Logo of westjmedLink to Publisher's site
West J Med. Oct 2000; 173(4): 240–243.
PMCID: PMC1071101

Recruitment and retention in the Navajo Area Indian Health Service


Objective To determine why physicians and midlevel providers join, leave, or stay in Navajo Area Indian Health Service (IHS). Design Cross-sectional analysis of data obtained from questionnaires. Setting Navajo Area IHS hospitals. Subjects Navajo Area health care physicians and midlevel health care providers. Main outcome measures The prevalence of physicians and midlevel providers who plan to leave the Navajo Area IHS, the demographic characteristics of these physicians and midlevel providers, and the most common reasons for staying or leaving. Results A total of 221 (64%) physicians and midlevel providers responded. Of these, 58% planned to leave eventually, and 47% of all physicians and midlevel providers planned to leave in the next 3 years. Physicians and midlevel providers planning to leave tended to be younger than those planning to stay (P = 0.009). The most common reason to join the IHS was a desire to work in the Southwest, to stay was the quality of the medical staff, and to leave was lack of administrative support. Conclusions A high turnover rate of physicians and midlevel providers may occur in the next 3 years. A combination of factors specific to the provider, the institution, and the environment attracts physicians to the Navajo IHS and encourages them to stay. Factors that push physicians and midlevel providers to leave tend to be specific to the institution and are potentially amenable.


The Indian Health Service (IHS), a division of the US Public Health Service, provides care to Native Americans, including the Navajo Nation. The IHS suffers from understaffing of physicians and midlevel health care providers; in 1998, an overall 16% vacancy rate existed for physicians.1 Understaffing is a serious problem because it can restrict the range and quality of services provided.

In the past, efforts to correct this problem have focused on recruitment at the federal level. The Health Professions Support Team was formed to help coordinate recruitment activities at various service units, and a survey performed in 1998 questioned a random sample of IHS physicians as to what motivated them to join the IHS.1 The Loan Repayment Program was designed to provide a financial incentive for physicians to join the IHS, and scholarship provides funds for medical school in return for a pledge to work in underserved areas. Title 38, federal legislation that allowed health service centers greater flexibility in setting pay scales, resulted in an increase in pay for many IHS physicians.

Although these efforts have probably assisted in correcting the understaffing problem, the problem still exists. In 1998, a 50% physician vacancy rate existed at one Navajo facility for 5 months. One hypothesis for the continued problem is that understaffing is a function of both recruitment and retention rates; if retention is excellent, recruitment rates will not necessarily affect staffing. However, if recruitment is excellent and retention is poor, a facility can remain understaffed. Therefore, efforts focused mainly on recruitment will not necessarily succeed.

Another explanation is that factors unique to a particular location may not be easily amenable to federal level intervention because individual facilities are increasingly responsible for their own recruitment and retention. Rates of understaffing vary between locations; some areas have ongoing 65% vacancy rates, and others have none.2 An example of such a unique factor is that certain tribal authorities are taking control of the IHS unit in their area, with continued federal funding. This might influence staffing in particular areas, without any influence on other locations. Another example is that different IHS locations have different pay scales and financial incentives, although all physicians and midlevel providers are employees of the Public Health Service. Therefore, financial programs aimed at IHS physicians and midlevel providers as a whole may not address the needs of a particular service unit.

To determine what factors were associated with provider recruitment and retention in the Navajo Area, I conducted a survey of Navajo Area IHS physicians and midlevel health care providers. The Navajo Nation is unique in that it has more than 200,000 members and is the second largest in the country. Also, it is the largest geographically, covering more than 64,000 km2 (>25,000 mi2), the size of West Virginia.3 The objectives of the survey were to determine what provider demographics and institutional variables were associated with both recruitment and retention. Questions examined factors in recruitment found to be significant in other surveys of physicians of underserved groups.4,5,6,7,8 Also, the survey asked specifically about the effects of Public Law 93-638.9 This federal law allows tribal authorities to manage their health care system with continued federal funding, and the Navajo Nation plans to assume control of their system in the next 2 years.


