, 94–98

Blackwell Science, LtdOxford, UKAJRThe Australian Journal of Rural Health1038-52822002 Blackwell Science Asia Pty Ltd10451Australian Journal of Rural HealthComparison of Metropolitan and Rural Medical Schools in China: L. WANG10.1046/j.1038-5282.2002.00451.xOriginal Article BEES SGMLOriginal Article

A COMPARISON OF METROPOLITAN AND RURAL MEDICAL SCHOOLS IN CHINA: WHICH SCHOOLS PROVIDE RURAL PHYSICIANS?

Lexin Wang

School of Biomedical Sciences, Charles Sturt University, Wagga Wagga, New South Wales, Australia

ABSTRACT: To compare the role of metropolitan and rural medical schools in the provision of rural physicians,a survey was conducted in 12 metropolitan and 10 rural medical schools. Rural medical schools enrolled fewerstudents ( P = 0.019), and produced fewer graduates ( P = 0.023) than metropolitan medical schools. Students inrural medical schools were mainly from surrounding regional cities and counties, whereas those in metropolitanschools were from cities nationwide ( P < 0.001). All rural medical schools produced rural physicians; one ruralschool reported that of its 256 graduates, 88 (34.4%) entered rural practice. Ten of the 12 metropolitan medicalschools did not produce any rural physicians, whereas the remaining two metropolitan schools registered a total of 73graduates who selected a rural practice location. These results indicate that rural medical schools may play a keyrole in overcoming the shortage of physicians in rural communities in China.

KEY WORDS: medical education, rural medical schools, rural physicians.

INTRODUCTION

A shortage of physicians in rural and remote area hasbeen a long-standing and serious problem affecting manycountries.1,2 Attitudinal and perceptual barriers arerecognised factors that discouraging medical graduatesfrom entering rural practice.1 Work stress, professionalisolation and family issues often make the retention of ruralphysicians difficult, if not impossible.3

Ways to boost the number of rural physicians havebeen vigorously sought by healthcare policy makers andeducators. Financial incentives, such as, grants for relocation,training and locum support have been used with variablesuccess.4 Programs to increase students’ exposure to ruralhospital and medical services have been developed andpromoted because students with a rural background or

rural practice experience are more likely to pursue ruralpractice as a long-term career.5,6

Furthermore, studying medicine in a rural location hasbeen shown to increase the retention of rural physicians.7,8

Mainland China is known to have established a numberof rural medical schools and rural physician educationprograms in an effort to provide its 900 million ruralpopulation with quality healthcare.9 The primary aim ofthis study was to characterise some of these rural medicalschools and to compare them with metropolitan medicalschools in the provision of rural physicians.

METHOD

Definition of rural medical schoolsA complete list of medical schools in mainland China wasobtained from the Ministry of Education in April 2001.These schools, which spread across 31 provinces and auton-omous territories, were offering a 5-year undergraduatecourse of conventional (Western) medicine. Some of theseschools were also offering a 7-year course, in which students

Correspondence: Dr Lexin Wang, School of BiomedicalSciences, Charles Sturt University, Wagga Wagga, NSW 2678,Australia. Email: lwang@csu.edu.au

Accepted for publication November 2001.

COMPARISON OF METROPOLITAN AND RURAL MEDICAL SCHOOLS IN CHINA: L. WANG 95

completed a research project in the final year of study andgraduated with a bachelor’s and master’s degree of medicine.

Four military medical schools were excluded from thisstudy, because their graduates had military obligationsfor many years after graduation. We divided the remainingmedical schools into three groups, according to theiraffiliation and the geographical location of the main cam-puses. The first group of 14 medical schools was affiliatedto the Ministry of Health or Education, and located inmajor metropolitan cities. The second group of 28 schoolswas affiliated to the provincial government and located inthe capital cities of provinces and autonomous territories.

