See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/236636183
Factors that influence midwifery students in Ghana when deciding where to
practice: A discrete choice experiment
Article  in  BMC Medical Education · May 2013
DOI: 10.1186/1472-6920-13-64 · Source: PubMed
CITATIONS READS
16 118
5 authors, including:
Peter Ageyi-Baffour Sarah Rominski
University of Michigan University of Michigan
1 PUBLICATION   16 CITATIONS    58 PUBLICATIONS   393 CITATIONS   
SEE PROFILE SEE PROFILE
Emmanuel Kweku Nakua Mawuli Gyakobo
Kwame Nkrumah University Of Science and Technology Tetteh Quarshie Memorial Hospital, Mampong-Akuapem
39 PUBLICATIONS   291 CITATIONS    13 PUBLICATIONS   271 CITATIONS   
SEE PROFILE SEE PROFILE
Some of the authors of this publication are also working on these related projects:
Alpha thalassaemia and the severity of Plasmodium falciparum infection in microcytic children from the Ashanti region of Ghana View project
Ghana-Michigan Study View project
All content following this page was uploaded by Sarah Rominski on 13 January 2014.
The user has requested enhancement of the downloaded file.
Ageyi-Baffour et al. BMC Medical Education 2013, 13:64
http://www.biomedcentral.com/1472-6920/13/64
RESEARCH ARTICLE Open Access
Factors that influence midwifery students in
Ghana when deciding where to practice: a
discrete choice experiment
Peter Ageyi-Baffour1, Sarah Rominski2*, Emmanuel Nakua1, Mawuli Gyakobo3 and Jody R Lori4
Abstract
Background: Mal-distribution of the health workforce with a strong bias for urban living is a major constraint to
expanding midwifery services in Ghana. According to the UN Millennium Development Goals (MDG) report, the
high risk of dying in pregnancy or childbirth continues in Africa. Maternal death is currently estimated at 350 per
100,000, partially a reflection of the low rates of professional support during birth. Many women in rural areas of
Ghana give birth alone or with a non-skilled attendant. Midwives are key healthcare providers in achieving the
MDGs, specifically in reducing maternal mortality by three-quarters and reducing by two-thirds the under 5 child
mortality rate by 2015.
Methods: This quantitative research study used a computerized structured survey containing a discrete choice
experiment (DCE) to quantify the importance of different incentives and policies to encourage service to deprived,
rural and remote areas by upper-year midwifery students following graduation. Using a hierarchical Bayes
procedure we estimated individual and mean utility parameters for two hundred and ninety eight third year
midwifery students from two of the largest midwifery training schools in Ghana.
Results: Midwifery students in our sample identified: 1) study leave after two years of rural service; 2) an advanced
work environment with reliable electricity, appropriate technology and a constant drug supply; and 3) superior
housing (2 bedroom, 1 bathroom, kitchen, living room, not shared) as the top three motivating factors to accept a
rural posting.
Conclusion: Addressing the motivating factors for rural postings among midwifery students who are about to
graduate and enter the workforce could significantly contribute to the current mal-distribution of the health
workforce.
Background Ghana has extreme need in areas related to maternal
A health worker with midwifery skills should be present health. Maternal death is currently estimated at 350 per
at every birth according to the joint statement by WHO/ 100,000, in part a reflection of the low rates of profes-
UNFPA/UNICEF/World Bank [1]. The UN’s Millennium sional or skilled support during childbirth [4]. While
Development Goal (MDG) 5, set in 2000, targeted a 75% there is high uptake of prenatal care, estimated as high
reduction in the maternal mortality ratio by 2015. Great as 94 percent, many women in Ghana give birth alone
strides have been made toward reaching MDG5 [2]. or with a non-skilled attendant [5,6]. Ghana’s health
Despite significant investments in resources and targeted worker density, estimated at 91 per 1,000 population,
interventions and a 56% reduction of maternal mortality falls far below the WHO recommended level of 2.28
in sub-Saharan Africa between 1990 and 2010, progress health care professionals per 1,000 population [7].
towards MDG5 has slowed in recent years and this Health worker density is negatively associated with
target will most likely not be met in many countries [3]. maternal mortality and potentially child mortality [8]. If
access to skilled a birth attendant improves, women’s
* Correspondence: sarahrom@umich.edu
2 lives could be saved and morbidity drastically reduced.Global REACH, University of Michigan Medical School, Ann Arbor, MI, USA
Full list of author information is available at the end of the article
© 2013 Ageyi-Baffour et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the
Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use,
distribution, and reproduction in any medium, provided the original work is properly cited.
