Social Science & Medicine 132 (2015) 30e37

Contents lists available at ScienceDirect

Social Science & Medicine

journal homepage: www.elsevier.com/locate/socscimed

Providing primary health care through integrated microfinance and

health services in Latin America
Kimberley H. Geissler a, b, *, Sheila Leatherman b
a
Department of Markets, Public Policy and Law, Boston University School of Management, 595 Commonwealth Avenue, Boston, MA 02215, USA
b
Department of Health Policy and Management, Gillings School of Global Public Health, University of North Carolina, 1101 McGavran-Greenberg Hall, CB
#7411, Chapel Hill, NC 27599, USA

a r t i c l e i n f o a b s t r a c t

Article history: The simultaneous burdens of communicable and chronic non-communicable diseases cause significant
Available online 10 March 2015 morbidity and mortality in middle-income countries. The poor are at particular risk, with lower access to
health care and higher rates of avoidable mortality. Integrating health-related services with microfinance
Keywords: has been shown to improve health knowledge, behaviors, and access to appropriate health care. How-
Latin America ever, limited evidence is available on effects of fully integrating clinical health service delivery alongside
Access to care
microfinance services through large scale and sustained long-term programs. Using a conceptual model
Microfinance
of health services access, we examine supply- and demand-side factors in a microfinance client popu-
Women's health
Social determinants of health
lation receiving integrated services. We conduct a case study using data from 2010 to 2012 of the design
of a universal screening program and primary care services provided in conjunction with microfinance
loans by Pro Mujer, a women's development organization in Latin America. The program operates in
Argentina, Bolivia, Mexico, Nicaragua, and Peru. We analyze descriptive reports and administrative data
for measures related to improving access to primary health services and management of chronic dis-
eases. We find provision of preventive care is substantial, with an average of 13% of Pro Mujer clients
being screened for cervical cancer each year, 21% receiving breast exams, 16% having a blood glucose
measurement, 39% receiving a blood pressure measurement, and 46% having their body mass index
calculated. This population, with more than half of those screened being overweight or obese and 9% of
those screened having elevated glucose measures, has major risk factors for diabetes, high blood pres-
sure, and cardiovascular disease without intervention. The components of the Pro Mujer health program
address four dimensions of healthcare access: geographic accessibility, availability, affordability, and
acceptability. Significant progress has been made to meet basic health needs, but challenges remain to
ensure that health care provided is of reliable quality to predictably improve health outcomes over time.
© 2015 Elsevier Ltd. All rights reserved.

1. Introduction Organization, 2012; Peters et al., 2008).

The simultaneous burdens of communicable and chronic non-
communicable diseases (NCDs) cause significant morbidity and
mortality in Latin America (World Health Organization, 2009).
Large disparities in disease, risk factors, access to healthcare and
utilization of health services exist both between and within coun-
tries (World Health Organization, 2012). The poor are at particular
risk, with lower health care access and worse outcomes including
higher rates of maternal and child mortality (World Health
* Corresponding author. Boston University School of Management, 595
Commonwealth Avenue, Boston, MA 02215, USA.
E-mail address: geissler@bu.edu (K.H. Geissler).
The capacity of health systems in Latin American countries to
reliably provide primary care for the identification and treatment of
chronic conditions is compromised. Factors such as an insufficiently
skilled workforce, inadequate numbers of healthcare providers
unevenly distributed geographically, and inadequate facilities limit
this capacity (World Health Organization, 2012; O'Brien and Gostin,
2011). High rates of elevated blood pressure and blood glucose (risk
factors for diabetes and cardiovascular disease) and obesity persist,
and NCDs such as diabetes, cancer, and cardiovascular disease are
becoming increasingly common (World Health Organization, 2012,
2009). These chronic diseases cause substantial avoidable
morbidity and premature mortality, with economic implications
due to compromised productivity and high medical costs (Pan

