The literature on nurses' intent to migrate out of the nursing profession or their native country was expertly summarized by Kingma (2006), and recent research continues to update information about this complex problem (El-Jardali, Dimassi, Dumit, Jamal, & Mouro, 2009; Nguyen et al., 2008). Studies have primarily explored two sets of variables as predictors of nurses' intent to migrate. These include demographic characteristics (i.e., age, marital status, and level of education) and job characteristics (i.e., salary, opportunities for advancement, and work conditions—a set of variables often expressed as “job satisfaction”). The study described here explores a new construct, perceived HIV stigma, as a potential predictor of nurses' intent to migrate. This study asks the question, “Does perceived HIV stigma contribute to nurses' intent to migrate in five African countries?”
The World Health Organization (WHO, 2006) has United Nations nurses trained in Africa currently works outside of her or his home country, mainly in 30 industrialized countries. Migratory trends of nurses from low-resource countries to fill staffing needs in high-income countries is well documented (Kingma, 2006). Kingma described nursing as a mobile profession, with reasons for migration varying from better paying jobs to better working conditions to career mobility as well as the simple desire to do something different. In the current nursing literature, the most commonly cited reason nurses give for leaving nursing is job dissatisfaction (Aiken et al., 2001; Larrabee et al., 2003).
Definitions of job dissatisfaction have ranged from how employees feel about their jobs (Locke, 1976) to the extent to which employees report dislike for their jobs (Stamps, 1986). Factors contributing to nurses' job dissatisfaction that have been frequently cited in the nursing literature include low autonomy, job stress, and poor nurse-physician collaboration (Ferrand et al., 2003; Schmalenberg et al., 2005; Wells, Roberts, & Medlin, 2002). Uys, Minnaar, Reid, and Naidoo (2004) reported that nurses in South Africa were most satisfied with their contributions to patient care and least satisfied with salary. A study based in Uganda reported that certain job-related factors predicted job satisfaction (Hagopian, 2007). Those factors included having a match between skills and experience, satisfaction with salary, satisfaction with supervisor, satisfaction with workload, having a stimulating/fun job, and having a sense of job security. Some training activities designed to increase job satisfaction, such as providing opportunities for advanced clinical training, are impossible for some hospitals to support because they cannot spare the few nurses they have to attend trainings (Anderson & Isaacs, 2007).
In their review of literature on nurse migration, Nguyen et al. (2008) explored the “push” and “pull” factors related to nurses' migration. Push factors included low salaries, unsafe working conditions, lack of opportunities for educational advancement, unstable governments, lack of autonomy and respect for competency, and poor management. Pull factors included opportunities for higher salaries, safe practice environments, educational advancement, availability of positions, and reported higher job satisfaction. Nguyen et al. (2008) explored student nurses' intent to migrate in Uganda and reported that students desiring to work in rural or public facilities reported less intent to leave nursing or to migrate out of Uganda.
Governments and nongovernmental organizations are struggling to maintain a satisfied and adequate nursing workforce amid extreme poverty, governmental debt, low human development indices, and very few nurses and midwives for their populations. The nurse data collected in this study come from five of the poorest countries in the world: Lesotho, Malawi, South Africa, Swaziland, and Tanzania. Based on the United Nations Development Programmes' (2008) 2007 to 2008 Human Development Index ranking of 177 countries, South Africa had the highest ranking of this group at 121, and Malawi had the lowest ranking at 164 (See Table 1).
HIV and Stigma
HIV is an additional burden on these countries that are already highly affected by poverty and poor health. The disease has created an extraordinary demand for health care services in areas where health systems are already severely compromised and overwhelmed. At the same time, however, the disease is decimating the health care workforce. In Lesotho, Mozambique, and Malawi, death is the leading cause of health worker attrition, with a significant proportion being related to HIV (Medecins Sans Frontieres, 2007). HIV has also been associated with the intent to migrate. The HIV pandemic is considered a major push factor for nurses to leave specific countries because of the extreme demands of the work and fear of contracting HIV (Aiken, Buchan, Sochalski, Nichols, & Powell, 2004; Buchan & Sochalski, 2004). No one has directly linked the potential contribution of HIV stigma to nurses' intentions to migrate.
