DIABETICMedicine

DOI: 10.1111/dme.13099

Research: Care Delivery

Delivery of Type 2 diabetes care in low- and
middle-income countries: lessons from Lima, Peru

M. K. Cardenas1, J. J. Miranda1,2 and D. Beran3

1
CRONICAS Center of Excellence in Chronic Diseases, Universidad Peruana Cayetano Heredia, 2Department of Medicine, Universidad Peruana Cayetano Heredia,
Lima, Peru and 3Geneva University Hospitals and University of Geneva, Division of Tropical and Humanitarian Medicine, Geneva, Switzerland

Accepted 11 February 2016

Abstract
Aims The health system’s response is crucial to addressing the increasing burden of diabetes, particularly that affecting
low- and middle-income countries. This study aims to assess the facilitators and barriers that help or hinder access to
care for people with diabetes in Peru.
Methods We used a survey tool to design and collect qualitative and quantitative data from primary and secondary
sources of information at different levels of the health system. We performed 111 interviews in Lima, the capital city of
Peru, with patients with diabetes, healthcare providers and healthcare officials. We applied the six building blocks
framework proposed by the World Health Organization in our analysis.
Results We found low political commitment, as well as several barriers that directly affect access to medicines, regular
laboratory check-ups and follow-up appointments for diabetes, especially at the primary healthcare level. Three major
system-level barriers were identified: (1) the availability of information at different healthcare system levels that affects
several processes in the healthcare provision; (2) insufficient financial resources; and (3) insufficient human resources
trained in diabetes management.
Conclusion Despite an initial political commitment by the Peruvian government to improve the delivery of diabetes
care, there exist several key limitations that affect access to adequate diabetes care, especially at the primary healthcare
level. In a context in which various low- and middle-income countries are aiming to achieve universal health coverage,
this study provides lessons for the implementation of strategies related to diabetes care delivery.
Diabet. Med. 33, 752–760 (2016)

2015 with the goal of a one third reduction in premature

Introduction
The health system’s response is crucial to addressing the
increasing burden of diabetes, especially that affecting low-
and middle-income countries (LMIC) [1]. Globally, diabetes
was responsible for 1.5 million premature deaths in 2012
[2]; between 2013 and 2035, the prevalence is predicted to
increase from 387 million to 592 million [3]. In LMICs, an
80% increase is predicted during this period.
In order to address this growing burden, in 2011, the
World Health Organization (WHO) developed the Global
Action Plan for the Prevention and Control of Noncommu-
nicable Diseases (NCDs) 2013–2020 [4]. The overarching
aim of the Global Action Plan is to reduce premature
mortality from NCDs by 25% by 2025. NCDs were also
included in the Sustainable Development Goals adopted in
Correspondence to: Maria Kathia Cardenas.
E-mail: maria.cardenas.g@upch.pe
death secondary to NCDs by 2030, as well as the achieve-
ment of universal health coverage (UHC) [5].
UHC assumes a political commitment with the ability to
operationalize the provision of affordable and quality
healthcare services to all [6]. It entails three dimensions of
extending universal services: providing services to people not
covered; increasing the number of services provided; and
finally, reducing direct payments, including user fees. [7].
UHC has yet to be fully implemented in many LMICs despite
its inclusion in a 2012 United Nations Resolution [8]. The
challenge remaining is how LMICs will provide UHC with
the concurrent burden of NCDs [9].
Peru exemplifies some of the challenges of UHC imple-
mentation [10] while providing diabetes care in a LMIC.
Peru has a population of 30 million and has seen rapid and
steady economic growth since 2001 with a gross domestic
product per capita of US$6621 in 2013, according to the
World Bank. A UHC programme called the Seguro Integral