http://dx.doi.org/10.1016/j.socscimed.2015.03.013
0277-9536/© 2015 Elsevier Ltd. All rights reserved.
 K.H. Geissler, S. Leatherman / Social Science & Medicine 132 (2015) 30e37
American Health Organization, 2007). NCDs are not isolated to
wealthier segments of the population; in Latin America, low-
income individuals are adopting a high calorie, low activity life-
style as frequently, if not more so, than other segments of the
population (Popkin, 2001) with increases in associated health risks.
One way to counter inadequate healthcare access, particularly
for poor populations, is through the integration of health services
into existing platforms and infrastructure. This may include using
educational or church related communities (Derose, 2010; Biesma
et al., 2009), work related settings or, for the informal employ-
ment sector, other types of networks such as microfinance in-
stitutions (MFIs). MFIs are broadly defined as providers of financial
services to those outside the reach of formal banks. Given their
large number, expansive geographic presence globally, and routine
contact with the very poor, MFIs may be uniquely situated to serve
as a new distribution channel to impact population health by
incorporating health-related services with financial services for the
poor (Leatherman and Dunford, 2010). MFIs have an increasing and
inherent interest in offering health-related services, as maintaining
the work-related productivity of clients is in their business interest
as well as their social mission. Worldwide, over 3700 MFIs provide
financial services to more than 204 million households
(Microcredit Summit Campaign, 2014). With the accepted multi-
plier of 5.5 individuals per household, integrating health promotion
and services through the platform of MFIs could impact the health
of nearly one billion worldwide.
Although the inextricable relationship between poverty and
health is widely accepted, cross-sectoral programs to simulta-
neously address both for the poor are not widespread. Even for
women engaged in a microfinance program, access to health care
remains a challenge. At a regional health fair held by an MFI in
Bolivia, 24% of attendees stated this was the first formal medical
care they had received (Leatherman et al., 2013). Research from
Latin America, Africa, and Asia shows that carefully designed in-
terventions implemented through MFIs may improve health
knowledge, change health-related behaviors, and increase access to
health services (Leatherman et al., 2012). However, most existing
Fig. 1. Description of Pro Mujer health services.
31
research on microfinance and health integration evaluated new
programs that were either at small scale or, more commonly,
implemented as part of the evaluation. Little evidence exists on the
health or economic impacts of fully incorporating the direct de-
livery of health services with core financial services (Churchill and
Roth, 2006; Matin, 2002; Leatherman et al., 2012) in long running
programs that may provide useful design and implementation
guidance for MFIs and development organizations.
Pro Mujer operates one of the few long-term and fully inte-
grated microfinance and health programs. Pro Mujer is a non-
governmental organization that provides microfinance with
health and human development services for women in urban and
peri-urban areas of Argentina, Bolivia, Nicaragua, Peru, and Mexico.
Pro Mujer has direct primary care delivery and a new screening
program focused on early diagnosis with prevention and manage-
ment of NCDs. Primary care is delivered by physicians and nurses,
and in line with international definitions, is health services char-
acterized by first contact, accessibility, longitudinal relationships,
and comprehensiveness (Starfield, 1992; Kruk et al., 2010). The
design of Pro Mujer's health programs were influenced by multiple
factors; primarily perceived basic health needs of clients, ongoing
financial sustainability of the organization, and assorted factors in
local context.
Programmatic details of services and costs by country are pro-
vided in Fig. 1. Variation in health services across countries is driven
by local context and historic differences in country leadership. Pro
Mujer clinics are co-located with loan services in focal centers to
directly provide preventive and primary care services attentive to
the needs of female adult clients and their children. Starting in late
2010, Pro Mujer standardized the availability of basic health ser-
vices across countries including annual health screenings, coun-
seling, and delivery of modules of health education (i.e., the
“universal” component of the health offerings). In most regions,
primary care with referrals to secondary and tertiary care is
included. The costs of these offerings are covered by the interest
rates charged on Pro Mujer microfinance loans, and are thus paid by
clients prior to the point of care. Pro Mujer trained microfinance
32 K.H. Geissler, S. Leatherman / Social Science & Medicine 132 (2015) 30e37

loan officers as health educators; this frees nurses' time for clinical
care and expands the network of health educators as there are
many more loan officers than affiliated nurses. The annual health
screenings are free at the point of service for body mass index,
blood pressure, clinical breast exam, and glucose measures (for
high-risk groups), and fee-for-service at a nominal cost to clients
for Pap smears.
We examine the universal program of services within the
framework of healthcare access as a case study on the design of a
large and established integrated microfinance and health program.
This case study, in which we examine the structure and composi-
tion of Pro Mujer services as a means to impact access for clients,
provides evidence on how program design of integrated health and
microfinance services directly contributes to healthcare access.