The third group of 40 schools was also affiliated tothe provincial government or other government agencies,however, this group was located in regional cities outsideof provincial capitals with a population ranging from300 000 to 5 million. Rural and regional counties oftensurrounded these medical schools. The third group ofschools had affiliated teaching hospitals with fewer in-hospital beds than the first two group’s, and oftenappointed hospitals in smaller regional cities as theirteaching hospitals.

In contrast to the first two groups of schools, the thirdgroup of medical schools only offered a 5-year undergrad-uate program, although some schools also had a 6-yearcourse (5-year medicine and 1-year English) at the time ofthe survey.

For the purpose of this study, we defined the thirdgroup of schools as rural medical schools and the first twogroups as metropolitan schools. We also defined county ortownship hospitals, which were often surrounded by ruralpopulations, as rural hospitals. Physicians practicing inthese hospitals were defined as rural physicians.

Data collectionWe randomly selected 12 metropolitan schools, six fromthose affiliated to the Ministry of Health and six from thoseaffiliated to provincial governments. We also randomly

selected 12 rural medical schools. Between 1 and 20 April2001, we carried out an Internet search on the officialInternet site of the selected schools, to obtain informationabout undergraduate courses, student enrolment and gradu-ate employment in 2000. We only collected student andcourse information related to clinical medicine, which wasthe mainstream course of all medical schools. Informationon other courses, such as, medical imaging, family planning,public health and preventive medicine was not registeredbecause the number of students in these courses wassmall, and the offering of these courses was highlyvariable among schools.

Some schools only enrolled students from their ownprovinces or autonomous territories. To facilitate com-parison of student origin between all schools, we definedstudents from other provinces or autonomous territories asforeign students.

A questionnaire was sent to schools without an acces-sible official Internet site, or schools with incompleteinformation on their Internet site.

StatisticsData variables were expressed as mean ± SD. Variablesbetween schools were compared by an ANOVA test.P < 0.05 was considered to be statistically significant.

RESULTS

The Internet search and questionnaire survey were com-pleted in all but two rural medical schools, whose studentinformation was incomplete and excluded from final analysis.

Origin of studentsStudent enrolment at all medical schools was regulated bya quota system developed by the Ministry of Education orprovincial government. Schools were given specific guide-lines on the number of enrolments and geographical areasfrom which a school could enrol students.

TABLE 1: Characteristics of three types of medical schools

Features M1 (n = 6) M2 (n = 6) R (n = 10)

No. of enrolments 409 ± 117 430 ± 101 301 ± 97*

No. of graduates 413 ± 46 480 ± 71 258 ± 86**

No. of schools with 7-year program 6 4 0

No. of schools with designated rural students 0 2 10

M1, metropolitan medical schools affiliated to the Ministry of Health; M2, metropolitan medical schools affiliated to the provincial

government; R, rural medical schools. *P = 0.019 versus M1 and M2; **P = 0.023 versus M1 and M2.

96 AUSTRALIAN JOURNAL OF RURAL HEALTH

Metropolitan schools enrolled more students thanrural schools (Table 1). They also enrolled more foreignstudents (from an average of 11 ± 3 provinces) than ruralschools (Fig. 1). Medical schools affiliated to the Ministryof Health had the highest percentage of foreign students(Fig. 1).

Only three rural medical schools enrolled foreignstudents. Two of these schools, who each had 19% foreignstudents, were affiliated to the Ministry of Coal Industryand the Ministry of Transport. The third rural school enrolledthree (1.6%) of its 168 students from a neighbouringprovince. The remaining seven rural schools enrolledstudents mostly from the surrounding rural cities orcounties.

All rural schools had designated places, ranging from9.1% to 15.1% of annual intake, for students from ruralareas (Table 1). Two of the 12 metropolitan schools alsohad such openings, which accounted for 1.0% and 6.7%of the annual enrolment of each of these schools. Designatedrural students, whose entry standard was slightly lowerthan the remainder of the class, were required to completethe same training process as other students, and contractedto practice at the place of origin after graduation.

Duration of undergraduate courseTen metropolitan schools had both 5- and 7-year under-graduate medical courses (Table 1). The number of studentsenrolled in the 7-year course ranged from 15 to 40. Noneof the rural schools offered a 7-year course (Table 1).