752 ª 2016 Diabetes UK

 Research article
What’s new?
 This study presents an assessment of the facilitators and
barriers to diabetes care at different levels of the health
system in Lima, Peru.
 The methods and analytical framework are both
promoted by the World Health Organization.
 Analysis of a complex system using a comprehensive
review of secondary sources and interviews.
 The findings provide lessons for the implementation of
diabetes care programmes in other low- and middle-
income countries.
 Despite political commitment and the adoption of
universal health coverage in Peru, limitations affecting
the adequate delivery of diabetes care at the primary
healthcare level exist.
de Salud (SIS: Comprehensive Health Insurance) was intro-
duced in 2002 offering a variety of services: pharmaceutical,
emergency, inpatient and outpatient primary-level and spe-
cialty care for the poor and extremely poor. In 2009, Peru
approved the Universal Health Insurance Law to establish an
obligatory basic service plan for all insurance companies to
provide greater healthcare access to more people. Mean-
while, in the last decade, Peru has faced an increasing
prevalence of NCDs and diabetes [3].
Historically, health care in Peru and other LMICs has
focused mainly on the control of communicable diseases and
acute care, and has not been designed or prepared to deliver
comprehensive services to patients with NCDs and co-
morbidities [1,11,12]. Moreover, the Peruvian healthcare
system is characterized by high complexity and fragmenta-
tion [13]. As a country, Peru faces challenges that are
common to most LMICs; it has started health system reform
and UHC programmes during a period of sustained eco-
nomic growth, but still has inequities in access to care [10].
In order to provide lessons for other LMICs, the aim of this
study was to assess the facilitators and barriers that help or
hinder access to care for people with Type 2 diabetes in
Lima, Peru.
Methods
The organization of the Peruvian health system was taken
into account when designing the study because it is charac-
terized by the participation of various stakeholders [13] at
different levels, including: social security (EsSalud), Ministry
of Health (MINSA) and public insurance SIS, regional and
local healthcare offices, police and armed forces, and private
providers/insurers.
The manual How to Investigate Access to Care for
Chronic Noncommunicable Diseases in Low- and Middle-
ª 2016 Diabetes UK
DIABETICMedicine
Income Countries [14] was used to design and collect
qualitative and quantitative data from primary and sec-
ondary sources at different levels of the health system. The
manual is a practical field guide that was developed
considering past implementations of the Rapid Assessment
Protocol for Insulin Access to study access to diabetes care
and medicines in Kyrgyzstan, Mali, Mozambique, Zambia,
Nicaragua, Vietnam and the Philippines. This type of
document is particularly appropriate for countries like Peru
that have limited time and resources to conduct pragmatic
research to assess the development of healthcare services
and policies for diabetes management. The survey tool
consists of an 11-module field manual that guides
researchers in planning and implementing descriptive
health system studies focused on one or more NCDs
(Fig. 1). The manual provides questionnaires that were
translated and then adapted taking into account the
characteristics of the Peruvian health system and the
diabetes situation in Peru.
The WHO’s six building blocks framework [15] was
used as an analytical framework. The framework describes
each health system in terms of the following compo-
nents: information; leadership and governance; financing;
health workforce; service delivery; and medical products,
vaccines and technologies. Each of these components
includes relevant topics about health system performance
(Fig. 1).
Data collection
A detailed review of secondary sources (i.e. regulations, the
web pages of health institutions, national statistics, drug
prices databases, health reports and investigations) was
carried out to assess the prevalence, availability, price, access
and healthcare response to diabetes at a country level.
Primary information was collected in 2013 through semi-
structured interviews (Table 1).
Primary data questionnaires included the 11 themes of
study and covered different aspects of health service
provision such as: training for healthcare workers;
perceived difficulties for the different actors regarding
healthcare delivery; and the availability of medicines,
equipment and consumables, among others. The inter-
views provided key information from diverse health
system actors from both the public and private sectors:
people with diabetes, healthcare providers, directors of
healthcare facilities and healthcare officials from MINSA
and the Diabetes Association. Primary data were collected
using purposive sampling in Lima, the capital city of
Peru, where one third of the total population of Peru live.
Nine facilities, at both the primary and secondary
healthcare response levels, were selected across three
different zones in Lima.
In each facility, we implemented a ‘snowball’ sampling
strategy to reach key participants until data saturation was
753
DIABETICMedicine
Data Collection
Tool: A survey manual to assess noncommunicable
diseases in low- and middle-income countries [14]
Multiple sources of information
• Secondary information: local literature and secondary
data.
• Primary data: interviews at different levels of the health
system.
4 levels of assessment:
• National;
• Intermediate (regional/local offices);
• Individual (facilities, health workers);
• Local (patients).
11 survey themes:
1. General information;
2. Healthcare structure;
3. Financial issues;
4. Health insurance;
5. Disease-related policies, programmes and activities;
6. Supply/procurement systems;
7. Resource allocation/ availability of care;
8. Price/affordability of care;
9. Disease management and treatment issues;
10. Referral issues;
11. Patient issues.
FIGURE 1 Methodology for data collection and data analysis.
achieved. Table 1 shows the participant distribution within
the sample, in which 111 participants were interviewed. The
average duration of each interview was 45–60 min. Each
interview was initiated only after obtaining informed
consent.
Data analysis
The collected data were registered in an Excel database.
Qualitative data were indexed and synthesized according to
the study themes. Quantitative data were analysed using
frequencies and descriptive statistics. Quantitative and qual-
itative data from primary and secondary sources were
triangulated and analysed according to the six building
blocks framework [15].
Ethics
Ethics approval was obtained from the ethics committee of
the Universidad Peruana Cayetano Heredia of Lima, Peru.
Administrative authorization was obtained for each partic-
ipating facility site.
754
Delivery of diabetes care in Peru  M. K. Cardenas et al.
Data Analysis
Framework:
Six building blocks (World Health Organization) [15]
Some relevant topics of study in each building:
1. Information
Facility- and population-based information and
surveillance systems (health systems performance,
determinants and health status).
2. Leadership and governance:
Health sector policies; regulations; political environment;
health system design.
3. Finance:
National health financing policies; data on health
expenditures; health financing system.
4. Health workforce:
Availability, distribution and performance of the national
workforce; training; education programmes.
5. Service delivery:
Demand for services; packages; delivery models; provider
network; infrastructure; management; safety and quality.
6. Medical products, vaccines and technologies:
Access; prices; national policies; procurement system;
essential medicines, commodities and equipment;
adherence.
Results
The results are presented using the six building blocks
framework [15].
Information
Health information on NCDs and risk factors is scarce in
Peru. There is no current official figure for the prevalence of
diagnosed and undiagnosed diabetes in the country from an
official population-based epidemiological survey. However,
since 2014 the National Demographic and Family Health
Survey, performed nationwide among persons aged 15 years
and older, has included questions about diagnosed diabetes.
According to this survey, the national prevalence of
diagnosed diabetes was 3.2% in 2014 and the highest
prevalence was found in Lima (4.5%) [16]. The last official
survey to diagnose diabetes at a national level was carried
out in 2005 among 4206 persons aged 15 years and older,
and the estimated prevalence of diabetes was 2.8% [17]. This
survey also assessed the prevalence of risk factors. The
International Diabetes Federation (IDF) [3] estimated a
ª 2016 Diabetes UK
 Research article DIABETICMedicine