2. Conceptual framework

We use two complementary models of healthcare access in low-

and middle-income countries (Jacobs et al., 2012; Peters et al.,
2008) to develop a hybrid conceptual framework addressing both
supply and demand side barriers to use of health services (Fig. 2).
Access is defined as “the timely use of service according to need”,
with utilization interpreted as realized access (Jacobs et al., 2012;
Peters et al., 2008).
The four dimensions of access are: geographic accessibility of
services; availability of healthcare (i.e., care available at appropriate
level at time needed with supplies required); affordability of health
services and products; and acceptability of services to target pop-
ulation (i.e., responsiveness of providers to social/cultural needs
and preferences of users). Achieving reliable quality of care, at both
the individual and population levels, is dependent upon the per-
formance of all four dimensions simultaneously (Jacobs et al., 2012;
Peters et al., 2008).
As shown in Fig. 2, individual and household characteristics
including socioeconomic status, female empowerment and auton-
omy, and poverty contribute not only to the health status of the Fig. 2. Conceptual model of access to healthcare for microfinance clients.
microfinance client and/or household members but also to the four
dimensions of access. Local, regional, and national factors e such as
health expenditures, characteristics of the health workforce, and We calculated descriptive statistics and examined time trends of
local economic conditions e directly influence the four dimensions care provided and clients served. Health status information from
of access. These factors also contribute to the health status of routinely collected screening data was used to profile individual
microfinance clients or household members; for example, local and population risk. Clinical outcomes are considered consider
food availability and prices may contribute to a client's weight and positive for select NCDs if the measures are above certain thresh-
nutrition status, and water and sanitation conditions to illness olds. Elevated blood glucose was above 126 mg/dl fasting or
occurrence. Access to care influences utilization of care, and utili- 200 mg/dl for random testing. Elevated blood pressure was above
zation of effective care leads to changes in health outcomes. An 140/90 mmHg. Positive Pap smears were results requiring follow-
additional feedback loop leads from health outcomes to current up with colposcopy or biopsy e that is, cervical intraepithelial
health status. neoplasia class I, II, III, or cancer and/or Bethesda LIE low, LIE high,
Jacobs et al. (2012) describe supply and demand side barriers to or Cancer. Patients were overweight if body mass index (BMI)
healthcare access for each of the four dimensions, and identify exceeded 25 and obese with BMI greater than 30.
possible interventions to ameliorate the barriers. As the Jacobs et al. To calculate the percent of clients served, we divided the
framework focused on interventions implemented by government number of new measurements conducted by the total number of
health sector organizations, we combined this set with a previous Pro Mujer clients in a given time period, all of whom were eligible
framework examining access barriers for microfinance clients to receive the universal program components. We matched the set
(Leatherman et al., 2012). of barriers with programmatic interventions delivered by the Pro
Mujer program to analyze the program design and identify con-
3. Methods tributions of the Pro Mujer program to healthcare access for clients.
We limited details of barriers to those specifically addressed rather
We use the Pro Mujer experience in Argentina, Bolivia, Mexico, than providing an exhaustive list. We included supply and demand
Nicaragua, and Peru as a case study in analysis of program design in side factors as Pro Mujer's integrated program with direct delivery
addressing social determinants of health (e.g., poverty) and of clinical services addresses both. We did this with careful analysis
expanding access to health-related services through integrated of what was and was not included in the design of the program and
microfinance and health services. As data sources, we used how these aligned with aspects of the conceptual framework. We
administrative data, reports of external consultants, and Pro Mujer used binary indicators of whether or not certain aspects of the
internal process and clinical monitoring data. program alleviated a barrier identified by the framework, and then
 K.H. Geissler, S. Leatherman / Social Science & Medicine 132 (2015) 30e37 33