Employment policies for graduatesUnder the general policy framework set by the Ministry ofEducation, medical graduates were encouraged to practicein their place of origin. Although they were free to practiceanywhere across the nation, major capital cities restrictforeign graduates from seeking permanent employment inmetropolitan areas.

In the nation’s capital city, for example, only 10% ofall foreign graduates, mostly those with higher academicachievements, were invited to work permanently in themetropolitan area in 2000. A similar policy on foreigngraduates was found in other provincial capitals.

Two metropolitan and one rural schools in a southernprovince offered financial incentives to graduates whowished to practice in any of the 16 rural counties of thatprovince. The amount of the one-off financial grant wasdependent on a graduate’s qualification, being approximatelyone-half of a year’s salary for a graduate with a bachelor’sdegree working in a public hospital.

Graduates entering rural practiceTwo metropolitan schools with provincial governmentaffiliation indicated that 4.9% (21/430) and 9.8% (52/530) of their graduates secured employment in ruralhospitals. The remaining metropolitan schools registeredno graduates entering rural practice.

Rural schools produced fewer graduates than metro-politan schools (Table 1). All these schools reported thatsome graduates were being employed by rural hospitals.One school in an eastern province provided employmentdetails for its 256 graduates (Fig. 2).

FIGURE 1: Comparison of foreign students enrolled in the

three groups of medical schools in 2000. M1, metropolitan

medical schools affiliated to the Ministry of Health; M2,

metropolitan medical schools affiliated to the provincial

government; R, rural medical schools.

FIGURE 2: Practice location of year 2000 graduates of a rural

medical school. Military, military medical service; Postgrad,

postgraduate study; Unknown, with unknown practice location.

COMPARISON OF METROPOLITAN AND RURAL MEDICAL SCHOOLS IN CHINA: L. WANG 97

DISCUSSION

The primary finding of the study was that rural medicalschools ran a smaller undergraduate course of clinicalmedicine, and produced fewer graduates than metropolitanschools in 2000. Rural medical schools, however, aremore likely to offer designated places for rural studentsand produce rural physicians than metropolitan schools.

Location of medical schools and rural physiciansThe location of rural medical schools seems crucial to astudent’s decision to become a rural physician. Previousobservations have shown that a student’s rural backgroundor experience is associated with the likelihood of choosingrural practice as a long-term career.6,10 Physicians oftenpractice near where they are trained.11,12 Furthermore,the location of a medical school in a rural area appears tohave a strong association with the number of graduatesentering rural practice.7,13,14 The rural medical schools inour study were surrounded by rural or remote counties,from which most students were enrolled. A higher per-centage of graduates from these schools choose to practicein the surrounding rural areas, because that is whereemployment opportunities lie.

Mission of medical schools and rural physiciansThe mission and structure of a medical school also influ-ence a student’s decision in selecting practice locationafter graduation. The goals, course structure and inherentvalues of a medical school can shape the student’s behav-iour, and have the potential to increase the supply of ruralphysicians.15 Medical schools with a strong biomedicalresearch environment produce fewer rural physicians,whereas those with strong family physician programsproduce more rural physicians.13 Although the researchintensity of schools was not evaluated in this study, it iscommon knowledge that metropolitan medical schools inChina have a better research infrastructure, receive moreresearch funds from government agencies, and producemore postgraduate students than rural schools. The factthat most metropolitan schools have a 7-year course, whilenone of the rural schools have such a program, also indicatesthe stronger research focus of metropolitan schools.

Role of governmentGovernment seems to play a key role in the provisionof rural physicians. It is the national and local governmentswho determine the number and location of rural medicalschools, which account for almost half of total conventional

medical schools in mainland China. Government canalso help to develop educational priorities throughfunding distribution to medical schools for differentfunctions.13 Furthermore, nationwide restrictions onpopulation (including physicians) migration from rural tourban areas, a policy that is perhaps unique to China,has largely contributed to the steady flow of medicalgraduates into remote areas where qualified physiciansare in high demand.