Ageyi-Baffour et al. BMC Medical Education 2013, 13:64 Page 2 of 6
http://www.biomedcentral.com/1472-6920/13/64
In rural areas of Ghana, poor road conditions, long hypothetical job posting scenarios each characterized by
distances to a health facility, and lack of transportation a variety of attributes (e.g., salary, housing, offer of a car
make accessing health care services a challenge. Contrib- allowance). Respondents are then asked to select their
uting to the lack of access, many health workers are preferred scenario. The benefit of this model is that of
unwilling to locate to rural areas where the majority of opportunity cost. We know individuals want the best of
Ghanaians live [9]. The lack of access to a skilled birth everything, but when resources are limited, the DCE
attendant adds to the non-uniform distribution of ma- gives a weighted relevance to distinguish which attri-
ternal mortality in Ghana. According to the State of the butes are the most highly incentivizing [14] to motivate
World’s Midwives (SOWM) report [4], Ghana has individuals to locate to rural areas. Well established in
reduced maternal mortality by 44% since 1990, but the the use of inferring patients’ preferences, this technique
rate remains high, in large part because of limits to has recently been used to test providers’ preferences as
access in care driven by the lack of adequate numbers of well [15].
midwives. In designing the DCE, we selected attributes (motiva-
Identified as a key priority for the Ghana Ministry of tors) to increase attraction to rural practice based on con-
Health, the SOWM [4] report highlights the need to versations with the Ghana Ministry of Health and eight
recruit and retain midwives especially in the rural focus groups with final-year midwifery students (n = 49) at
northern portion of the country. Differences between the two largest Ghanaian midwifery training colleges. The
urban and rural areas are striking. While less than 20% seven attributes identified by content analysis [16]
of urban births were attended by untrained personnel, included: salary, study leave, housing, supportive manage-
over 60% of births in rural areas were attended by ment, infrastructure, transportation and children’s educa-
someone with no formal training. Recognizing this tion. The DCE was then designed to estimate the relative
problem, in 2005, the Ministry of Health targeted im- value or utility of different work conditions that might
proving basic obstetrical care training for midlevel pro- incentivize students to locate to rural areas to practice
viders [10]. As part of this commitment, the Ministry of after graduation. The survey consisted of demographic
Health accredited and opened 14 new midwifery train- and background questions followed by a series of 11
ing colleges. discrete choice questions. In these questions, students
Various strategies, including educational interventions were asked to compare two hypothetical job postings
such as selecting students with rural background, in- (Figure 1).
creasing financial incentives, and professional develop- Participants were asked to imagine that upon comple-
ment incentives, have been implemented in an attempt tion of their midwifery training, they were offered two
to recruit and retain health professionals to rural and postings in two rural deprived areas by the Ministry of
remote areas [11-13]. These strategies have been poorly Health. Deprived area was defined as an area that is dis-
evaluated and largely unsuccessful at reversing within- tant from a big city with few social amenities such as
country maldistribution of the health workforce. Under- schools, roads, or pipeborne water. Participants were
standing the preferences of future midwives for posting asked to imagine themselves making a real decision
to rural, deprived areas is key to achieving the MDGs, between two rural postings and were asked of the two
specifically in reducing maternal mortality by three- offered, which they felt was better. Further, students
quarters and reducing by two-thirds the under 5 child were asked to answer whether or not they would accept
mortality rate by 2015. this posting if it were offered.
