The impact of rural-exposure strategies on the intention of dental students and dental graduates to practice in rural areas: a systematic review and meta-analysis

Figures & data

Table 1 Search terms used in Medline, Embase, and Scopus

Database	Search terms
Medline	(((“Rural Health”[MeSH] OR “Rural Health Services”[MeSH]) OR rural[All Fields] OR (underserve[All Fields] OR underserved[All Fields] OR underserved’[All Fields] OR underserves[All Fields]) OR remote[All Fields]) AND ((“students, dental”[MeSH Terms] OR (“students”[All Fields] AND “dental”[All Fields]) OR “dental students”[All Fields] OR (“dental”[All Fields] AND “students”[All Fields])) OR “Students, Dental”[MeSH] OR “Education, Dental”[MeSH])) AND hasabstract[text] AND (“2000/01/01”[PDAT]: “2015/12/31”[PDAT]) AND “humans”[MeSH Terms] AND English[lang]
Embase	(exp rural health care/or exp rural population/or exp rural area/) AND (exp dental student/OR exp dental education)
Scopus	((TITLE-ABS-KEY (dental education)) OR (TITLE-ABS-KEY (dental student*))) AND ((TITLE-ABS-KEY (underserve*)) OR (TITLE-ABS-KEY (rural))) AND (LIMIT-TO (PUBYEAR, 2015) OR LIMIT-TO (PUBYEAR, 2014) OR LIMIT-TO (PUBYEAR, 2013) OR LIMIT-TO (PUBYEAR, 2012) OR LIMIT-TO (PUBYEAR, 2011) OR LIMIT-TO (PUBYEAR, 2010) OR LIMIT-TO (PUBYEAR, 2009) OR LIMIT-TO (PUBYEAR, 2008) OR LIMIT-TO (PUBYEAR, 2007) OR LIMIT-TO (PUBYEAR, 2006) OR LIMIT-TO (PUBYEAR, 2005) OR LIMIT-TO (PUBYEAR, 2004) OR LIMIT-TO (PUBYEAR, 2003) OR LIMIT-TO (PUBYEAR, 2002) OR LIMIT-TO (PUBYEAR, 2001) OR LIMIT-TO (PUBYEAR, 2000)) AND (LIMIT-TO (DOCTYPE, “ar”)) AND (LIMIT-TO (SUBJAREA, “DENT”)) AND (LIMIT-TO (SRCTYPE, “j”))

Figure 1 Overview of the article-selection process.

Table 2 Data-extraction table presenting key characteristics of each study

Study	Site	Design	Participants	Data collection	Exposure	Event	Definition of “rural”
DeCastro et al^Citation28	USA	Cross-section	Dental graduates	Self-administered questionnaire	Alumni of the Community-Oriented Dental Education (CODE) program	Reporting interest in helping communities after graduation	Underserved area where the distance from schools is about 145–195 km
Johnson and Blinkhorn^Citation29	Australia	Cross-section	Dental students	Self-administered questionnaire	Fourth-year students who participated in a rural placement program	Intention to work in rural location after they graduate	Not clearly described
McFarland et al^Citation32	USA	Retrospective cohort	Dental graduates	Tracking alumni records	Rural background (based on domicile where respondents were brought up or the location of their high school)	Currently working in a rural county after graduation	Counties with populations of 50,000 or fewer
McMillan and Barrie^Citation33	South Africa	Cross-section	Dental students	Self-administered questionnaire	Rural background (based on domicile where respondents were brought up)	Intention to work in rural areas	Administrative areas smaller than a town or city
Johnson and Blinkhorn^Citation30,b	Australia	Cross-section	Dental graduates	Tracking records of dental graduates from the Australian Health Practitioner Regulation Agency	Graduates who participated in rural placement program	Currently working in rural areas	Area with a population of fewer than 100,000
Thammatacharee et al^Citation34	Thailand	Cross-section	Dental graduates	Self-administered questionnaire	Rural background, defined as having been brought up in a rural area during the first 15 years of life	Intention to practice in a community (district) hospital given, not contract-bound	Nonprovincial capital district of any province outside Greater Bangkok
McQuistan et al^Citation31	USA	Cross-section	Dental graduates	Self-administered questionnaire	Participating in Community-Based Dental Education (CBDE) program (a 10-week rotation in community health centers or nursing homes)	Current involvement in charitable dental service in community	Not clearly described

Notes:

a Questionnaire did not directly ask respondents about their intention to serve rural areas. The respondents were asked to report whether they agreed that “interest to serve the communities” was an important factor in selecting a workplace after graduation.

b 2008 data not included, since in 2008 the clinical rotation program had not been fully implemented.