Limitations of the studyAlthough all rural medical schools produced rural physi-cians, only one school provided details about graduateemployment. The number of rural physicians producedfrom other rural schools is unclear. However, given thesimilarities between the 10 rural medical schools ingeographical location, course structure, student origin andgraduate employment policy, one would not expect toomuch deviation in the number of rural physicians fromthese schools. Although we cannot compare the number ofrural physicians produced by metropolitan schools withthe number produced by rural schools, we are almostcertain that rural schools have produced more ruralphysicians because the 12 metropolitan schools registeredonly 73 graduates entering rural practice, which is lessthan that from one rural school (Fig. 2).

In conclusion, a combination of rural location, students’rural background or origin, and a government-regulatedemployment policy has made rural medical schools betterproviders of rural physicians. Maximising the capacity ofexisting rural medical schools, or establishing more med-ical schools in areas where rural and remote populationslive, can be an effective way of overcoming the shortage ofrural physicians.

ACKNOWLEDGEMENTS

The author whishes to thank Dr Juan Chen at Wei FangMedical College, People’s Republic of China, for heradvice and information on graduate employment. Theauthor also thanks Professors Mark Burton and DavidBattersby at Charles Sturt University for the criticalreview of the manuscript.

REFERENCES

1 WONCA. Working Party on Training for Rural Practice.Policy on training for rural practice. Canadian FamilyPhysician 1996; 42: 1181–1183.

2 Hays RB. Common international themes in rural medicine.Australian Journal of Rural Health 1999; 3: 191–195.

98 AUSTRALIAN JOURNAL OF RURAL HEALTH

3 Hays RB, Veitch PC, Cheers B, Crossland L. Why doctorsleave rural practice. Australian Journal of Rural Health1997; 5: 198–203.

4 Hutten-Czapski P. Rural incentive programs: a failingreport card. Canadian Journal of Rural Medicine 1998; 3:242–247.

5 Haug MR, Lavin B, Breslay N. Practice location preferencesat entry to medical school. Journal of Medical Education1980; 55: 333–338.

6 Rolfe IE, Pearson SA, O’Connell DL, Dickinson JA. Findingsolutions to the rural doctor shortage: the roles of selectionversus undergraduate medical education at Newcastle.Australian and New Zealand Journal of Medicine 1995; 25:512–517.

7 Magnus JH, Tollan A. Rural doctor recruitment: does med-ical education in rural districts recruit doctors to ruralareas? Journal of Medical Education 1993; 27: 250–253.

8 Rabinowitz HK, Diamond JJ, Markham FW, Hazelwood CE.A program to increase the number of family physicians inrural and underserved areas: Impact after 22 years. Journalof American Medical Association 1999; 281: 255–260.

9 Umland B, Waterman R, Wiese W, Duban S, Mennin S,Kaufman A. Learning from a rural physician program inChina. Academic Medicine 1992; 67: 307–309.

10 Connor RA, Hillson SD, Kralewski JE. Associationbetween rural hospitals’ residencies and recruitment andretention of physicians. Academic Medicine 1994; 69:484–488.

11 Dorner FH, Burr RM, Tucker SL. The geographic relationshipsbetween physicians’ residency sites and the locations of theirfirst practices. Academic Medicine 1991; 66: 540–544.

12 Burfield WB. Hough DE, Marder WD. Location of medicaleducation and choice of location of practice. Journal ofMedical Education 1986; 61: 545–554.

13 Rosenblatt RA, Whitcomb ME, Cullen TJ, Lishner DM,Hart LG. Which medical schools produce rural physicians?Journal of American Medical Association 1992; 268: 1559–1565.

14 Inoue K, Hirayama Y, Igarashi M. A medical school for ruralareas. Journal of Medical Education 1997; 31: 430–434.

15 Bruce TA. Physicians for the American homelands. Aca-demic Medicine 1990; 65 (Suppl.): S10–S14.