The purpose of this study is to further our under-
standing of the factors that motivate professional mid- Setting and sample
wifery students to move to rural areas with the final goal Midwifery education in Ghana is a three-year, post-
of assisting the government in Ghana to develop an secondary school diploma program. Since 2003, fourteen
incentive package to improve health worker distribution. midwifery training schools in Ghana have been
accredited. Of the ten regions in Ghana, each is home to
Methods at least one midwifery education program. There is a na-
Used for many years in market research, one tool with tional curriculum with the first three semesters focused
potential to suggest working condition priorities is on general nursing and the final three semesters devoted
conjoint, or trade-off, analysis. This constellation of to midwifery knowledge and skills. Students spend at
techniques provides the researcher with the ability to least one clinical rotation at a rural district hospital.
elicit individuals’ stated preferences. A common form of We chose third-year midwifery students (n = 238)
conjoint analysis is the discrete choice experiment about to graduate and considering employment perspec-
(DCE), in which respondents (e.g. midwifery students) tives for our sample. We used purposive sampling to
are presented with a choice of several competing obtain a wide diversity of experiences and opinions.
Ageyi-Baffour et al. BMC Medical Education 2013, 13:64 Page 3 of 6
http://www.biomedcentral.com/1472-6920/13/64
Figure 1 Sample DCE question.
Students at two of the largest midwifery training schools incentive packages. The software calculates total utilities of
in Ghana were invited to participate in the study. These the simulated options for each respondent by summing
two schools combined graduate the largest number of attribute utilities. The respondents were repeatedly sam-
midwifery students per year. This survey was part of a pled to stabilize these preferences. In addition, we added a
larger collaboration between the University of Michigan, random error term to the estimates of utilities to correct
the Kwame Nkrumah University of Science and Tech- for any similarities in scenarios. We used Sawtooth’s
nology and the Ghana Ministry of Health. The research Choice-Based Conjoint with Hierarchical Bayes statistical
was approved by the Ghana Health Service Ethical program to estimate coefficients for the individual utilities
Review Committee, the Kwame Nkrumah University of of each attribute level.
Science and Technology Committee on Human Re-
search, Publications and Ethics, the University of Ghana Results
Medical School, and the University of Michigan Ethical Two-hundred and thirty eight upper level midwifery
Review Board. students completed our survey for a response rate of
79.8%. See Table 1 for select demographic data.
Data collection While less than 5% of our respondents currently had
Informed consent was obtained prior to participation in children, nearly all (99.2%) reported they plan to have chil-
the DCE. Each computerized survey took approximately dren in the future. The vast majority of the sample
30–45 minutes. Students were given an incentive of 10 (98.7%) attended public high school with only 1.3% (n = 3)
Ghana Cedis (approximately 7 US dollars) upon comple- reporting they attended private school prior to midwifery
tion of the survey. Students signed in and the names training. Almost half (47.1%) would like to be practicing
were compared to a class list generated by the head of general midwifery in 10 years with the remainder
each college to determine response rate. reporting they preferred a position in administration
(21.8%), public health (13.9%), teaching (7.1%) or nursing
Data analysis (1.7%). One hundred and seventy seven (74.4%) respon-
Sawtooth Software (Orem, UT) was used to construct, field dents reported they believe midwives are “very valued”
and score the surveys. Using market simulator software in in their society, with another 23.5% reporting that mid-
Sawtooth’s Choice-Based Conjoint with Hierarchical Bayes wives are “somewhat valued”. Only 2.1% reported that
module, we used individual-level utilities to estimate the midwives are either somewhat or very unvalued. The
proportion of respondents who would prefer specific vast majority of our sample (99.3%) reported they
Ageyi-Baffour et al. BMC Medical Education 2013, 13:64 Page 4 of 6
http://www.biomedcentral.com/1472-6920/13/64
Table 1 Select demographic data Table 2 DCE results
Characteristics (n = 238) Mean or No (%) Parameter 95% CI
Age 24.5 years Salary1 Mean 0.72 0.60 0.84
Range (18–33 years) SD 0.99
Gender Allowance for Children's Mean −0.93 −1.03 −0.83
education2
Female 238 (100%)
SD 0.83
Male 0 (0%)
Basic Infrastructure, equipment, Mean −1.07 −1.20 −0.93
Marital Status and supplies3
Married 8 (3.4%) SD 1.11
Living with a partner 4 (1.7%) Management are supportive4 Mean 0.96 0.85 1.07
In a relationship, not living together 135 (56.7%) SD 0.93
Not in a relationship 89 (37.4%) 2 years before study leave5 Mean 1.24 1.02 1.45
Rather not say 2 (0.8%) SD 1.78
Number of Children No Housing6 Mean −1.88 −2.05 −1.71
None 227 (95.4%) SD 1.39
One 1 (0.4%) Superior Housing6 Mean −0.87 −0.99 −0.75
Two 1 (0.4%) SD 1.02
Three 2 (0.8%) No access to car7 Mean −0.88 −0.98 −0.77
Ever lived in a rural area SD 0.89
Yes 91 (38.2%) 1 continuous variable; coefficient represents the magnitude of increase in
No 146 (61.3%) utility for every 10% increase in salary.