further analyzed documents to determine more specifically how Table 1

successful these were. Process indicators of preventive screenings and clinical care in five countries,
2010e2012.
The study was reviewed by the University of North Carolina
Office of Human Research Ethics, which determined that the study Total Average per Average %
was not human subjects research and did not require Internal Re- month of clients
per yearc
view Board approval.
Panel A: Preventive screenings
Pap smears 116,407 3234 (1245) 13%
4. Results Breast exams 174,314 4842 (3103) 21%
Blood glucose measurements 129,056 3585 (2764) 16%
4.1. Client characteristics Blood pressure measurements 460,127 12,781 (6566) 39%
Body mass index measurementsa 353,454 14,727 (3504) 46%
Panel B: Clinical care
Operations in five Latin American countries served more than Adult visits (acute or chronic care) 858,926 23,859 (6852)
250,000 clients in 2012. As of December 2012, 39% of clients were in Gynecological consultations 164,242 4562 (2019)
Bolivia and 22% in Peru (Fig. 3). Due to Pro Mujer's focus on mi- Family planning consultations 218,991 6083 (2014)
croenterprise loans, women served are not the ultra-poor; how- Prenatal visits 23,234 645 (188)
Child well visitsb 98,201 2728 (712)
ever, they require ongoing loans and economic assistance to
Pediatric acute care visitsb 100,977 2805 (965)
maintain and grow their small enterprises. The women served have
an average age of approximately 40 years. Internal and consultant Standard deviation in parentheses.
a
Information on body mass index measurements was not collected until January
reports showed general demographic characteristics of clients. In 2011.
Nicaragua [2008], 53.4% of clients had a primary education or less b
Pediatric services are not provided in Mexico or Nicaragua.
and 29% lived in poverty as measured by a poverty index (Delgado, c
Calculated as (# new measures/# of clients). Some measures may have been
2008). In Bolivia [2009], 65% of clients had a monthly income below conducted for non-clients, but these are a very small proportion and client status
was not recorded in the data.
the national poverty line (Conly et al., 2010). In two regions of
Argentina, 14% of clients had incomes below the national poverty
line and over 90% were of low-to lower-middle class income visits for family planning (Table 1B). Again, there was a sharp in-
(Aparicio, 2010). In Peru in 2008, the median client interviewed had crease over time as Pro Mujer capacity increased (Fig. 4B). Signifi-
a monthly household expenditure higher than the country's min- cant variation in utilization across countries existed, with Pro Mujer
imum living wage but not exceeding that of basic family needs (Ma clinics serving significantly fewer clients in Argentina and Mexico
et al., 2008). No information was available on client income or (results not shown). This is due to low internal supply capacity,
education in Mexico. which constrained service delivery due to use of third-party private
providers (Argentina) and health campaigns (Mexico). Pro Mujer
4.2. Utilization of health related services recently (2013) established greater capacity by employing nurses to
begin to generate and meet demand for these screening services.
From January 2010 to December 2012, Pro Mujer clinics con- Both the preventive care and primary care measures showed
ducted a substantial number of preventive screenings for repro- seasonality, with visits declining in the December to February
ductive and general health, including 460,127 blood pressure period each year; this is in line with seasonality of meeting atten-
measurements and 116,407 Pap smears (Table 1A). The number of dance related to business and credit related activities, holidays, and
preventive screenings increased substantially over time (Fig. 4A) as school schedules.
the health program became more standardized across countries,
clinics expanded capacity for conducting preventive screenings, 4.3. Health status and screening results
and outreach to clients encouraged use. Over this time period,
approximately 46% of clients used at least one screening service. Results from the preventive screenings show Pro Mujer micro-
Over the same period, Pro Mujer also increased the clinical care finance clients have substantial risk factors for current and future
provided to clients and their families, with 858,926 visits for adult chronic diseases such as diabetes and cardiovascular disease
primary care, 164,242 visits for gynecological needs, and 218,991 (Table 2). More than half (58%) of body mass indices measured
(including new and repeat measurements) were categorized as
overweight or obese. In 2012, overweight and obese categories
were separated for Peru, Bolivia, and Argentina; approximately half
are in the overweight (but not obese) category and half fall above
the cutoff for obesity. Nine percent of glucose tests were clinically
positive for elevated blood glucose, which is an indicator of either
impaired glucose tolerance (i.e., “prediabetes”) or diabetes mellitus.
One percent of breast exams were referred for further examination
and possible treatment; three percent of Pap smears were clinically
positive for possible disease and required follow-up. Four percent
of blood pressure measurements were considered elevated.

