Determining factors in the retention of physicians in rural and underdeveloped areas: a systematic review

Background Imbalance in distribution of Health Care Workers (HCWs) in a country is a global challenge. Almost all of the rural and underdeveloped areas are struggling with the shortage of HCWs, especially physicians. Therefore, this study aimed to identify factors governing the retention of physicians in rural and underdeveloped areas. Methods International databases including Scopus, PubMed, Web of Science, Proquest, and Embase were searched using Mesh terms in order to find peer-reviewed journal articles addressing physicians’ retention factors in rural and underdeveloped areas. The records were screened, and any duplicate results were removed. The quality of the studies was assessed according to the Critical Appraisal Skills Program developed for different types of studies. Then, through content analysis, the related factors were identified from finally selected papers, coded, and categorized. Results The initial search resulted in 2312 relevant articles. On the basis of specific selection criteria, 35 full-text articles were finally reviewed.. The major affecting factors in physicians’ retention in rural and underdeveloped regions were classified into the following six categories: 1) financial; 2) career and professional; 3) working conditions; 4) personal; 5) cultural; and 6) living conditions factors. Conclusion There is a complex interplay of factors governing physicians’ retention in rural and underdeveloped areas. If health organizations are concerned with physicians’ retention in deprived areas, they should take into account these main factors. Moreover, they should develop policies and strategies to attract and retain physicians in rural and underdeveloped areas.


Background
Health Care Workers (HCWs) are the backbone of health organizations [1] and are considered as the main factor in public access to health services [2,3]. In most countries, 10% of the total government employment is allocated to the health sector [4]. The World Health Organization (WHO) defines Health Workers (HWs) as all the people who are engaged in actions primarily intended to improve health [5]. The imbalance in the distribution of HCWs within countries is a common serious problem and can lead to challenges such as shortage of physicians in underdeveloped areas which in turn could pose challenges for the public access to health care services [6,7].
Physicians play a vital role among the HCWs [4]. Although they form a major group of HCWs, they have been neglected in the discussions about the health system functions in many developing as well as developed countries [8]. Migration of physicians from rural and underdeveloped areas to affluent ones has caused major concerns for health policymakers. A disproportionate distribution of physicians can cause a disturbance to the performance and stability of health systems [9]. In most of the rural and underdeveloped areas, lack of physicians is a long-standing challenge with serious consequences for the quality and quantity of provided services and equity of access [10]. Moreover, it restricts access to health services for people in these areas.
Previous experiences suggest that, in general, increasing the number of physicians in a country is not a determining factor in the increase in their retention rate in rural and underdeveloped areas [11,12].
Conducted in 1998 in the United States, one study revealed that there was a 75% shortage of physicians in rural areas [13]. The results of a similar study in Argentina in 2014 showed that 21% of the physicians had a strong desire to leave the deprived areas, 57.3% had a moderate desire to desert these areas, and 21.5% were reluctant to leave the aforementioned areas and were willing to be active in their relevant areas [12]. In Iran, the distribution of HCWs is not equitable between and within provinces, especially in deprived and underdeveloped areas [14,15]. Based on a study conducted in Kerman in 2014, the proportion of doctors who had quit working in deprived and rural areas was 26, and 77.3% of them intended to leave health services in the near future [16].
Many countries suffer from critical shortage of physicians in marginal and underdeveloped rural areas. Moreover, little is known on determining factors in the retention of physicians in these areas. Therefore, this study aims to identify pivotal factors governing physicians' retentions in rural and underdeveloped areas in order to develop evidence-informed policy interventions to deal with the issue of physicians' inaccessibility.

Search strategy
A systematic review of the literature was conducted in order to identify the factors that influence physicians' retention in rural and underdeveloped areas. We searched five distinct databases including Scopus, Web of Science, PubMed, Embase, and ProQuest. The search focused on the articles which had been published in the relevant databases from their inception until December, 2019. The reference lists of the articles were also checked to find additional studies to complete the search process and to find all of the relevant studies. The detailed search terms were developed in consultation with a librarian. The search terms and the sample of search strategy, taken from the PubMed database, are provided in Table 1 in the Additional file 1.