The authors' earlier work in developing and validating an instrument to measure HIV stigma for nurses, the HIV/AIDS Stigma Instrument—Nurse (HASI-N; Uys et al., 2009), brought to light the importance of stigma in nurses' lives. The items in the instrument were developed based on verbatim quotations from nurses and people living with HIV (PLWH), and two factors emerged. The first factor, Nurses Stigmatizing Patients, recognized the difficult fact that both patients and nurses report that nurses do in fact stigmatize patients. These scale items reflected the potential effect of nurses' personal feelings and beliefs about HIV and PLWH and included statements such as, “A nurse provided poorer quality care to an HIV/AIDS patient than to other patients.” The second factor, Nurses Being Stigmatized, recognized that nurses were themselves stigmatized by family, colleagues, and friends simply because of the care they provided for PLWH. This factor reflected community views of HIV stigma that were aimed directly at nurses, and included items such as, “People said that nurses who work in home care are HIV-positive.” In the validation of the HASI-N, the authors showed that nurses who reported lower stigma scores reported high job satisfaction (Uys et al., 2009). However, no one has linked HIV stigma experienced by nurses to their intent to migrate.
On the basis of this earlier research on HIV stigma and the lack of investigation into its relationship to migration, the purpose of this study was to explore whether HIV stigma perceived by nurses would contribute to explanations for intention to migrate, as evidenced by changes in odds/risk ratios, controlling for personal and job characteristics. Data were collected from nurses who worked with PLWH in Lesotho, Malawi, South Africa, Swaziland, and Tanzania. This study was part of a larger study in HIV-related stigma in these countries (Chirwa et al., 2009; Dlamini et al., 2007; Greeff et al., in press; Greeff & Phetlhu, 2007; Greeff, Phetlhu, et al., 2008; Greeff, Uys, et al., 2008; Holzemer & Uys, 2004; Holzemer, Uys, Chirwa, et al., 2007; Holzemer, Uys, Makoae, et al., 2007; Kohi et al., 2006; Makoae et al., 2008; Naidoo et al., 2007; Uys et al., 2005; Uys et al., 2009).
This study used a cross-sectional, descriptive, survey design. The questionnaires were self-administered by the participating nurses. All the measures were self-reported by participants.
Setting and Sample
Participants were recruited in 2005 through convenience sampling methods in each of the five countries. In Malawi, the participating nurses were from the city of Lilongwe. In South Africa, nurses were recruited in Potchefstroom, an urban area, and Kayakulu, a rural area. In Tanzania, the participating nurses came from the cities of Dar-es-Salaam and Mbeya. In the smaller countries of Lesotho and Swaziland, nurses were recruited from the entire country. The five principal investigators from different countries recruited potential participants mainly from hospital settings that provided HIV inpatient and clinic services and, in some cases, through health care centers, all of which specifically provided care to PLWH. Because of the nature of hospital-based nursing in these countries, the principal investigators first approached the nursing director/chief matron at each site to introduce the project and garner institutional support. This nurse then helped to distribute questionnaires to all nurses at the site. To the extent possible, the convenience sample of participants was balanced in terms of urban and rural settings. Each principal investigator sought to recruit 300 nurses per country, for a sample goal of 1,500 nurses.
Protection of Human Subjects
Permission to conduct the study was obtained from appropriate local and central government authorities. In addition, the protocol was approved by the institutional review boards or ethical committees of the seven collaborating universities. Local and community approvals were also sought as appropriate. The nurses were given information about the background of the study and were told that participation was voluntary and that they could withdraw from the study at any time. Because this was the first observation in a larger longitudinal study, participation was not anonymous, but the nurses were assured of the confidentiality of all information obtained because only subject codes were placed on the survey instruments. All participants signed a written consent form.
Each nurse completed a questionnaire in his or her own language. Languages included Sotho in Lesotho; Chichewa in Malawi; Tswana, English, and Zulu in South Africa; Swazi in Swaziland; and Kiswahili in Tanzania. Questionnaires were distributed to nurses at work, and in most cases they were completed on-site and returned to one of the members of the research team. The overall questionnaire included four sections relevant to the analysis here.
The first section of the questionnaire included demographic items. These included, for example, gender, age, marital status, education, and years working as a nurse.