4.4. Analysis of program design e effect on access

We examined the match of program elements on the four di-

Fig. 3. Number of Pro Mujer clients by country (2008e2012).
mensions of access and determined ways health programs address
supply- and demand-side barriers (Fig. 5).
Pro Mujer improved access on the first dimension e geographic
accessibility of services e through co-location of health education
and clinical service provision within focal centers. This co-location
34 K.H. Geissler, S. Leatherman / Social Science & Medicine 132 (2015) 30e37

improved access for rural clients; loan repayment meetings are

25,000 held in rural clients' home villages and thus clients do not travel to
the focal centers on a regular basis. Access to primary care in rural
Blood pressure measurements
areas is facilitated by Pro Mujer through regularly scheduled “jor-
Number of Screenings Conducted

20,000
nadas” when physicians/nurses go to rural areas together with
credit officers and, in the case of Peru, using mobile clinics with a
15,000 Body mass index primary care physician or nurse, radiologist, lab technician, and
measurements
pharmacy. Differences in definitions of rurality across countries
10,000
Breast exams mean that rural clients in Bolivia may have significantly lower
access.
Glucose measurements Health training provided by loan officers contributed to
5,000 geographic accessibility. In Bolivia, where nurses deliver health
Pap smears
training, three health education sessions were delivered per client
0
per year. In Nicaragua, credit officers deliver health training and
clients received an average six sessions per year. Health education
provided to clients included information on topics including
reproductive and sexual health, self-esteem, health behaviors, and
40,000 nutrition, as well as availability of Pro Mujer provided health ser-
vices. This education may influence utilization by reinforcing health
35,000
related messages and increasing awareness of future care needs.
Number of consultations provided

30,000
Some evidence of behavior change e utilization of PAP smears and
other screening services, drinking more water rather than caloric
25,000 Adult beverages, more time spent exercising e was found among a small
group of respondents surveyed in Nicaragua in 2011 after receiving
20,000
education from loan officers (Smith, 2012).
15,000
Pro Mujer addressed the second dimension of access e avail-
ability of healthcare e in an innovative manner with direct delivery
Family planning
10,000 of clinical services. Focal centers had operating hours centered
around loan repayment meeting times, and clinics aimed to keep
Gynecological
5,000 wait times short to improve client use of services and reduce op-
Pediatric
Child Well Visits portunity costs. Pro Mujer clinicians provided referrals to more
0 Prenatal
specialized providers when indicated by screening results or
Sep-10

May-11

Sep-11

Sep-12
May-10
Jul-10

Nov-10

Jul-11

Nov-11

May-12
Jul-12

Nov-12
Jan-10
Mar-10

Jan-11
Mar-11

Jan-12
Mar-12

diagnosis; maintaining coordination of care after making a referral

was an ongoing challenge.
Fig. 4. Health services provided by Pro Mujer in five countries, 2010e2012. Panel A: The third dimension of access e affordability of health services
Pro Mujer provided preventive screening services (Note: Monthly measures were not and products e was approached from a supply side perspective
available for January to June 2010, so the total number of screenings during this period
with fee-for-service primary care prices designed to cover costs but
divided by six months was used to give monthly values. Body mass index measure-
ment began in January 2011.). Panel B: Pro Mujer provided primary care services (Note: remain affordable for clients. Additionally, the timing of payment
Adult consultations are visits for primary care (acute or chronic). They do not include was an important factor, with many of the costs of the health
visits for family planning, prenatal care, or gynecological needs. Monthly measures program built into loan interest rates and fees. This contrasts with a
were not available for January to June 2010, so the total number of each type of visit fee-for-service model, where clients have to at the point of care.
during this period divided by six months was used to give monthly values. Pediatric
services are not provided in Mexico or Nicaragua.).
Certain preventive screenings were provided free at the point of
care with those requiring supplies (e.g., PAP smear) offered at a
nominal fee. Results were available to clients at the time of
Table 2 screening or, in the case of Pap smears, after the lab results are
Health risk indicators from screenings in five countries, 2010e2012. returned. The success of matching the price of services with client
Indicator Percent of cases ability to pay was evidenced in part by the volume of services
provided. Pro Mujer microfinance clients were all women oper-
Pap smear positive cases 2.7%
Breast exams referred for further testing 0.8% ating small businesses who likely value their time as an opportu-
Elevated blood glucose (as a % of total 9.0% nity cost; if they do not work, they do not have income. Thus,
tests-new and repeat) decreased wait times over government services and the lack of
High blood pressure 4.4%
need to travel elsewhere for services makes the aggregate cost of
Overweight or obesea 58.1%
the services lower than the direct price paid for health services may
a
Body mass index measurements were not conducted until January 2011 and indicate. Access to essential drugs is an issue for many clients; for
were not conducted during the 2011e2012 period in Mexico.
urban, peri-urban, and many rural clients, drugs are available but
the lack of cash at the time the medication is needed remains
was designed to simultaneously decrease supply and demand side problematic.
barriers to access for clients who regularly attend financial meet- Pro Mujer improved the fourth dimension of access e accept-
ings at these locations. It likely reduced indirect costs to most Pro ability of health services to the target population e with direct
Mujer households (living in urban and peri-urban settings) as they delivery of care. As clients were already actively engaged with the
are already in the location of services, so do not lose further pro- organization, they may be more trusting of services than they
ductive time in transit. Improvement in access due to co-location is would be from a separate private provider. Transparency of costs
less reliably true for acute care, which requires clients to travel to was designed to increase acceptability to clients, as prices of ser-
reach the focal centers when care is needed outside of standard vices were posted in clinics and paid prior to consultations. The
meeting times. Additionally, co-location of services may not have group based model of loan repayment and health education may be
 K.H. Geissler, S. Leatherman / Social Science & Medicine 132 (2015) 30e37 35