Study selection process
All of the retrieved citations were imported to EndNote (V. X8; Clarivate Analytics, Philadelphia, PA) after searching in the relevant databases. The title and abstract (if available) of each unique citation were screened by one reviewer (NM) according to the prespecified inclusion and exclusion criteria. The final stage of the screening of the articles (full-text review) was carried out by the two researchers (NM and LD). The researchers (NM and LD) independently extracted relevant data from the articles. In cases of disagreements, the researchers' views on the elimination of the studies were compared in order to reach a consensus. There were few discrepancies in the data extractors' views.

Inclusion and exclusion criteria
On the basis of the inclusion criteria, the selected studies: 1) were published in English language; 2) focused on physicians, including general practitioners, general physicians, family physicians and specialists as the target population; 3) were related to deprived, remote, rural and underdeveloped areas; 4) involved any type of article including letters to the editor, any type of review, original articles, debates, and perspectives; and 5) were fulltext articles.
Moreover, based on the exclusion criteria, the excluded articles were: 1) the ones whose target populations were not physicians for example they were administrative staff, managers, or nurses in the health sector; 2) full-text articles were not available.; and 3) were low quality based on the quality appraisal analyses.

Data extraction
In order to extract the relevant data, a data extraction checklist was used. Title, author(s), referencing style, country, year of publication, study type, study population, data collection method, extracted factors, quality appraisal, effect of the context on how the factors influenced the retention of physicians in rural and underdeveloped areas, and the positive or negative effects of the extracted factors on the retention of physicians were noted. In addition, the checklist contained a column for presenting research findings which were related to the factors governing the retention of physicians in the underdeveloped areas and their effects.

Quality appraisal
The Critical Appraisal Skills Program (CASP) checklists were used to assess the quality of the articles. CASP checklists are quick, simple, and straightforward tools for answering common questions on a scientific study [17]. We downloaded various relevant standard checklists from the website of the program (www.casp-uk.net) for the various types of studies (i.e., cross-sectional, review, qualitative, and analytical) which were included in our study.
The used checklists included 10 questions on the aims of the research, appropriateness of the methodology, research design, data collection, and data analysis. In order to assess the quality of studies, each question was scored 1 point for Yes or 0 for No. Then, the studies were categorized as low quality (i.e., studies with a score range of 0 to 3), medium quality (i.e., studies with a score range of 4 to 6), and high quality (i.e., studies with a score range of 7 to 10). Two of the researchers conducted quality appraisal independently. Finally, the differences in the researchers' appraisals were resolved by discussion.

Data analysis
Inductive content analysis was performed in order to understand, extract, and categorize factors governing the retention of physicians in rural and underdeveloped regions. Two members of the study team coded and categorized the data. To this end, they: 1) examined the selected full-text studies carefully and extracted the determining factors, 2) developed potential themes which involved the extracted factors, 3) generated a thematic map, 4) gave a name to the themes and defined and generated examples for each of them, and 5) assessed the reliability of the analyses through discussions regarding the contradictions to achieve full agreement.

Results
The searching process from the five databases provided the researchers with a total of 2312 articles: (1737 in Scopus, 289 in Embase, 215 in PubMed, 62 in Proquest, and 9 in ISI Web of Science,). A total of 490 duplicate articles were identified, leaving us a total of 1822 unique citations on the topic. The screening process of the extracted studies is shown in Fig. 1.
Moreover, 1311 titles were excluded due to the fact that they were not relevant to the purpose of study. The abstracts of 511 remaining studies were reviewed. Furthermore, 406 studies, whose aims were different from our purpose, were excluded. The remaining 105 full-text studies were carefully examined. On the basis of this examination, 71 studies were excluded because they only described the distribution of physicians and their desertion rates. In addition, these studies did not identify factors associated with retention or desertion. However, Fig. 1 Flow chart of study selection process two related studies were identified using a manual search (reference by reference) and were added to the study. At the end of this phase, 36 studies were included in the present study. One article did not get an acceptable quality score and was excluded from the corpus. A careful examination of the selected articles showed that 45 and 55% of the studies were concerned with the exploration of the factors governing the retention of physicians in underserved settings of the developed countries and the developing or underdeveloped countries, respectively. The characteristics and findings of all included studies are presented in Table 1.
The identified themes were classified into six main categories of factors: financial, career and professional, working conditions, personal, living conditions, and cultural.
Financial factors focused on the physicians' income level, salary, payment interval, reimbursement of tuition fees, various allowances and loan repayment. Personal factors were related to the physicians' demographic factors such as age, sex, education level, marital status, number of children, rural background, previous exposure or service in a deprived area and birthplace. Career and professional factors included factors such as job performance, evaluation of superiors, job improvement and educational opportunities, job position, willingness to change the performed job or the employer or support for professional development or research marketing and opportunities to achieve professional goals.
Working conditions or environment referred to the employment relationships and comprised all of the existing circumstances which affected physicians' work in the workplace, including work schedules, breaks, work hours, legal rights, and responsibilities, possibility of relocation and job security and job flexibility. Living conditions or recruitment situation encompassed the conditions in which physicians lived such as suitable housing and accommodation, educational opportunities for the children, access to refreshment facilities as well as recreational activities. Finally, cultural factors involved the customs, traditions, beliefs, moral values, code of communication and used language(s) of the rural community. Table 2 provides the influential factors in the retention of physicians in rural and underdeveloped areas based on the identified category of factors. Moreover, Fig. 2 provides information on the percentage of the influential factors in the selected studies.