Table 3 Potential biases and methodological limitations in each study

Study	Critical limitations or potential biases
DeCastro et al^Citation28	• The question used to gather information on practice settings might be subject to misinformation bias. For instance, “community-based practice setting” could describe either a private practice in a community or a public community health center.
	• The questionnaire asked a respondent to report if he/she agreed with the statement that interest in serving rural areas was the key factor in their workplace selection, rather than asking his/her willingness to practice in rural areas.
Johnson and Blinkhorn^Citation29	• There was the potential for recall bias and judgment error because the timing for completing the questionnaire was different between the exposure and control groups.
McFarland et al^Citation32	• The college instigated numerous special programs to encourage rural placement of graduates. However, the study showed only the effect of a rural background on students, while other measures to encourage rural placement were not assessed.
	• The study did not clearly indicate whether the practice after graduation referred to the job placement immediately after graduation or the job placement of graduates at the time of completing the study.
McMillan and Barrie^Citation33	• Respondents who were unsure of their future workplace were not included in the analysis (n=11).
	• The study participants were dental students from different years. About a quarter (n=80) of participants were first-year students. The first-year students’ willingness to work in rural areas might have changed when they became final-year students.
Johnson and Blinkhorn^Citation30	• As the study explored the records of the workplace location of dental graduates at just one point in time, this might have been subject to bias, since job vacancies for dentists varied by month.
Thammatacharee et al^Citation34	• The survey was conducted at the annual health-workforce meeting arranged by the Ministry of Public Health (MOPH). This is the meeting where the MOPH assigns new graduates to available workplaces. Taking an MOPH post is obligatory for all doctors and dentists graduating from public institutions. Graduates who intend to enter private practice right after graduation are required to pay a fine and need not attend the workplace-selection meeting. Therefore, the survey might have encountered selection bias.
McQuistan et al^Citation31	• The nonevent group comprised a mixture of participants, ie, those not providing charitable service and those providing charitable service, but not in a community setting.
	• The study did not clearly indicate the operational definition of “community” setting.
	• The survey suffered from a large volume of nonrespondents (response ratê41%).

Figure 2 Meta-analysis of all seven studies with random-effect model, showing results in terms of OR.

Abbreviations: OR, odds ratio; CI confidence interval.

Table 4 Number of “exposure” and “control” participants in each study

Study	Total participants, n	Exposure participants, n	Control participants, n	Exposure participants with event, n	Exposure participants without event, n	Control participants with event, n	Control participants without h event, n
DeCastro et al^Citation28	111	46	65	38	8	39	26
Johnson and Blinkhorn^Citation29	62	32	31	31	1	11	20
McFarland et al^Citation32	1,361	708	653	361	347	107	546
McMillan and Barrie^Citation33	284	14	270	11	3	94	176
Johnson and Blinkhorn^Citation30	77	30	47	13	17	7	40
Thammatacharee et al^Citation34	202	45	157	35	10	97	60
McQuistan et al^Citation31	541	427	114	320	107	61	53

Figure 3 Meta-analysis of all seven studies with random-effect model, showing results in terms of RR.

Note: A P-value of 0.000 is a P-value of less than 0.001 due to the standard STATA output.
Abbreviations: RR, risk ratio; CI confidence interval.

Figure 4 Subgroup meta-analysis between rural background and rural clinical rotation.

Abbreviations: OR, odds ratio; CI confidence interval.

Figure 5 Funnel plot of the seven studies.

Abbreviations: SE, standard error; CI confidence interval.

DeCastroJEMathesonPBPanagakosFSStewartDCFeldmanCAAlumni perspectives on community-based and traditional curriculaJ Dent Educ200367441842612749571

 PubMedGoogle Scholar

JohnsonGBlinkhornAAssessment of a dental rural teaching programEur J Dent20126323524322904650

 PubMedGoogle Scholar

McFarlandKKReinhardtJWYaseenMRural dentists: does growing up in a small community matter?J Am Dent Assoc201214391013101922942149

 PubMed Web of Science ®Google Scholar

McMillanWJBarrieRBRecruiting and retaining rural students: evidence from a faculty of dentistry in South AfricaRural Remote Health201212185522449176

 PubMed Web of Science ®Google Scholar

JohnsonGBlinkhornAThe influence of a clinical rural placement programme on the work location of new dental graduates from the University of Sydney, NSW, AustraliaEur J Dent Educ201317422923524127764

 PubMed Web of Science ®Google Scholar

ThammatachareeNSuphanchaimatRWisaijohnTLimwattananonSPutthasriWAttitudes toward working in rural areas of Thai medical, dental and pharmacy new graduates in 2012: a cross-sectional surveyHum Resour Health2013115324148109

 PubMed Web of Science ®Google Scholar

McQuistanMRMohamadAKuthyRAAssociation between dentists’ participation in charitable care and community-based dental educationJ Dent Educ201478111011824385530

 PubMed Web of Science ®Google Scholar