Fig. 5. Match of detailed supply and demand side barriers with Pro Mujer programs.

particularly important for fostering acceptability of services,
allowing women to learn from one another and creating peer ef-
fects related to behavior changes or seeking care.
In each dimension of access, there were disparities across
countries, and within country regional variation. Some of this was
due to the variability of historical programming; standardization of
programs may reduce disparities. In Argentina, screenings were
conducted by third parties (i.e., private physicians with fee for
service contracts paid by Pro Mujer) until the second half of 2011,
after which these were conducted in the focal centers. As this
transition was systematized and clients made aware of it,
increasing numbers of clients made use of the services.
5. Discussion
The Pro Mujer integrated model represents a promising
36 K.H. Geissler, S. Leatherman / Social Science & Medicine 132 (2015) 30e37
approach to early intervention through health education, facilitates
access to primary health services, and allows for screening and
outreach with populations of women at high risk of NCDs. The
match of Pro Mujer service provision to the dimensions of access
showed the strategy addressed supply and demand side barriers.
The number of services provided, particularly surrounding
screening for NCDs, demonstrated direct delivery of clinical care is
geographically accessible, affordable, and acceptable for a large
subset of clients.
Between 2011 and 2012, 46% of clients had at least one screening
procedure, representing a high level of engagement with the
available health services given previous rates of contact with health
services for similar populations (Leatherman et al., 2013). Using
monitoring data on service provision and clinical outcomes, we
found substantial increased rates of preventive screenings and
primary care for acute and chronic conditions among clients with
increased use as knowledge of the program grew and services were
standardized across countries and the universal package of health
services was implemented. Demographic and health status data
indicated many of these clients have risk factors putting them in
danger of developing long-term health conditions. Primary care for
NCD diagnosis and treatment has been shown to effectively reduce
NCD risk factors and prevent advanced stage disease through early
detection and treatment (Di Cesare et al., 2013). A recent study
found screening for cervical cancer highly cost-effective in Mexico,
and blood pressure control was highly cost-effective for diabetes
(Salomon et al., 2012). Implementing “universal, financially and
physically accessible, high-quality primary care” is considered a key
action to reduce NCDs (Di Cesare et al., 2013), and the health care
offered by Pro Mujer may be pivotal for improving the health status
of microfinance clients.
Simultaneously addressing multiple barriers through the use of
an integrated microfinance, health education, and clinical service
provision program appeared to have multiple positive impacts on
client access. Unusual for MFIs, another notable leader in integrated
microfinance and health, BRAC in Bangladesh, has done similar
work on microfinance and health integration (BRAC, 2013).
Analyzing the design of such programs focusing on influences on
healthcare access provides more information about potential scope
and scalability to other programs in pilot form. In this case, the
microfinance platform provides the distribution channel for health
services, but schools or churches might also substitute if they could
likewise meet the demand and supply needs.
Although the program we study is not that of a strictly finan-
cially focused MFI, it provides lessons that could be useful not only
to other MFIs but also to other human services and development
organizations as they expand their reach into health promotion and
services. In particular, although many MFIs do not have the capacity
to add direct delivery of health services, the Pro Mujer model in-
cludes replicable interventions that improve health care access
such as training credit officers as health educators, combining
health education with group based microfinance to foster accept-
ability of health services, and partnering with health or govern-
ment organizations to conduct health campaigns in low-access
areas.
Rural populations, which form a minority of Pro Mujer's clients,
remained at risk of low or non-existent access to health services.
This is an ongoing challenge that Pro Mujer addresses through