Discussion
In this study, factors governing the retention of physicians in rural and underdeveloped areas were identified through a systematic literature review. According to our findings, the following six major categories govern the retention of physicians in the mentioned areas: financial, professional, working condition, living condition, cultural, and personal factors.

Financial factors
Most of the studies had examined the desertion and retention of physicians in rural and underdeveloped areas mainly based on financial factors, such as income, salary, loans and appropriate reimbursement. One of the major incentives for most of the physicians is their revenue. Many of them neglect the well-being of cities in order to have more income, although this is not the only motivational factor. Physicians who were employed in underdeveloped and remote areas had numerous complaints regarding their low income, reimbursements and compared them with their counterparts in the urban areas.
Financial rewards and incentives have a significant effect on the desertion and retention of physicians in the aforementioned areas. It makes sense for people to prefer urban and well-developed areas to rural in terms of equal income.
Two studies reported that the inappropriateness of income rates and interrupted payments cause the physicians' reluctance to serve in rural areas [16,20]. In most of the studies, salary was a factor significantly affecting the retention rate in rural and underserved areas [46,47]. Furthermore, the physicians in rural and deprived areas had fewer opportunities for working in private sector and seemed to be less satisfied with their incomes [23]. A study by Colleen Morken et al. (2018) showed that loan repayment was the least important factor in physicians' retention in rural areas [48]. Pathman and Konrad (2004) indicated that economic vitality of communities attracted physicians. However, it did not influence physicians' subsequent retention [21]. The result of another in Turkey demonstrated that an increase in physicians' income reduced their willingness to work in rural areas [25]. Thus, the solution to this challenge is to increase the physicians' salaries or to change reimbursement methods [26].

Career and professional factors
Providing educational and skill improvement opportunities was another factor in the satisfaction of physicians who served in deprived areas [5]. Nonetheless, there was no public agreement on it [26]. Medicine is a science that needs to be updated based on the newest validated information and knowledge. Although there are many possibilities for being updated in today's world, these facilities are less accessible in rural areas. Some physicians, who worked in rural and deprived areas, argued that activities in rural areas prevented them from developing professional skills [25]. According to the evidence, educational factors were the most effective factors          [34]. Therefore, the relevant educational and professional policies should be considered. These policies include reserving the post-graduate seats for physicians who have worked in remote areas, providing the physicians in rural areas with the opportunity to teach certain courses at universities of medical sciences, and arranging continuing educational programs for rural physicians [8].
On the other hand, the results of Colleen Morken's study (2018) showed that professional development opportunities and opportunities to teach medical students or residents' courses did not have a great effect on the physicians' retention in rural areas [48].