Traynor and Wade's (1993) 38-item scale was used to measure five dimensions of job satisfaction. The scales and subscales were standardized so that scores ranged from 0 to 4, where high scores indicated satisfaction with particular aspects of the job. The five dimensions included Personal Satisfaction (10 items, Cronbach's alpha, .89), Workload (7 items, Cronbach's alpha, .85), Professional Support (9 items, Cronbach's alpha, .89), Pay and Prospects (8 items, Cronbach's alpha, .89), and Training (4 items, Cronbach's alpha, .84). The first factor, Personal Satisfaction, measured feelings of worthwhile accomplishment, contributions to patient care, the use of skills, amount of perceived challenge, quality of work, and the extent to which the job was seen as varied and interesting. The second factor, Workload, evaluated the time available to complete the work, staffing levels, and the amount of time spent on administration. Professional Support measured perceived support and respect from both managers/supervisors and colleagues, together with social aspects of work. Pay and Prospects focused on satisfaction with pay, clinical grading, and prospects for promotion. The fifth factor, Training, measured amount of time off provided for training, funding for training, opportunities to attend courses, and the extent to which training was considered important for the job.
Migration planning was measured by a series of brief questions. All respondents were asked the question, “Are you considering working outside of this country?” The response options were, “Yes, I am in the process of leaving” (n = 55); “Yes, I am considering it” (n = 452); “No, I have never seriously considered leaving” (n = 442), and “No, I am not interested” (n = 425). These responses were then transformed to create a dichotomous variable “Yes, considering migrating” (n = 507) and “No, not considering migrating” (n = 867).
The HASI-N (Uys et al., 2009) is a 19-item instrument composed of two factors, Nurses Stigmatizing Patients and Nurses Being Stigmatized, which was developed and validated in a sample of 1,474 nurses in five African counties. Respondents rate the frequency of events from 0 (never) to 3 (most of the time). Factor 1, Nurses Stigmatizing Patients, has 10 items and a Cronbach's alpha reliability estimate of .91. Sample items from this factor include, “A nurse provided poorer quality care to an HIV/AIDS patient than to other patients,” and, “A nurse shouted at or scolded an HIV/AIDS patient.” Factor 2, Nurses Being Stigmatized, has 9 items and a Cronbach's alpha reliability estimate of .90. Sample items from this factor include, “People said nurses who provide HIV/AIDS care are HIV-positive,” and, “Someone said that nurses who care for HIV/AIDS patients spread the disease.”
Data were collected from 1,374 nurses in five countries including Lesotho (n = 281), Malawi (n = 288), South Africa (n = 231), Swaziland (n = 288), and Tanzania (n = 286). Of those nurses, 37% reported that they were considering migrating (See Table 2). There were significant differences between the countries. About 58% of the nurses from Lesotho reported that they planned to leave the country, whereas only 18.6% of the nurses from South Africa reported intent to migrate. This reporting of country-level differences must be interpreted cautiously because the country-level samples were not randomly drawn from a population of nurses in each country.
The nurse sample was divided into those considering migration (n = 507, 36.9%) and those not considering migration (n = 867, 63.1%). F tests or Chi-square tests were used to compare the differences between these two groups on the demographic variables and job characteristic variables (See Table 3).
As shown in Table 3, nurses were primarily women (78.8%) and in their late 30s (M = 37.6, SD = 9.49 years), not married (78.7%), and with a diploma education (45.3%). The majority of the nurses worked in hospitals (63%), which were primarily located in urban settings (89.2%). The average number of years worked as a nurse was 11.62 years (SD = 9.35, range: 1-43), and 27.7% of the nurses reported that they worked on an identified HIV-care unit.
There were no differences in years as a nurse, gender, or working on an HIV unit between the groups considering and not considering migration. The mean age of the nurses who were considering migrating (36.31 years) was significantly younger than the mean age of the nurses who were not considering migrating (37.84 years). A higher proportion of unmarried nurses (81.7%) were considering migrating. Almost 43% of the nurses with degrees were considering migrating, and only 17.7% of the nurses with certificate/diploma-level education were considering migrating. Those who were considering migrating reported significantly higher scores on the factor of Nurses Stigmatizing Patients (M = .41) than nurses not considering migrating (M = .33). There were no differences between the groups on the factor of Nurses Being Stigmatized. Two job characteristics (site and working on HIV units) were not significantly related to intent to migrate. A significantly higher proportion of nurses working in an urban area (91.6%) were considering migrating than those in the rural areas (8.4%). The nurses considering migration reported significantly lower scores on all five dimensions of the job satisfaction scale than those not considering migration.