community health fairs and mobile clinics. However, it is compli-
cated to reach remote rural populations in a cost-effective way, not
only for health but also for financial services. Despite increases in
initial access from this integrated strategy, there were significant
unknowns in terms of the effectiveness of healthcare provided. We
hope some of the outstanding and important questions will be
addressed through systematic data collection, monitoring, and
evaluation recently implemented by Pro Mujer.
The significant increased rates of screening may lead to earlier
diagnosis and treatment of NCDs; however, we were not able to
systematically link increases in screenings e free at the point of
care e to follow-up treatment for clients who tested positive for
risk factors. Some risk factors, such as being overweight or obese,
having elevated blood sugar, or having elevated blood pressure
were addressed directly by referral to an onsite physician at the
time of screening, and indirectly through the provision of health
education and behavior modification strategies delivered as mod-
ules during loan repayment meetings. Future research will examine
whether these strategies culminate in knowledge and behavioral
changes leading to improved health status. Positive Pap smears and
breast exams were referred for follow-up either by a Pro Mujer
provider or a specialist outside of Pro Mujer. These follow-ups were
informally tracked at the clinic level in some areas, but were not
monitored at the central level. If referral systems to outside pro-
viders were not adequate, this may result in women not receiving
adequate follow-up care after a positive result due to lack of fa-
miliarity with the health system, inability to afford care, or high
time costs of seeking care.
Additionally, some data points raised questions regarding the
reliability of screening values. The proportion of blood pressure
measurements considered elevated (4.4%) was much lower than
expected given country and region level population averages
(World Health Organization, 2012; Miranda et al., 2013). Since we
had access only to aggregated data with an indicator variable for
high blood pressure rather than the actual measurement, we were
not able to determine why these measures were not in line with
expectations and other clinical health status indicators. However,
follow-up tests have shown that the percentage of clients with
elevated blood pressure was consistent across regions and when
conducted by different health providers. Addressing this is an
important consideration going forward to ensure that clients are
being given accurate information about their health status.
We were limited in this research to supply side information
from existing administrative data on service provision. Future
research will address questions of access from a client perspective.
As clinical indicators were recorded as binary indicators of health
outcomes, we are not able to examine the severity of these con-
ditions or to determine correlation between risk factors and so-
cioeconomic characteristics. The Pro Mujer experience may be
somewhat unique, in that they provide microfinance, but are also
non-governmental and have a focus on client health. As such, they
may be more suited for the direct provision of clinical services than
other MFIs that are financially focused. However, the experience
can provide guidance for MFIs considering linkages to health pro-
viders; taking into account how health interventions will affect
each component of access is critical for improving client health
outcomes. We did not have the data to address the cost-
effectiveness of service provision, which is an important consid-
eration for many MFIs. Future research on Pro Mujer should address
the cost-effectiveness of programs, longer-term implications for
health outcomes (e.g., blood pressure control, changes in obesity
status), and systematically evaluate the effect of programs on client
utilization and health. Future research regarding microfinance and
health integration should examine not only the effects of such
programs on healthcare access, but also on the clinical quality of
care provided and the costs of care both to the clients and the MFI.
The findings from Pro Mujer have important policy implications
for future work in improving healthcare access, particularly with
the growing importance of NCDs in many developing countries.
Utilizing existent but nontraditional distribution platforms such as
MFIs may be an effective way to provide larger populations with