Working conditions factors
Our findings showed that working conditions factors are the factors that have significant influence on the retention of physicians in rural or under development areas.
The working conditions factors have been mentioned in most of the studies. In some studies, working conditions, environment, and opportunities for professional improvement were mentioned as affecting factors on physicians' recruitment and retention in rural areas [44,49]. Our study showed that unavailability of surgical equipment and the lack of essential drugs are discouraging factors in the physicians' work in the rural and underdeveloped areas [50]. In   (20,23,18,25,27,29,41,43,45) ➢ Adequate and fair pay scale (20) ➢ Transparency of the payment system (16) ➢ High salaries (9,24) ➢ High Earnings from the practice (9) ➢ Bolstering incomes for employed physicians (14) ➢ Favorable salary levels (14) ➢ Increasing reimbursement levels (14) Remuneration (12,18,23,27) ➢ Desirable average monthly payments (7) ➢ Various allowances (7) ➢ Financial bonus (7) ➢ Good revenues/patient volume (13) ➢ Financial support (8 (25,34,19) ➢ Supporting systems (19) ➢ Staff function as team (19) ➢ Supportive family and work/life balance (19) ➢ Supportive administration (19) ➢ Effective support staff (19) ➢ Social justice (19) ➢ Developing a locum service to allow rural doctors to take annual leave and sabbaticals (20) ➢ A reasonable level of infrastructure (20) ➢ Working security (16,20) ➢ A secure working environment (20) ➢ Opportunities for continuing medical education (20) ➢ Compulsory services commitment (16) Educating native manpower (16) ➢ Workplace with a high level of progressive healthcare (9) ➢ Good working environment (9) ➢ Good infrastructure (9) ➢ Rich instrumental background (9) ➢ Reasonable workload (9) ➢ Incentive benefits (9) ➢ Good equipment (9) ➢ Access to medical information and consultations (9) ➢ Opportunities to achieve professional goals (9)   (7) ➢ Increasing level of the facility (7) ➢ Call coverage (13) ➢ Provision of improved working conditions (8) ➢ Improving living and working conditions and environment (8) ➢ Good service infrastructure (25) Educational incentives to promote working (12) ➢ Personal security at workplace (12) ➢ Access to online information technology for bibliographic search (12) ➢ Ability to use new cutting edge technologies (12) ➢ Access to specialist referral network via telephone and web (12) ➢ Adequate infrastructure at workplace (12) ➢ Access to high-complexity regional hospital to refer patients ( Fig. 2 Percentage of the influential factors in the selected studies other words, the developmental condition of the region, where the physicians worked, was a very important variable adversely shaping their preferences [25]. The lack of job security, heavy workload, length of working hours, and the multiplicity of the physicians' responsibilities in rural health centers were among the other important factors in physicians' unwillingness to stay in the deprived areas. In Vietnam, there was a distinct lack of job replacement among physicians. The physicians who worked at a certain level of an organization could only be transferred within the same region and geographical area from one level to another level in other organizations [5].
In rural areas, usually, there is not any medical specialist that could help or guide other physicians specially GPs. Thus, they should visit patients and make decisions by themselves at times even without any preclinical exams or have to work seven days of the week.
Job security and the possibility of relocation were introduced in the Shankar's (2010) review as factors in the retention of physicians [8].
Evidence showed that there was a strong relationship between the physicians' occupational group and choice of workplace. For example, specialists who worked in hospitals with access to advanced technologies were reluctant to work in rural or underdeveloped areas where technological facilities were meager [25]. Therefore, it was stated that an increase in specialization contributed significantly to the imbalance in the geographical distribution of physicians [51].

Personal factors
Personal factors were introduced in a limited number of studies [9] as influencing factors in physicians' retention in deprived areas. Prasad and Amatya (2015) argued that the personal factors including demographic factors are more significant than the financial factors [28]. Among the demographic factors, gender and age had the greatest impact In general; male physicians' tendency to serve in rural and underdeveloped areas was greater than the female physicians [12,26]. This issue may stem from the fact that female physicians harbor more concerns on workplace conditions, and their choices are often more affected by factors such as security issues, responsibilities, family commitments, living conditions and life considerations in comparison to their male colleagues [25,52].
In the literature, the findings concerning the effect of age on working in rural and underdeveloped areas are contradictory. According to Mollahaliloglu (2015), young physicians were the groups that could be easily motivated to work in rural areas [25]. However, most of the studies showed that older physicians had a greater tendency to serve in deprived areas in comparison with the younger ones [12,22]. It can be argued that younger physicians often place a higher value on education, career progression, and leisure activities compared to older physicians.
According to Joseph Lee)2016(, family ties were one of the factors which influenced the physicians' decision on retention in rural areas [44]. Therefore, the selection of physicians based on their readiness and desire to work in underdeveloped areas as a result of either interest or family ties in these areas can enhance their recruitment and retention in underdeveloped areas. The physicians, who were from the ethnic communities in rural areas along with the physicians who were born and raised in rural places (i.e., had a rural background), were more inclined to serve in rural and underdeveloped areas [24]. One possible explanation is that these physicians were more familiar with the deprived environments and were able to easily adapt themselves to the conditions and problems of the deprived environments. Furthermore, the physicians, who passed their courses in the deprived areas or had a record of serving in deprived and rural areas, tend to serve in these areas more readily in comparison with other physicians without such backgrounds. This issue stems from the fact that during their studies, these physicians had seen and experienced the living conditions in the rural environments and have kept preparing themselves to deal with such problems [2,34]. Similarly, the findings of a number of other studies confirmed that rural background, rural origins, and rural lifestyle were the factors highly associated with recruitment and retention of health professionals in rural areas [12].
Some physicians enjoy serving in deprived areas because they had a high motivation to treat the patients who were in need of help [13]. In other words, these physicians were happy to serve in deprived areas because they believed that their efforts helped to establish a mutual understanding among community and society members.