A stepwise logistic regression analysis was used to identify a final set of predictors of intent to migrate (See Table 4). Five personal characteristics (age, gender, live in an urban area, years as a nurse, and education), two stigma variables (Nurses Stigmatizing Patients and Nurses Being Stigmatized), and three job characteristic variables (work on dedicated HIV-care unit, hospital site, and total job satisfaction) were entered stepwise as predictors, and the final model was significant (χ2 = 25.23, degrees of freedom = 15, p < .047). Odds ratios (ORs) of greater than 1 indicate that the odds are greater that the respondent would consider working outside of the country when the condition is present, when the other factors in the model remained equal. The highest ORs were, in descending order: living in an urban area (OR = 2.115), working in a hospital rather than a clinic (OR = 1.461), and having witnessed nurses stigmatizing patients (OR = 1.224).
Conventional migration literature supports the contention that individuals migrate primarily to exploit wage differentials. When the costs of migration are offset by expected higher wages, migration occurs. Although this result may be true for nurses, the authors' data suggest that other factors are also operating in the nurses' intent to migrate in these five African countries. Buchan and Sochalski's (2004) work suggested that nurses seek opportunities for improved professional development as a main reason for migration and not necessarily higher wages, but these data do not support that observation. Neither pay nor training increased the odds of nurses in this study reporting that they intended to migrate out of their home countries in the subscale analysis of the job satisfaction scale.
This study presents a somewhat different profile of the nurses who intended to migrate from these five African countries. The odds of intent to migrate in this sample were greater for more experienced nurses working in urban hospitals. These results present a challenge for underfunded human resource planners to develop retention strategies that focus on experienced nurses who work primarily in urban hospitals. These nurses represent a key group for directed retention strategies because they are the most clinically experienced and have greater opportunities to migrate.
Further, the results show that nurses who witness other nurses stigmatizing patients because of HIV status also reported a higher OR of intent to migrate. In the authors' bivariate analyses, all five measures of job satisfaction were negatively related to intent to migrate. However, in logistic regression modeling, holding other variables constant, these bivariate relationships did not significantly increase the odds of the nurses reporting that they intended to migrate out of their countries.
When one examines retention strategies proposed in the literature (i.e., Medecins San Frontieres, 2007), the focus is often on increasing salaries, upgrading living and working conditions, improving management, and enhancing career advancement. Although such strategies are important options for Ministries of Health to undertake, it is doubtful that these strategies are sufficient to stem the stream of outmigration. In light of country-level health and workforce indices, the “bigger picture” of the country's economic future looks unpromising to people who are experienced and educated and who could make lives for themselves and their families elsewhere. Increasing the upper-level salary ranges for experienced nurses seems to be an important retention strategy that is rarely used.
It is challenging to make recommendations to stem the flow of nurses out of these five countries. If outmigration is linked to the level of social and economic development of the country rather than health sector work factors only, one might have to accept outmigration as a reality for the foreseeable future. This might then demand a significant increase in the training of nurses and the development of strategies that attempt to retrieve economic resources spent on such training from professionals who leave the system. For instance, countries might have to retain trained government-sponsored nurses by binding them to an employment contract for at least the number of years they were trained.
One of the strategies suggested by results from this study is to increase opportunities for further education about HIV-related stigma and its effect on nurses and their job satisfaction. We are only beginning to develop an understanding of the negative effect of HIV stigma on nurses. Another strategy that emerged from these data was to consider supporting increased training of more experienced nurses who are already embedded in their communities and who would find it more difficult to leave. It is a massive challenge to reinforce a threadbare nursing workforce in countries with high HIV prevalence and poor economic conditions. These data provide the first hint that HIV stigma by association may increase the odds of a nurse's intent to migrate, a result that requires further research.