primary care, particularly where current health care supply is
 K.H. Geissler, S. Leatherman / Social Science & Medicine 132 (2015) 30e37
inadequate. As integrated microfinance and health programs
become more common worldwide, it is important to consider how
evidence gleaned from current practices can inform design and
implementation of new programs, as well as to conceptualize and
implement pragmatic evaluation for the many naturally occurring
experiments in integrating microfinance and health.
Acknowledgment
We would like to thank Gabriela Salvador and Jana Smith for
their assistance with providing the data and helpful comments on
earlier drafts.
References
Aparicio, S., 2010. Report of Impact Study. Pro Mujer, Argentina.
Biesma, R.G., Brugha, R., Harmer, A., et al., 2009. The effects of global health ini-
tiatives on country health systems: a review of the evidence from HIV/AIDS
control. Health Policy Plan. 24, 239e252.
BRAC, 2013. About BRAC 2013 [cited 2013 25 January 2013]. Available at: http://
www.brac.net.
Churchill, C., Roth, J., 2006. Microinsurance: opportunities and pitfalls for micro-
finance institutions. In: Churchill, C. (Ed.), Protecting the Poor, a Microinsurance
Compendium. International Labor Organization, Geneva.
Conly, C., Door, D., Hwang, J., et al., 2010. Antares/Pro Mujer Bolivia Analysis and
Health Recommendations. Harvard School of Public Health, Cambridge, MA.
Delgado, E., 2008. Survey report for poverty measurement. Pro Mujer, Leon,
Nicaragua.
Derose, K.P., 2010. The Role of Faith-based Organizations in HIV Prevention and Care
in Central America. RAND Corporation, Santa Monica, CA.
Di Cesare, M., Khang, Y.H., Asaria, P., et al., 2013. Inequalities in non-communicable
diseases and effective responses. Lancet 381, 585e597.
Jacobs, B., Ir, P., Bigdeli, M., et al., 2012. Addressing access barriers to health services:
an analytical framework for selecting appropriate interventions in low-income
Asian countries. Health Policy Plan. 27, 288e300.
Kruk, M.E., Porignon, D., Rockers, P.C., Van Lerberghe, W., 2010. The contribution of
primary care to health and health systems in low- and middle-income
37
countries: a critical review of major primary care initiatives. Soc. Sci. Med. 70
(6), 904e911.
Leatherman, S., Dunford, C., 2010. Linking health to microfinance to reduce poverty.
Bull. World Health Organ. 88, 470e471.
Leatherman, S., Geissler, K., Gray, B., et al., 2013. Health financing: a new role for
microfinance institutions? J. Int. Dev. 25 (7), 881e896.
Leatherman, S., Metcalfe, M., Geissler, K., et al., 2012. Integrating microfinance and
health strategies: examining the evidence to inform policy and practice. Health
Policy Plan. 27, 85e101.
Ma, C., Escobedo, S., Cordero, L., et al., 2008. Market Study for the Health Product for
Pro Mujer Peru (Peru).
Matin, I., 2002. Finance for the poor: from microcredit to microfinancial services.
J. Int. Dev. 14, 273e294.
Microcredit Summit Campaign, 2014. Data Reported: State of the Campaign Report
2014. Microcredit Summit Campaign [cited 2014 October 6]. Available at: http://
stateofthecampaign.org/data-reported/.
Miranda, J.J., Herrera, V.M., Chirinos, J.A., et al., 2013. Major cardiovascular risk
factors in Latin America: a comparison with the United States. The Latin
American Consortium of Studies in Obesity (LASO). PLoS One 8, e54056.
O'Brien, P., Gostin, L., 2011. Health Worker Shortages and Global Justice. Millbank
Reports. Millbank Memorial Fund, New York, NY.
Pan American Health Organization, 2007. Health conditions and trends. In: Orga-
nization PAH (Ed.), Health in the Americas. Pan American Health Organization,
Washington, DC.
Peters, D.H., Garg, A., Bloom, G., et al., 2008. Poverty and access to health care in
developing countries. Ann. N. Y. Acad. Sci. 1136, 161e171.
Popkin, B.M., 2001. The nutrition transition and obesity in the developing world.
J. Nutr. 131, 871Se873S.
Salomon, J.A., Carvalho, N., Gutierrez-Delgado, C., et al., 2012. Intervention strate-
gies to reduce the burden of non-communicable diseases in Mexico: cost
effectiveness analysis. BMJ 344, e355.
Smith, J., 2012. Use of Credit Officers as Health Trainers: The Experience of Pro
Mujer. Pro Mujer Inc, New York.
Starfield, B., 1992. Primary Care: Concept, Evaluation, and Policy. Oxford University
Press, New York.
World Health Organization, 2009. Global Health Risks: Mortality and Burden of
Disease Attributable to Selected Major Risks. World Health Organization,
Geneva.
World Health Organization, 2012. Global Health Observatory Data Repository.
Available at: http://apps.who.int/ghodata/?vid¼61180.