2 compared to no allowance for children’s education.
Don’t know 1 (0.4%) 3 compared to improved infrastructure, equipment, and supplies.
4 compared to unsupportive workplace and management.
Birth Location 5 compared to five years minimum work before study leave.
Urban 168 (70.6%) 6 compared to basic housing.
7 compared to access to a utility car.
Periurban 42 (17.6%)
Rural 19 (8%) incentives structure in place had a large negative utility
Don’t know 7 (2.9%) at −3.67.
Rather not say .2 (0.8%)
Limitations
As only two schools, both located in major urban areas,
would like to return to university in the future to were sampled, it is not clear to what extent the findings
pursue a university degree. can be extrapolated to all midwifery students in Ghana.
However, it is plausible that those students studying in
Discrete choice experiment Accra and Kumasi would be more reluctant to locate to
The full results of the DCE can be seen in Table 2. For rural areas, as they are accustomed to living and study-
each of the parameter estimates, the currently offered ing in well-appointed areas.
level was used as the reference.
In our sample, the highest parameter estimates for Discussion
rural placement by third year midwifery students were There is a great need to improve the provision of mater-
study leave after two years versus after five years, and nal and child health in rural Ghana. One important
having an “advanced” work environment (reliable electri- cadre of worker that can deliver this necessary care is
city, ultrasound, constant drug supply) with utilities of midwives. Finding incentives to motivate these midwives
1.24 and 1.07. Superior housing, defined as a two bed- to locate to rural areas is of the utmost importance.
room, one bath, not shared house) also had a relatively We explored the factors and their strengths that influ-
high parameter estimate of 0.87. In contrast, providing ence reported acceptance to rural postings among
no housing to midwifery students in rural area negatively graduating midwives. The top three factors identified in-
influenced their decision to practice in underserved clude: 1) study leave after two years versus five years of
areas. working in rural areas; 2) advanced working conditions
Increasing salary had a utility of 0.72 and access to a such as electricity, regular drug supplies, and equipment,
vehicle hire program had a utility of 0.88. Having no and; 3) a free superior housing scheme. These results
Ageyi-Baffour et al. BMC Medical Education 2013, 13:64 Page 5 of 6
http://www.biomedcentral.com/1472-6920/13/64
corroborate those found by Kruk and colleagues [15] other low-income countries, points to the importance of
among Ghanaian medical students and Kwansah and carrying out these kinds of policy experiments. The
colleagues [17] among Ghanaian nursing students. simulation presented in this paper can offer a starting
Considering Ghana has recently expanded midwifery point to create a series of packages for experimentation.
training by starting the first bachelor’s degree in midwif-
ery, educational incentives offer a promising direction Conclusion
for future interventions. However, study leave after two Maternal death remains high and one of the top public
years of service will ultimately impact the training health concerns in Ghana. With barely three years to go
schools. The needs of these schools must to be taken for the UN MDGs, there is a huge need for uncompromis-
into account when thinking through potential interven- ing effort to improve access to care in rural areas. If the
tions, as many schools in Ghana are currently at or past global strides made in reduction of maternal deaths are to
capacity. be sustained [21] and the targeted 75% reduction in the
Providing working conditions that offer a full range of maternal mortality ratio by 2015 achieved [1], greater
amenities to health workers is a priority of the govern- access to skilled care in rural Ghana is needed.