Table 1 Number of interviews performed at the different levels of the Peruvian health system (2013)

Ministry of Health
(MINSA) Social Security
(EsSalud) Armed Forces Private
Post (01)/ Hospitals Polyclinics Hospital Clinic Total
Interviewee Centres (02) (02) (02) (01) (01) (N = 111)

Patients with diabetes

Patients with Type 1 diabetes mellitus 0 3 0 0 0 3
Patients with Type 2 diabetes mellitus 5 10 8 5 8 36
Facilities
Directors 3 2 2 1 1 9
Health workers
Physician (general practitioner) 4 0 1 0 0 5
Physician (internal medicine) 0 2 1 1 1 5
Physician (endocrinology) 0 2 0 1 1 4
Nurse* 3 4 2 1 1 11
Nutritionist 2 2 1 1 1 7
Internal pharmacy staff† 3 1 2 2 1 9
Internal laboratory staff† 2 2 2 1 1 8
External pharmacy staff† 3 1 1 1 1 7
External laboratory staff† 3 0 0 0 1 4
Other key stakeholders
Diabetes Association – – – – – 1
Ministry of Health officers – – – – – 2

*Professional or technician nurse.

†‘Internal’ laboratories or pharmacies operate within the health facility, whereas ‘external’ laboratories or pharmacies are private and
operate outside the health facility.

national diabetes prevalence of 6.1%, suggesting that
~ 1 143 590 people aged between 20 and 79 years live with
diabetes in Peru. A diabetes prevalence of 7% was estimated
by the PERUDiab study [18] carried out between the 2011
and 2012 among 1677 people aged ≥ 25 years living in
diverse urban and peri-urban settings across Peru. Similar to
the IDF findings, a baseline assessment from the CRONICAS
cohort study between 2010 and 2011 [19] showed an overall
diabetes prevalence of 6.2%. This age-standardized figure
was estimated using a sample of 3238 people aged ≥ 35 years
from three settings (Lima, Puno and Tumbes) with varying
degrees of urbanization. This study also estimates that,
among people with Type 2 diabetes, 61.3% were aware of
their diagnosis and 71.4% were self-reported to be under-
going treatment [19]. A pilot diabetes surveillance at 18
hospitals was started in Lima in 2011 [20] and one
conclusive result was the prevalence of inadequate glycaemic
control and poor treatment adherence.
Other data sources include a public database from MINSA
with medication prices at the public and private facility level.
Furthermore, since 2013, MINSA provides information
regarding the monthly availability of essential medicines at
each of its facilities. Some difficulties that affected the
management of patients with diabetes at MINSA hospitals
were incomplete or misplaced patient files.
Leadership and governance
MINSA is the leading authority on health issues in Peru, in
both the public and private sectors. Since 2004, a National
Health Strategy for the Prevention and Control of Noncom-
municable Diseases (NCDs National Health Strategy) has
existed and an action plan for 2004–2012 has been created,
the updated version of which has not yet been approved. A
specific NCD unit within MINSA works to implement a
regulatory framework for the prevention and management of
NCDs. Among other duties, they also promote intersectorial
participation, NCD research, NCD registry and surveillance,
while strengthening primary-level healthcare services, med-
ication availability and access to insurance and health care
for poorer populations.
A general protection law for people with diabetes was
approved in 2005. This law proposes the creation of a patient
registry, which has yet to be fully implemented at a national
level. In November 2015, almost 10 years after the creation
of the protection law and the NCDs National Health
Strategy, a clinical practice guideline document for the
diagnosis, treatment and control of Type 2 diabetes at the
primary healthcare level was approved. Despite this achieve-
ment, no clinical practice guideline for the management of
diabetes complications at the secondary and tertiary levels
exists.
Other relevant governmental efforts include: exoneration
of import and value-added taxes for various glycaemic
control medications, implementation of a corporate purchas-
ing scheme for medicines, and the creation of a mechanism
called ‘Inclusive Pharmacies’. The exoneration was imple-
mented in 2006, but in 2010 it was found that most
exonerated medications were not sold at a lower price [21].
By contrast, the corporate scheme for purchasing medicines