In mid-August 1998, the questionnaire was distributed to the clinical directors of all Navajo Area IHS facilities, who distributed it to physicians and midlevel health care providers in their facility. A preaddressed envelope was provided so that each provider could mail the survey directly without the clinical director seeing the response. Twelve facilities are listed as Navajo Area IHS facilities and range from outpatient clinic-only services to tertiary care centers; all serve rural communities. The questionnaire was developed in a focus group consisting of a physician at a Navajo Area IHS facility and the clinical directors; it was pilot tested among the staff at a Navajo Area IHS facility. Eligibility was limited to health care professionals including physicians, nurse practitioners, physician assistants, nurse anesthetists, and midwives who were not under locum tenens contracts. Clinical directors were sent periodic reminders for their staff to finish the questionnaires. The surveys were collected until January 1, 1999.

The questionnaires were 3.5 pages long and began with questions on age, sex, race, marital status, number of children, upbringing in a rural or underserved area, the size of the community where they trained, degree, board certification, specialty, length of employ in the IHS and in the facility, number of hours worked, and amount of money still owed on medical school loans at the end of training. Participants were asked to rank on a 5-point Likert scale 36 factors that influenced their joining the IHS. They also were asked to rank by order of importance in fill-in-the-blank format the reasons they decided to join the IHS, if they planned to leave, and when. Finally, in order of importance, they were asked to list in a fill-in-the-blank format the reasons they decided to leave or stay.

The demographic characteristics of the participants were described using percentages for dichotomous variables. The characteristics of participants who planned to leave were compared with those who did not, using χ2 tests for dichotomous variables. Also, characteristics of participants who planned to leave in the next 3 years were compared with those who planned to leave after that time and with those who planned to stay. To adjust for possible confounding, a multivariate logistic model was used to evaluate variables associated with plans to leave and when leaving was planned. Open-ended responses were coded, and the most common reasons listed for joining, staying, or leaving were tabulated. In 3 instances, the reasons for both leaving and staying were listed without a preference for leaving or staying, and these results were not used. All statistical tests were carried out using commercially available statistical software (STATA; Stata Corp, College Station, TX).


The clinical directors were asked how many staff members eligible to answer the questionnaire existed at each site, and these totaled 344; 221 staff members (64%) responded. We did not collect data on nonresponders because several of the IHS facilities had few physicians and midlevel providers, and there was concern about confidentiality of the responses. In all, 129 planned to leave, 89 planned to stay with their IHS facility, and 3 did not comment. Of those who planned to leave, 108 (48%) planned to do so in the next 3 years. Characteristics of the respondents and respondents who intend to leave within 3 years did not differ significantly by sex, race, marital status, presence of children, employment status of spouse, rural or underserved upbringing, completion of a residency, location of residency, specialty, loan burden, or a work week of more than 40 hours. Increased age (P = 0.004) and increased time spent in the IHS (P = 0.01) were significantly associated with plans to leave within the next 3 years. However, in a multivariate analysis, only age remained significant: if the provider was between 45 and 55 years old, they were less likely to leave than a person between 25 and 35 years old (odds ratio: 0.30; 95% confidence interval, 0.16-0.88). Response rates between different clinics did not vary significantly (P = 0.06).

Reasons for joining

When open-ended responses were tabulated, the most common reasons for joining the IHS were the desire to work in the southwestern United States, followed by the desire to work with the underserved, schedule flexibility, loan repayment, and variety in work content. These differed slightly from the reasons listed as those most influencing the decision to join as elicited from the Likert scales, which were desire to work with the poor, quality of the medical staff, job challenges, job variety, and recreational opportunities. Important reasons for joining on both the open-ended questions and Likert scales were the desire to work with the poor and the variety of work content. From the Likert scales, the most negative influences on joining were the quality of the schools, distance from family and friends, quality of administration, low salary, and lack of housing. From the median score and the percentage of responses above zero, a lack of jobs outside the IHS was not an important motivator to join.