ment of Ghana. In qualitative work with a subset of mid- Findings from our study conclude the three most im-
wifery students, the need for professional, as well as portant factors graduating midwives consider in accepting
personal, support was noted [16]. This level of on-going postings to rural underserved areas include study leave
support requires intense planning and deployment after two years versus five years of working in rural areas,
efforts by governmental agencies to realize. advanced working condition and free superior housing
The strong desire identified by our sample for free scheme. Without addressing these needs to correct the
superior housing and the negative utility of no housing mal-distribution of skilled health staff in urban and rural
suggests these students are aware of poor housing options areas, Ghana is unlikely to realize the goals of MDG 5.
in rural areas and consider free superior housing a re-
quirement for accepting a rural posting. This finding is Competing interests
The authors declare that they have no competing interests.
consistent with previous work in Ghana and Ethiopia
which shows other cadres of students are also motivated Authors’ contributions
by the availability of housing [15,18-20]. Providing a PAB was involved in all aspects of the study including conceptualization of
free housing scheme requires cooperation between the the study, development of the study instruments, data collection, data
analysis, drafting and revising of the manuscript. SR was involved in all
health ministry, local government, and economic plan- aspects of the study including conceptualization of the study, development
ning ministries. Thus the need for intersectoral collabo- of the study instruments, data collection, data analysis, drafting and revising
rations in the recruitment and retention of health staff of the manuscript. EN was involved in data collection, data analysis and
drafting and revising of the manuscript. MG was involved in data collection
in rural areas becomes eminent. and drafting and revising of the manuscript. JL was involved in all aspects of
Interestingly, increased salary was not as important to the study including conceptualization of the study, development of the
these students as expected. Increased salary lagged be- study instruments, data collection, data analysis, drafting and revising of the
manuscript. All authors read and approved the final manuscript.
hind educational, professional and housing interventions
in our sample. Work in Ethiopia has suggested that for Author details
1
physicians, large salary increases are needed to motivate School of Medical Sciences, Kwame Nkrumah University of Science and
Technology, Kumasi, Ghana. 2Global REACH, University of Michigan Medical
work in deprived areas [18]. It does not appear from School, Ann Arbor, MI, USA. 3Medical School, University of Ghana, Accra,
these results that increasing salary will be as motivating Ghana. 4University of Michigan School of Nursing, Ann Arbor, MI, USA.
for young midwives as improving the clinic amenities,
Received: 18 June 2012 Accepted: 23 April 2013
offering superior housing, or enabling a return to school. Published: 4 May 2013
Allowances for children’s education were among the
least motivating incentives in the policy packages. References
1. WHO/UNFPA/UNICEF/World Bank: Reducing maternal mortality. A joint
Considering the majority of participants in this study statement by WHO/UNFPA/UNICEF/World Bank. Geneva: World Health
planned on having children in the future, this is some- Organization; 1999.
what surprising. Perhaps, as Kruk and colleagues [15] 2. Rosenfield A, Maine D, Freedman L: Meeting MDG-5: an impossible dream.
Maternal Survival: The Lancet; 2006. September 5-6.
found, given the young age of the participants and their 3. WHO: Trends in Maternal Mortality: 1990–2010. 2012. Available at: http://
plans to have children in the future, the well-being of whqlibdoc.who.int/publications/2010/9789241500265_eng.pdf Accessed
these future children does not factor as heavily as the May 25, 2012.
4. UNFPA: State of the World’s Midwifery.; 2011. Available at: http://www.unfpa.
participants’ own well-being. org/sowmy/report/home.html Accessed May 9, 2012.
This research points to future research including a 5. Ghana Statistical Service (GSS), Ghana Health Service (GHS), and ICF Macro:
policy experiment testing the packages that were most Ghana Demographic and Health Survey [DHS], 2008. Accra, Ghana: GSS, GHS,
and ICF Macro; 2009.
attractive to these participants. The enormous shortage 6. Adanu RMK: Utilization of obstetric services in Ghana between 1999 and
of health care workers practicing in rural Ghana, and 2003. Afr J Reprod Health 2010, 14(3):153–158.