ª 2016 Diabetes UK 755

DIABETICMedicine
from different public sector providers (MINSA, EsSalud,
police and armed forces) has been shown to be effective in
lowering the wholesale price. The mechanism of ‘inclusive
pharmacies’ was created on December 2013 as part of a
package of regulations to improve UHC in Peru. The aim
was to increase access to medication among patients with
public insurance (SIS) who suffer from diabetes or hyperten-
sion by allowing private pharmacies to participate in public
bidding. By the end of 2015, no progress had been made in
the implementation of this mechanism.
Financing
In Peru, the main sources of healthcare funding are taxes,
social security, private insurance and out-of-pocket expen-
diture. Previous evidence shows that out-of-pocket expen-
diture is the primary source of funding, followed by public
resources [22]. The total public budget spent on health
programming at the national, regional and local levels was
US$3 600 705 534 in 2013, according to Ministry of
Economics and Finance statistics (Portal de Transparencia
Econ omica). Of that, only 3.2% was allocated to NCD-
specific programmes such as the NCDs National Health
Strategy, despite NCDs being responsible for 66% of all
deaths [23], and only US$8 329 943 was specifically
allocated for the ‘treatment and control of diabetes’
according to the Ministry of Economics and Finance
statistics as part of the Budget for Results Programme.
Since 2011, the NCDs National Health Strategy has
allocated money as part of the Budget for Results Pro-
gramme to finance public programmes for the prevention,
diagnosis and treatment of diabetes and other NCDs
addressed by the National Health Strategy. Reliable infor-
mation about the number of patients and their demands on
medical goods and services is both necessary and relevant
for annual resource targeting under the Budget for Results
Programme.
The total health expenditure per capita was US$356 in
2013, below the average obtained by upper–middle income
countries (US$466), and out-of-pocket expenses represented
35% of the total health expenditure [24,25]. Besides the low
financial coverage for health care, one third of the Peruvian
population did not have any type of health insurance in
2013 according to the National Health Survey. Among those
with insurance, 39% had public insurance (SIS), 23% had
social security and 5.5% had another form of insurance.
Persons with SIS typically come from the lowest socio-
economic sector and use MINSA facilities for medical
attention.
Health insurance coverage among the interviewed patients
with diabetes was 77%, with monthly insurance costs
ranging from US$0 (those with SIS) to US$32.2 (those
patients with private insurance). Among those who reported
not having health insurance, their reasons included: lack of
financial capacity to afford a monthly payment, history of a
756
Delivery of diabetes care in Peru  M. K. Cardenas et al.
‘bad experience’ with the staff or care service, or lack of
interest in being affiliated to an insurance provider. People
with health insurance have access to a basic plan called the
‘Essential Health Insurance Plan’, which contains a minimum
list of conditions that are mandatorily covered by any
insurance. Type 2 diabetes is covered when there are no
major complications (damages to target organs). Armed
Forces and EsSalud, a social security insurance that covers
employees in the formal sector and their beneficiaries, offers
a wider list of conditions covered, including major compli-
cations of Type 2 diabetes.
Health workforce
In 2012, in Peru, there were 33 669 doctors (112 per 100,000
population) and 33 491 nurses (111 per 100 000 population),
according to the Peruvian National Human Resources for
Health Observatory. In comparison, the number of nutrition-
ists was very low, only 1019. Peru has 17 401 practising
medical specialists, most of whom focus on maternity and
infant health. Of the specialists that help manage diabetes,
there were 564 cardiologists, 242 endocrinologists, 1483
internists, 216 nephrologists and 605 ophthalmologists. One
third of the available endocrinologists worked in the private
sector and 8 of 25 regions lacked endocrinologists.
The majority of health workers reported attending con-
ferences, seminars or courses in order to receive training to
manage their patients with diabetes. These events were often
organized or supported by the Peruvian Medical College,
professional societies and universities. The health workers
interviewed reported interest in other training, especially in
preventing and managing diabetes complications. Training
events were also organized by pharmaceutical companies,
specifically on the use of insulin and oral hypoglycaemic
agents. However, the professionals interviewed from the
public sector lamented the few opportunities to participate in
diabetes-related training.
Service delivery
According to the interviewed patients with diabetes, they
were diagnosed between the ages 43–59 years, mostly via a
fasting blood glucose test. One third of the patients were
diagnosed in the emergency department of a public or private
hospital after presenting with a diabetes-related complication.
After their diabetes diagnosis, 77% of the patients received a
prescription for an oral hypoglycaemic and 36% were
prescribed insulin. The medications used most by patients
were metformin and human isophane insulin (NPH insulin).
Patients with diabetes face several problems when seeking
medical attention. Although it was not difficult to find
available primary care appointments, the practitioners them-
selves highlight the insufficient resources (specialty diabetic
care equipment) to provide proper diabetes diagnosis and
control. As a result, many patients denied having had a
ª 2016 Diabetes UK
 Research article DIABETICMedicine