The aim of this study was to examine whether a relationship exists between HIV stigma perceived by nurses and their intent to migrate. The results presented here contribute a new perspective regarding patterns of nurse migration and the relationship to HIV and stigma. The results have been used to inform programmatic work to reduce HIV stigma, as perceived by both nurses and PLWH in health care settings (Uys et al., in press).
- A substantial number of nurses in Lesotho, Malawi, South Africa, Swaziland, and Tanzania were considering migrating to other countries, which could have a significant effect on clinical care for PLWH.
- The nurses who were most likely to be planning to migrate were those living in urban areas, working in hospitals, and having witnessed nurses stigmatizing PLWH.
- Health care delivery sites in these countries should consider strategies to improve retention of nurses who provide the majority of clinical care for PLWH.
This work was supported by NIH Research Grant #R01 TW06395 funded by the Fogarty International Center, the National Institute of Mental Health, and the Health Resources and Services Administration, U.S. Government.
Aiken, L.H., Buchan, J., Sochalski, J., Nichols, B., & Powell, M. (2004). Trends in international nurse migration
. Health Affairs, 23,
Aiken, L.H., Clarke, S.P., Sloane, D.M., Sochalski, J.A., Busse, R., Clarke, & H., et al. (2001). Nurses
' reports on hospital care in five countries. Health Affairs, 20,
Anderson, B.A., & Isaacs, A.A. (2007). Simply not there: The impact of international migration
and midwives—Perspectives from Guyana. Journal of Midwifery & Women's Health, 52,
Buchan, J., & Sochalski, J. (2004). The migration
: Trends and policies. Bulletin of the World Health Organization, 82,
Chirwa, M.L., Greeff, M., Kohi, T.W., Naidoo, J.R., Makoae, L.N., Dlamini, & P.S., et al. (2009). HIV stigma
and nurse job satisfaction in five African countries. Journal of the Association of Nurses in AIDS Care, 20,
Dlamini, P.S., Kohi, T.W., Uys, L.R., Phetlhu, R.D., Chirwa, M.L., Naidoo, & J.R., et al. (2007). Verbal and physical abuse and neglect as manifestations of HIV
in five African countries. Public Health Nursing, 24,
El-Jardali, F., Dimassi, H., Dumit, N., Jamal, D., & Mouro, G. (2009). A national cross-sectional study on nurses
' intent to leave and job satisfaction in Lebanon: Implications for policy and practice. BMC Nursing, 8,
Ferrand, E., Lemaire, F., Regnier, B., Kuteifan, K., Badet, M., Asfar, & P., et al. (2003). Discrepancies between perceptions by physicians and nursing staff of intensive care unit end-of-life decisions. American Journal of Respiratory and Critical Care Medicine, 167,
Greeff, M., & Phetlhu, R. (2007). The meaning and effect of HIV
for people living with AIDS
involved in their care in the North West Province, South Africa
. Curationis, 30,
Greeff, M., Phetlhu, R., Makoae, L.N., Dlamini, P.S., Holzemer, W.L., Naidoo, & J.R., et al. (2008). Disclosure of HIV
status: Experiences and perceptions of persons living with HIV
involved in their care in Africa
. Qualitative Health Research, 8,
Greeff, M., Phetlhu, D.R., Makoae, L.N., Dlamini, P.S., Holzemer, W.L., Naidoo, & J., et al. (in press). Disclosure of HIV
status: Experiences and perceptions of persons living with HIV
involved in their care in five Africa
countries. Qualitative Health Research
Greeff, M., Uys, L.R., Holzemer, W.L., Makoae, L.N., Dlamini, P.S., Kohi, & T.W., et al. (2008). Experiences of HIV
of PLWA and nurses
from five African countries. African Journal of AIDS Research, 10,
Hagopian, A. (2007). Recruiting primary care physicians from abroad: Is poaching from low-income countries morally defensible? Annals of Family Medicine, 5,
Holzemer, W.L., & Uys, L.R. (2004). Managing AIDS stigma
. Journal of Social Aspects of HIV/AIDS, 1,
Holzemer, W.L., Uys, L.R., Chirwa, M.L., Greeff, M., Makoae, L.N., Kohi, & T.W., et al. (2007). Validation of the HIV
Instrument - PLWA (HASI-P). AIDS Care, 19,
Holzemer, W.L., Uys, L., Makoae, L., Stewart, A., Phetlhu, R., Dlamini, & P.S., et al. (2007). A conceptual model of HIV
from five African countries. Journal of Advanced Nursing, 58,
Kingma, M. (2006). Nurses on the move: Migration and the global health care economy.