Reasons to stay or leave

From the open-ended responses, the most common reasons to stay were the quality of the medical staff, schedule flexibility, desire to work in a rural area, desire to work in the Southwest, and the variety of work.

The most common reasons to leave were lack of administrative support and quality, fear of PL 93-638, distance from family and friends, excessive working hours, and the poor local school system. This did not vary significantly between different facilities in the IHS. The second most common reason for leaving was fear of PL 93-638 or the impending turnover of the Navajo Area IHS to the Navajo Nation. This fear was not elaborated on, and most often the comments stated that they did not know what to expect, although some were concerned that non-Navajo Native Americans would take their jobs or that a large drop in salary or negative change in benefits would occur.

Comments from physicians and midlevel providers

Although unsolicited, many physicians and midlevel providers elaborated extensively on their reasons to stay or leave. Because these comments come from unsolicited written responses on a survey with a fill-in-the-blank format, I did not subject them to formal qualitative coding.

The lack of housing was commented on by 10 respondents. One respondent summarized: “when one considers what physicians are being offered elsewhere, I think that fewer and fewer will be willing to accept the marginal housing facilities. Most of us come from an urban or community background and expect more. I know of young urban physicians who became very happy in rural settings [when adequate housing was provided].”

Lack of health care was also a negative influence, and one provider commented: “Health care is not provided for [those who are not] commissioned officers. It is insulting to new staff that we cannot provide basic health care for employees who are non-Native American. [These people] have to take an entire day of sick leave to go to a medical appointment, the nearest being 60 miles [away].”

Several physicians and midlevel providers who had worked in the IHS for more than a decade remarked that their situation had improved since they had started. One physician stated, “My generation of physicians has a hard time identifying with the concerns of younger physicians. Compared to years ago, work hours are much shorter, pay is much higher, educational and shopping opportunities are much greater, housing is more available and of better quality, etc.”

Younger physicians agreed about the pay not being a problem in the decision to stay, although it did not encourage them to join. One stated, “Title 38 [institution of premium special pay for physicians] has made a huge difference; I make more money now than most of the other residents who finished at the same time as me.” Physicians and midlevel providers who did not benefit from Title 38 stated that pay was inadequate. One nurse practitioner stated, “[We receive] no extra pay for hardship location when we live right next to medical doctors who get a bonus for same.”

Although exceptions were noted, one provider wrote, “Administrators are 1. slow to adapt to change, 2. totally lack respect for physicians, and 3. should be replaced by trained persons rather than the usual `on the job training' persons we've had.” Another elaborated, “They are polite, but they manage to do absolutely nothing to make their nursing or medical staff feel appreciated. There is minimal communication regarding issues that directly impact on patient care; in fact, our opinions are never requested; and if we express them, we are perceived as threatening and hostile.” This sentiment was also expressed by a provider who stated, “I am just a body, not a person to some.” Poor administration was thought to contribute to increased administrative duties for physicians and midlevel health care providers and excessive hours. One respondent stated, “A big frustration was not being given administrative time on a regular, predictable basis.”


This survey found that 108 (47%) of all Navajo IHS physicians and midlevel providers who responded to the survey planned to leave in the next 3 years. This low retention rate is lower than rates of retention in other rural health care facilities. In a 1992 survey, Pathman and colleagues found that roughly 25% of rural physicians not in the National Health Service Corps left their index practice and that about 80% left their index community after 3 years. Rates of retention for rural National Health Corps physicians were lower: about 20% remained in their index practice after 3 years, and about 40% remained in their index community.4

Concerning recruitment in 1998, Navajo Area IHS physicians and midlevel providers had personal reasons for joining, as indicated by their principal reasons for joining—desire to work with the poor and in the Southwest. However, institutional factors such as loan repayment, schedule flexibility, and work variety were important in recruitment. The attractions of the Navajo Area IHS and those cited as reasons for staying were similar and included the desire to work in the Southwest, the schedule flexibility, the variety of work, and the quality of the medical staff. These overcame concerns about the distance from family and friends, poor school quality, lack of easily accessible health care, comparatively lower salaries, and lack of housing.