consultation with a general practitioner (GP) within the last referrals. Self-monitoring of blood sugar levels is rare.
year. Most people with diabetes are therefore referred to Patients cannot afford the costs of the equipment and
hospital, creating a burden on secondary and tertiary care consumables, and neither are covered by health insurance.
systems. Given the increased numbers of patients and the The diabetic medicines included on the national list of
reduced numbers of specialists in public facilities, appoint- essential medicines are: glibenclamide 5 mg, metformin
ments are scheduled with long delays. As a social security 850 mg, insulin human regular and NPH insulin. In 2013,
physician stated: ‘there is a shortage of medical specialists, MINSA bought glibenclamide and metformin at US$0.0035
and there is patient need, which is steadily increasing and the and US$0.0131 per tablet and a 10 ml vial of 100 IU/ml
staff that should be attending them are not increasing in NPH insulin at US$2.54 [26]. Data about the availability of
the same proportion’. Because of this mismatch, patient these medications at the studied facilities showed a lack of
waiting-times fluctuated considerably depending on the available insulin and syringes at primary healthcare facilities;
facility. The shortest waiting time was found at the MINSA insulin was only found at hospitals and private pharmacies.
primary healthcare level (73 min on average), followed by Further, the majority of people who sought care at MINSA
private clinics and an armed forces hospital (both close to facilities bought their medicines in pharmacies outside their
80 min). The waiting time in social security facilities was on visited facility. The reasons for this were: the prescription
average 113 min, with one patient waiting up to 225 min. In pills or insulin were not sold by the facility’s pharmacy (e.g.
contrast to the prompt attention granted in the MINSA brand name or combination pills), the existence of a gap
primary care facilities, patients attending at MINSA hospi- between the date of the patients’ next appointment and the
tals reported the longest waiting time, with an average wait expiration date of the last prescription, and a general stock
time of 160 min and a maximum of 315 min. shortage at some times of the month. People requiring insulin
In the private sector, it was not difficult to find an available faced other challenges because syringes and needles were not
appointment with a specialist. However, the high consulta- available, at any point, at MINSA facilities.
tion costs can be prohibitory. The maximum price that The poor availability of medicines relates to problematic
interviewed patients paid for a medical consultation in a planning and purchasing. The quantity of medications and
private clinic was US$17.6. By contrast, people with public supplies to be purchased were estimated according to the
health insurance paid between US$0 and US$2.5 for a effective demand of the previous year. Sometimes, the
consultation. In addition to the consultation fee, transporta- projected demand underestimated actual demand, a probable
tion costs varied among patients; with a lower (US$0.27 one- occurrence when considering the health registry deficiencies
way fare) average estimated transportation cost for those at the patient and facility levels. The acquisition of medica-
who attended a primary healthcare system (MINSA health tions also presented various problems, such as: delays in
posts or centres) compared with those who attended a private determining the referential price, non-compliance of a
clinic or an armed forces hospital (US$1.6 and US$1.1, selected supplier, voided purchase processes due to insuffi-
respectively). cient participants or product failures, requiring additional
The interviewed healthcare workers and patients agreed paperwork for returns or laboratory trials.
that little or no time is given to patient education during the
Table 2 Prices of oral medications and insulin in the public and private
medical consultation. Rather, it was found that patients
sector (2013)
receive little information about healthy lifestyles. The main
healthy lifestyle recommendations received by patients were
(A) Median (B) Median
to lower their intake of sugar and refined carbohydrates, price at price at
increase their consumption of fruit and vegetables, and private public
increase their physical activity. However, interviewees denied pharmacies pharmacies Ratio
Medication (US$) (US$) A/B
receiving any additional instructions or strategies on how to
manage their diabetes; rather, some information was poten- Glibenclamide 5 mg 0.11 0.01 15.0
tially misleading, as one public hospital physician shared: Metformin 850 mg 0.30 0.02 16.6
Glimepiride 2 mg 0.81 0.01 56.5
‘some patients think that they are going to manage diabetes NPH human insulin 19.43 6.88 2.8
only with medications’. (100 IU/ml 10 ml vial)
Insulin human regular 23.60 8.56 2.8
(100 IU/ml 10 ml vial)
Medical products and technologies Insulin lispro 14.21 Not on sale N.A.
(100 IU/ml 10 ml vial)
In primary healthcare systems, several barriers to the Insulin glargine 55.17 Not on sale N.A.
(100 IU/ml 10 ml vial)
diagnosis and control of diabetes relate to a lack reagents
for laboratory tests, e.g. HbA1c, creatinine and microalbumin N.A., not available.
to test average blood sugar levels and renal function. The Estimated with data from: Observatory of Medicine Prices of
current stock of reagents was insufficient to meet monthly the General Directorate of Medication, Supplies and Drugs
(DIGEMID) from the Ministry of Health (MINSA).
demand for laboratory tests, leading to frequent hospital