Ithaca, NY: Cornell University Press.
Kohi, T., Makoae, L., Chirwa, M., Holzemer, W.L., Phetlhu, D.R., Uys, & L., et al. (2006). HIV
and AIDS stigma
violates human rights in five African countries. Nursing Ethics, 13,
Larrabee, J.H., Janney, M.A., Ostrow, C.L., Withrow, M.L., Hobbs, G.R., Jr., & Burant, C. (2003). Predicting registered nurse job satisfaction and intent to leave. Journal of Nursing Administration, 33,
Locke, E.A. (1976). The nature and causes of job satisfaction. In: M.D. Dunnette (Ed.), Handbook of industrial and organizational psychology
(pp. 1325-1334). Chicago: Rand McNally.
Makoae, L.N., Greeff, M., Phetlhu, R.D., Uys, L.R., Naidoo, J.R., Kohi, & T.W., et al. (2008). Coping with HIV
in five African countries. Journal of the Association of Nurses in AIDS Care, 19,
Medecins Sans Frontieres (2007). HELP WANTED: Confronting the health care worker crisis to expand access to HIV/AIDS treatment: MSF experience in southern Africa.
Johannesburg, South Africa
Naidoo, J.R., Uys, L.R., Greeff, M., Holzemer, W.L., Makoae, L., Dlamini, & P., et al. (2007). Urban and rural differences in HIV
in five African countries. African Journal of AIDS Research, 6,
Nguyen, L., Ropers, S., Nderitu, E., Zuyderduin, A., Luboga, S., & Hagopian, A. (2008). Intent to migrate among nursing students in Uganda: Measures of the brain drain in the next generation of health professionals. Human Resources for Health, 6,
Schmalenberg, C., Kramer, M., King, C.R., Krugman, M., Lund, C., Poduska, & D., et al. (2005). Excellence through evidence: Securing collegial/collaborative nurse-physician relationships, part 2. Journal of Nursing Administration, 35,
Stamps, P.L. (1986). Nurses and work satisfaction: An index for measurement.
Ann Arbor, MI: Health Administration Press.
Traynor, M., & Wade, B. (1993). The development of a measure of job satisfaction for use in monitoring the morale of community nurses
in four trusts. Journal of Advanced Nursing, 18,
United Nations Development Programme. (2008). Statistics of the Human Development Report. Retrieved May 14, 2008 from http://hdr.undp.org/en/statistics/
Uys, L.R., Chirwa, M., Dlamini, P., Greeff, M., Kohi, T., Holzemer, & W.L., et al. (2005). Eating plastic, winning the lotto, joining the WWW: Descriptions of HIV
. Journal of the Association of Nurses in AIDS Care, 16,
Uys, L.R., Chirwa, M.L., Kohi, T.W., Greeff, M., Naidoo, J.R., Makoae, & L.N., et al. (in press). Evaluation of a health setting-based HIV stigma
reduction intervention in five African countries. AIDS
Patient Care and STDs.
Uys, L.R., Holzemer, W.L., Chirwa, M.L., Dlamini, P.S., Greeff, M., Kohi, & T.W., et al. (2009). The development and validation of the HIV
Instrument - Nurse (HASI-N). AIDS Care, 21,
Uys, L.R., Minnaar, A., Reid, S., & Naidoo, J.R. (2004). The perceptions of nurses
in a district health system in KwaZulu-Natal of their supervision, self-esteem and job satisfaction. Curationis, 27,
Wells, N., Roberts, L., & Medlin, L.C. (2002). Issues related to staff retention and turnover. Seminars for Nurse Managers, 10,
World Health Organization (2006). The World Health Report 2006: Working together for health.
Geneva, Switzerland: Author.
World Health Organization. World Health Organization Statistical Information System (WHOSIS), Retrieved May 14, 2008 from http://www.who.int/whosis/en/index.html