The most common reasons cited for leaving, however, were institutional, such as lack of administrative support, quality and fear of impending tribal management, and excessive working hours. Other factors were environmental and located outside the IHS, such as the local school system. These findings are similar to those of a 1992 report examining understaffing in the IHS that found high vacancy rates due to inferior pay, “community and environment factors,” and “administrative support.”2

This survey has several limitations. Because no easily accessible statistics exist on health care professionals on the Navajo reservation, there is little information on nonresponders. Thus, respondents may not be representative of health care professionals on the reservation as a whole, despite the high 64% response rate. Also, we do not know how the respondents' answers corresponded with their decisions to stay on the reservation. Finally, themes could have been further clarified with more qualitative analysis. Particularly, further discussion of dissatisfaction with the administration may have yielded more specific and potentially amenable reasons.

Many unique factors draw applicants to the Navajo Area IHS, including the southwestern US location and work content variety. The rapport with the medical staff often encourages physicians and midlevel providers to stay. These reasons are not enough, however, to stem an efflux of physicians and midlevel providers, an estimated 47% in 3 years. Unique issues to the Navajo Area IHS are important in physicians and midlevel providers' intention to leave, such as the fear of Title 638, whereas factors such as low pay are not as important as in other areas. Further qualitative research on provider focus groups that emphasized confidentiality of responses would help clarify themes of dissatisfaction and possibly identify solutions.


I thank the participants of the survey for their time and effort and Dr Douglas Peter.


Funding This study was funded by the Robert Wood Johnson Foundation.

Competing interests: None declared

At the time this survey was done, Dr Kim was affiliated with the Crownpoint Healthcare Facility, Crownpoint, NM.


1. Bolte M. Factors influencing physician selection of rural or nonrural practice sites in the Indian Health Service. IHS Prim Care Prov 1998;23: 90-91.
2. Quality Management Recruitment and Retention of Healthcare Professionals Workgroup. Rockville, MD: Health Resources and Service Administration, Public Health Service, US Dept of Health and Human Services; 1993.
3. Etcitty D. Navajo Nation FAX 1993. Window Rock, AZ: Division of Economic Development, Navajo Nation; 1994.
4. Pathman DE, Konrad TR, Ricketts TC 3d. The comparative retention of National Health Service Corps and other rural physicians: results of a 9-year follow-up study. JAMA 1992;268: 1552-1558. [PubMed]
5. Verby JE. Improving the supply of physicians to rural areas [editorial]. JAMA 1992;268: 1597-1598. [PubMed]
6. Xu G, Veloski J, Hojat M, Politzer RM, Rabinowitz HK, Rattner SL. Factors influencing primary care physicians' choice to practice in medically underserved areas. Acad Med 1997:72: S109-S111. [PubMed]
7. Li LB, Williams SD, Scammon DL. Practicing with the urban underserved: a qualitative analysis of motivations, incentives, and disincentives. Arch Fam Med 1995;4: 124-133. [PubMed]
8. Gessert C, Blossom J, Sommers P, Canfield MD, Jones C. Family physicians for underserved areas: the role of residency training. West J Med 1989;150: 226-230. [PMC free article] [PubMed]
9. The Indian Self-Determination and Education Act amendments: notice of proposed rulemaking, 61 Federal Register 2037-2077 (1996). [PubMed]

Articles from The Western Journal of Medicine are provided here courtesy of BMJ Group
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


  • PubMed
    PubMed citations for these articles

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...