ª 2016 Diabetes UK 757

DIABETICMedicine
Because of the lack of resource availability, patients who
bought their medications often ended up going to private
pharmacies with considerably higher prices than public
facilities. Table 2 shows that the costs of oral medications
such as metformin 850 mg can be 16.6 times higher at a
private pharmacy compared with a public pharmacy, and
human isophane insulin can cost 2.8 times more at a private
pharmacy.
Access to medication hinders compliance, with many
interviewees reporting private pharmacy costs as their reason
for not adhering to their medication regimen. As stated by a
physician from a private clinic: ‘many patients cannot pay for
the medications they need to take. On those occasions, the
patients take generic or the amount they can afford’.
Discussion
Our assessment was conducted at various levels of the health
system where ongoing efforts by the Peruvian government to
enhance diabetes management in the context of UHC are
being implemented. This study was conducted to identify the
facilitators and barriers to the implementation of compre-
hensive diabetes care in Peru. We found that the Peruvian
healthcare system operates with high patient out-of-pocket
expenses and several limitations that directly affect access to
medicines, regular laboratory check-ups and follow-up
appointments for diabetes, especially at the primary health-
care level. These findings expand on, and potentially explain,
observations made in a diabetes surveillance study conducted
by the Ministry of Health, in which up to 30% of patients
with diabetes attending public hospitals had at least one
micro- or macrovascular complication [20]. Of note, one
third of the patients interviewed in our study reported having
been diagnosed with diabetes after being admitted at the
hospital’s emergency department. This suggests that Peru can
improve its early diagnosis, to help prevent the complications
that cause significant diabetes-related morbidity and
mortality.
Although we found weaknesses in all six building blocks,
we identified the low political commitment assumed by the
Peruvian government to be the main impediment to improv-
ing the health system response to delivery of Type 2 diabetes
care. In addition, three major system-level barriers affecting
diabetes and other related chronic conditions were also
identified, these related to: (1) the availability of information
at different healthcare system levels, (2) insufficient financial
resources, and (3) insufficient human resources. Lack of
information is a clear challenge at the patient, facility and
national levels. This has repercussions for other elements
such as medication supply and the organization of services
for healthcare delivery.
In terms of funding allocation, only 3.2% of the total
public health budget was assigned to NCDs, despite 66% of
total mortality being due to NCDs. Given that UHC aims to
ensure increased service coverage for the population, in the
758
Delivery of diabetes care in Peru  M. K. Cardenas et al.
case of diabetes, this can become a limiting factor consid-
ering the paucity of human resources well trained in the
management of diabetes. Indeed, our findings point towards
a diabetes healthcare delivery model fully based and depen-
dent on specialists who usually work at the tertiary care level
or in the private sector. All of these factors undoubtedly have
major financial consequences for patients who, in many
cases, report having to pay for services and medicines that, in
theory, should be fully covered by the government’s public
health system.
Overall, our results are in line with the assessment of the
health system response for NCDs in LMICs [11] and the
results of past health system assessments in several countries
across Asia, Africa and Central America [27]. Issues around
lack of insulin, trained healthcare workers and organized
delivery of care are all challenges that were found in these
previous studies in LMICs. Specific barriers for each building
block were identified and highlight the challenge of deliver-
ing services for diabetes. Because diabetes has been described
as a tracer condition [28–30], lessons from this assessment
may be relevant to other NCDs in Peru.
Our findings help illustrate some lessons. Increasing health
insurance coverage is not sufficient to ensure equitable access
to quality health care. However, an adequate budget and
finance system are necessary conditions for effective service
delivery that protects people from impoverishment [11]. It is
necessary to develop effective mechanisms of coordination
and articulation among the different sectors and levels of the
healthcare systems [27]. The competencies of GPs and nurses
must be enhanced to manage diabetes at the primary