Ageyi-Baffour et al. BMC Medical Education 2013, 13:64 Page 6 of 6
http://www.biomedcentral.com/1472-6920/13/64
7. Scheffler RM, Liu JX, Kinfu Y, Dal Poz MR: Forecasting the global shortage
of physicians: an economic- and needs-based approach. Bull World Health
Organ 2008, 86(7):516–523B.
8. Speybroeck N, Kinfu Y, Dal Poz MR, Evans DB: Reassessing the relationship
between human resources for health, intervention coverage and health
outcomes. Geneva: World Health Organization; 2006.
9. Ghana Ministry of Health: Reproductive health stratgetic plan, 2006–2010.
2005.
10. Prosser M, Sonneveldt E, Hamilton M, Menotti E, Davis P: United States Aid
for International Development. (USAID). (2006). In The emerging midwifery
crisis in Ghana: mapping of midwives and service availability highlights gaps
in maternal care. 2006.
11. Dolea C, Stormont L, Braichet J: Evaluated strategies to increase attraction
and retention of health workers in remote and rural areas. Bull World
Health Organ 2010, 88:379–385.
12. Globler L, Marais BJ, Mabunda SA, Marindi PN, Reuter H, Volmink J:
Interventions for increasing the proportion of health professionals
practicing in rural and other underserved areas. Cochrane Database Syst
Rev 2009, Issue 1:Art. No.: CD005314. doi:10.1002/14651858.CD005314.pub2.
13. Lehmann U, Dieleman M, Martineau T: Staffing remote rural areas in
middle- and low-income countries: a literature review of attraction and
retention. BMC Health Serv Res 2008, 8:19.
14. Orme B: Getting started with conjoint analysis. Madison, WI: Research
Publishers LCC; 2006.
15. Kruk M, Johnson J, Gyakobo M, Agyei-Baffour P, Asabir K, Kotha S, Kwansah
J, Nakua E, Snow R, Dzodzomenyo M: Rural practice preferences among
medical students in Ghana: a discrete choice experiment. Bull World
Health Organ 2010, 88:333–341.
16. Lori JR, Rominski SD, Gyakobo M, Muriu EW, Kweku NE, Agyei-Baffour P:
Perceived barriers and motivating factors influencing student midwives'
acceptance of rural postings in Ghana an exploratory qualitative study.
Hum Resour Health 2012, 10:17.
17. Kwansah J, Dzodzomenyo M, Mutumba M, Asabir K, Koomson E, Gyakobo
M, Kruk M, Snow R: Policy talk: incentives for rural service among nurses
in Ghana. Health Policy Plan 2012. doi:10.1093/heapol/czs016.
18. Hanson K, Jack W: Health worker preferences for job attributes in Ethiopia:
Results from a discrete choice experiment (working paper). Washington, DC:
Georgetown University; 2008.
19. Mensah K: Attracting and retaining health staff: a critical analysis of the
factors influencing the retention of health workers in deprived/hardship areas.
Accra: Yak-Aky Services; 2002.
20. Snow R, Asabir K, Mutumba M, Koomson E, Gyan K, Dzodzomenyo M, Kruk
M, Kwasnah J: Key factors leading to reduced recruitment and retention
of health professionals in remote areas of Ghana: a qualitative study and
policy recommendations. Hum Resour Health 2011, 10:138.
21. WHO: Millennium Development Goals: progress towards the health-related
Millennium Development Goals. 2011. Available at: http://www.who.int/
mediacentre/factsheets/fs290/en/index.html Accessed May 9, 2012.
doi:10.1186/1472-6920-13-64
Cite this article as: Ageyi-Baffour et al.: Factors that influence midwifery
students in Ghana when deciding where to practice: a discrete choice
experiment. BMC Medical Education 2013 13:64.
Submit your next manuscript to BioMed Central
and take full advantage of: 
• Convenient online submission
• Thorough peer review
• No space constraints or color figure charges
• Immediate publication on acceptance
• Inclusion in PubMed, CAS, Scopus and Google Scholar
• Research which is freely available for redistribution
Submit your manuscript at 
www.biomedcentral.com/submit
View publication stats