Living conditions factors
Deprivation rates of an area had a direct impact on physicians' tendency to stay in their workplaces or to leave them. The deprivation severity of the region was a major factor in the physicians' desertion or decisions on the desertion of their workplaces in the following years [53].
Lack of amenities, accommodations, and some deficiencies such as lack of communication and internet systems, recreational facilities and interesting places prompted the physicians to leave their workplaces [14,54].
In some cases, long distances between the rural areas and the urban centers restricted the physicians' access to shopping centers and stores which were necessary to buy their required food and clothing items. These factors made the physicians reluctant to continue to serve in these areas [13]. Furthermore, the physicians, who worked in the rural areas, faced a higher risk in terms of health and safety in comparison with the physicians who worked in the urban areas. More specifically, the risk of certain diseases was high in rural and remote areas due to the fact that people disregarded some health issues [5]. Evidence showed that lifestyle opportunities, including the availability of cycling opportunities, parks and restaurants together with adequate recreation affected the retention of physicians in rural areas. Therefore, providing the physicians with recreational opportunities in underdeveloped and rural areas should be taken into consideration [18,44].

Cultural factors
A limited number of studies have addressed cultural factors in the retention of physicians. The customs, traditions, beliefs, moral values, language, and laws of the rural community can help GPs to work in rural areas for a long period of time, but in some cases, they could be counterproductive.
The existence of appropriate bilateral relationships between physicians and patients was one of the important issues in these areas and could lead to these physicians' satisfaction and retention in deprived areas [33]. A number of physicians reported the people's breach of etiquette in deprived and rural areas [22]. This issue could stem from the people's poor living conditions and their lack of awareness of the rules and conventions of the etiquette in health centers. Therefore, it is necessary to develop a framework and code for communication between the physicians and their patients in order to increase the physicians' motivation to serve in disadvantaged areas [28]. The government policy makers and universities of medical sciences have a major effect on the culture of working in limited-resource and underprivileged areas of the countries [12].

Strengths and limitations
To the best of our knowledge, this is the first study of its kind to systematically explore all the factors governing the retention of physicians in underdeveloped or deprived areas in both developed and developing countries published papers, although we only included papers that were written in English. Strength of this review is conducting a systematic search in five databases, supplemented by a hand search of the bibliography in the articles identified. This review is limited by the difficulty of analyzing, identifying the category of some extracted factors and comparing findings of studies conducted in a wide diversity of contexts, with a variety of using different study designs.

Conclusion
This study demonstrates that there are different factors influences the physicians' interest in rural and remote areas and their retention in these areas. A good number of studies over this issue have been conducted in developing and developed countries. Working conditions factors and financial factors deserve healthcare policy makers' particular attention among the factors which are associated with the retention and willingness of physicians to serve in deprived areas. Recruiting physicians, who are from rural backgrounds and rural origins, is another determining factor in physicians' retention which has to be considered by the policy makers who aim to promote the physicians' retention in rural areas. There is no enough evidence regarding the cultural factors and their effect on the physicians' retention in the mentioned areas. Further studies are warranted to determine the quantitative effects of the identified determining factors on the retention of physicians in rural and underdeveloped areas.
Identifying the influential factors can be useful to policy makers and health planners who intend to develop evidence-informed policy interventions and executive programs in order to guarantee the retention of the healthcare workers in these areas which is a crucial factor in the improvement of population health.