healthcare level and to provide cost- and time-efficient
patient education to support people in better controlling
their diabetes. It is also necessary to monitor the availability
of the essential drugs, supplies and technologies at the
primary healthcare level, as well as prices of these medicines.
A public registry of medication prices might be a useful tool,
if presented in a patient-friendly format, to reduce asymme-
tries in pricing information. Health facilities and offices need
to enhance their capacities for demand forecasting based on
improved quality health information. For better procure-
ment, the capabilities of healthcare employees in the correct
planning of goods and services require improvement, if they
are to attend to the real demand, and it is also necessary to
strength and supervise the logistical process, providers and
the supply chain.
Some limitations are worth noting from this study. First, a
purposive sampling was implemented for primary data
collection, interviewing health workers, patients and other
key stakeholders from Lima. Although this approach sacri-
fices the representativeness of the sample studied in compar-
ison with more traditional population-based study designs,
we believe that our large sample derived from different
healthcare delivery sectors provides strong evidence about
the situation faced by patients with diabetes in an urban
context in Peru. Healthcare delivery may be worse in peri-
ª 2016 Diabetes UK
 Research article
urban and rural settings. Second, fitting the reporting of our
data and our results to the Building Block Framework is both
a strength and a limitation, because it does not convey the
complexity of health systems and the interaction within and
between building blocks [31]. However, the use of this
simple approach facilitates comparisons with other health
system assessments in similar LMIC settings, and provides a
framework by which to monitor the progress of diabetes care
delivery in Peru.
In summary, our evaluation of the Peruvian healthcare
system to identify the facilitators and barriers that help and/
or hinder the delivery of diabetes care has found low political
commitment, as well as several key limitations that affect the
adequate delivery of diabetes care at the primary healthcare
level. In the contemporary context, in which various LMICs
attempt to implement a UHC despite increasing NCD
prevalence, studies like this one can be helpful in assessing
the readiness, perceptions, responses and progress of the
implementation of UHC with a focus on NCDs.
Funding sources
This study was supported by the Alliance for Healthy Policy
and Systems Research, World Health Organization
(HQHSR1206660).
Competing interests
None declared.
Acknowledgements
The authors appreciate the work performed by the fieldwork
team composed by Dulce Moran, Carlos Huayanay, Sofia
Chumpitaz and Miriam Gonzales, as well as the valuable
time of all the informants that were interviewed of those who
facilitated the authorizations at the different facilities. We
also thank the comments received from different persons
during the development of the study: Maryam Bigdeli,
Rodrigo Carrillo, Lucero Cahuana, Romina Tejada, Jose
Luis Avilez and Elizabeth Abbs.
Author contributions
DB and JJM conceived, designed, obtained funding and
supervised the overall study. MKC implemented study as
project manager. All authors participated in writing of
manuscript, provided important intellectual content and gave
their final approval of the version submitted for publication.
References
1 Atun R, Jaffar S, Nishtar S, Knaul FM, Barreto ML, Nyirenda M
et al. Improving responsiveness of health systems to non-commu-
nicable diseases. Lancet 2013; 381: 690–697.
ª 2016 Diabetes UK
DIABETICMedicine
2 World Health Organization. Global Status Report on Noncommu-
nicable Diseases 2014. Geneva: World Health Organization, 2014.
3 International Diabetes Federation. IDF Diabetes Atlas – sixth
edition, update 2014. Brussels: International Diabetes Federation,
2014.
4 World Health Organization. Global Action Plan for the Prevention
and Control of NCDs 2013–2020. Geneva: World Health Orga-
nization, 2013.
5 United Nations General Assembly. Transforming Our World: the
2030 Agenda for Sustainable Development. Resolution adopted by
the General Assembly on 25 September 2015 [without reference to
a Main Committee (A/70/L.1)] 70/1. Available at http://www.
unfpa.org/sites/default/files/resource pdf/Resolution_A_RES_70_1_
EN.pdf Last accessed 3 October 2015.
6 World Health Organization. Plan of Action – Health Systems
Governance for Universal Health Coverage. Geneva: World Health
Organization, 2014.
7 World Health Organization. The World Health Report: Health
Systems Financing: The Path to Universal Coverage. Geneva:
World Health Organization, 2010.
8 United Nations General Assembly. Global Health and Foreign
Policy. Sixty-seventh session. Agenda item 123. (A/67/L.36) 2012.
9 Jamison DT, Summers LH, Alleyne G, Arrow KJ, Berkley S,
Binagwaho A et al. Global health 2035: a world converging within
a generation. Lancet 2013; 382: 1898–1955.
10 Atun R, de Andrade LOM, Almeida G, Cotlear D, Dmytraczenko
T, Frenz P et al. Health-system reform and universal health
coverage in Latin America. Lancet 2015; 385: 1230–1247.
11 Samb B, Desai N, Nishtar S, Mendis S, Bekedam H, Wright A et al.
Prevention and management of chronic disease: a litmus test for
health-systems strengthening in low-income and middle-income
countries. Lancet 2010; 376: 1785–1797.
12 Atun R. Transitioning health systems for multimorbidity. Lancet
2015; 386: 721–722.
13 Alcalde-Rabanal JE, Lazo-Gonz alez O, Nigenda G. The health
system of Peru. Salud Publica Mex 2011; 53(Suppl 2): s243–s254.
14 Beran D, Higuchi M. How to Investigate Access to Care for
Chronic Noncommunicable Diseases in Low- and Middle-Income
Countries. A survey manual based on a Rapid Assessment Protocol.
Draft for Field Testing 2012. Available at http://apps.who.int/
medicinedocs/en/d/Js20981en/ Last accessed 22 October 2015.
15 World Health Organization. Everybody’s Business. Strengthening
Health Systems to Improve Health Outcomes. WHO’s Framework
for Action. Geneva: World Health Organization, 2007.
16 Instituto Nacional de Estadıstica e Inform atica (INEI). Peru
Enfermedades No Transmisibles y Transmisibles 2014. Lima: INEI,
2015. Available at https://www.inei.gob.pe/media/MenuRecursivo/
publicaciones_digitales/Est/Lib1212/Libro.pdf Last accessed 31
October 2015.
17 Centro Nacional de Alimentaci on y Nutricion (CENAN). Peru
National Survey on Nutritional, Biochemical, Socioeconomic
Indicators Related to Chronic Degenerative Diseases. Lima: Minis-
terio de Salud, 2006. Available at http://www.minsa.gob.
pe/portada/Especiales/2007/nutricion/publicaciones/INFORME_
FINAL_ENIN.pdf Last accessed 31 October 2015.
18 Seclen SN, Rosas ME, Arias AJ, Huayta E, Medina CA. Prevalence
of diabetes and impaired fasting glucose in Peru: report from
PERUDIAB, a national urban population-based longitudinal study.
BMJ Open Diabetes Res Care 2015; 3: 1.
19 Bernabe-Ortiz A, Carrillo-Larco RM, Gilman RH, Checkley W,
Smeeth L, Miranda JJ. Contribution of modifiable risk factors for
hypertension and type-2 diabetes in Peruvian resource-limited
settings. J Epidemiol Community Health 2016; 70: 49–55.
20 Ramos W, L opez T, Revilla L, More L, Huamanı M, Pozo M.
[Results of the epidemiological surveillance of diabetes mellitus in
759
DIABETICMedicine
hospitals in Peru, 2012]. Rev Peru Med Exp Salud Publica 2014;
31: 9–15.
21 Meza E. The impact of the exoneration of tariffs and sales tax on
drug prices. Lima: Red Peruana por una Globalizaci on con
Equidad, Centro Peruano de Estudios Sociales, 2010. Available
at http://www.redge.org.pe/sites/default/files/estudio_impacto_
exoneraciones_web.pdf Last accessed 31 October 2015.
22 Lavilla H. Impoverishment due to out-of-pocket health expen-
diture. Lima: Consorcio de Investigaci on Econ omica y Social,
2012. Available at http://cies.org.pe/sites/default/files/investi
gaciones/empobrecimiento_por_gasto_de_bolsillo_en_salud_0.pdf
Last accessed 31 October 2015.
23 World Health Organization. Noncommunicable Diseases Country
Profiles. Geneva: World Health Organization, 2015.
24 World Bank. The World Development Indicators Database 2013.
Available at http://data.worldbank.org/data-catalog/world-devel-
opment-indicators Last accessed 26 September 2015.
25 World Health Organization. Global Health Expenditure Database
2015. Available at http://apps.who.int/nha/database Last accessed
26 September 2015.
760
Delivery of diabetes care in Peru  M. K. Cardenas et al.
26 Management Science for Health. International Drug Price Indica-
tor Guide. 2014 edition. Medford, MA: Management Science for
Health, 2015.
27 Beran D. The impact of health systems on diabetes care in low
and lower middle income countries. Cur Diabetes Rep 2015; 15:
20.
28 Kessner DM, Kalk CE, Singer J. Assessing health quality – the case
for tracers. N Engl J Med 1973; 288: 189–194.
29 Nolte E, Bain C, McKee M. Diabetes as a tracer condition in
international benchmarking of health systems. Diabetes Care 2006;
29: 1007–1011.
30 Beran D. Health systems and the management of chronic
diseases: lessons from Type 1 diabetes. Diabetes Manag 2012;
2: 323–335.
31 Mounier-Jack S, Griffiths UK, Closser S, Burchett H, Marchal B.
Measuring the health systems impact of disease control pro-
grammes: a critical reflection on the WHO building blocks
framework. BMC Public Health 2014; 14: 278.
ª 2016 Diabetes UK