diabetes research and clinical practice 1 4 7 (2 0 19 ) 4 7–54

Contents available at ScienceDirect

Diabetes Research
and Clinical Practice
journal homepage: www.elsevier.com/locat e/dia bre s

Diabetes education and health insurance: How they

affect the quality of care provided to people with type 1
diabetes in Latin America. Data from the International
Diabetes Mellitus Practices Study (IDMPS)

Juan José Gagliardino a,*, Jean-Marc Chantelot b, Catherine Domenger b, Hasan Ilkova c,
Ambady Ramachandran d, Ghaida Kaddaha e, Jean Claude Mbanya f, Juliana Chan g,
Pablo Aschner h, on behalf of the IDMPS Steering Committee
a
CENEXA, Center of Experimental and Applied Endocrinology (La Plata National University –National Scientific and Technical
Research Council), La Plata, Argentina
b
Sanofi, Paris, France
c
Istanbul University, Cerrahpasa Medical Faculty, Department of Internal Medicine, Division of Endocrinology Metabolism and
Diabetes, Turkey
d
India Diabetes Research Foundation, Dr. A. Ramachandran’s Diabetes Hospitals, Chennai, India
e
Consultant & Head of Diabetology Unit, Government of Dubai, Dubai Health Authority, Dubai, United Arab Emirates
f
Biotechnology Center, Doctoral School of Life Sciences, Health and Environment, and Faculty of Medicine and Biomedical Sciences,
University of Yaounde I, Yaounde, Cameroon
g
Department of Medicine and Therapeutics, The Chinese University of Hong Kong, The Prince of Wales Hospital, Shatin, Hong Kong
Special Administrative Region
h
Javeriana University School of Medicine and San Ignacio University Hospital, Bogotá, Colombia

A R T I C L E I N F O A B S T R A C T

Article history: Aims: This study aimed to evaluate the impact of diabetes education and access to
Received 9 May 2018 healthcare coverage on disease management and outcomes in Latin America.
Accepted 9 August 2018 Methods: Data were obtained from a sub-analysis of 2693 patients with type 1 diabetes mel-
Available online 15 August 2018 litus recruited from 9 Latin American countries as part of the International Diabetes Melli-
tus Practices Study (IDMPS), a multinational, observational survey of diabetes treatment in
developing regions.
Keywords:
Results: Results from the Latin American cohort show that only 25% of participants met
Diabetes education
HbA1c target value (< 7% [53 mmol/mol]). Attainment of this target was significantly higher
Management
among participants who had received diabetes education than those who hadn’t (28% vs.
Observational study
19%, p < 0.001), and among those who practiced self-management (27% vs. 21% no self-
Type 1 diabetes mellitus
management, p = 0.001). Multivariate analysis showed that participants who had received
Treatment
diabetes education were more likely to manage their diabetes (OR: 1.65 [95% CI: 1.24,

* Corresponding author at: CENEXA, Center of Experimental and Applied Endocrinology, (UNLP-CONICET), Facultad de Ciencias
Médicas UNLP, Calle 60 y 120, La Plata, Argentina.
E-mail addresses: cenexaar@yahoo.com.ar (J.J. Gagliardino), Jean-Marc.Chantelot@sanofi.com (J.-M. Chantelot), Catherine.
Domenger@sanofi.com (C. Domenger), ilkova@superonline.com (H. Ilkova), ramachandran@ardiabetes.org (A. Ramachandran),
ghaida_kaddaha@hotmail.com (G. Kaddaha), jcmbanya@yahoo.co.uk (J. Claude Mbanya), jchan@cuhk.edu.hk (J. Chan), pabloaschner@
gmail.com (P. Aschner).
https://doi.org/10.1016/j.diabres.2018.08.007
0168-8227/Ó 2018 The Authors. Published by Elsevier B.V.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
48 diabetes research and clinical practice 1 4 7 ( 2 0 1 9 ) 4 7 –5 4

2.19]; p = 0.001), and to attain HbA1c target values (OR: 1.48 [95% CI: 1.14, 1.93]; p = 0.003).
Conclusions: Given the association between uncontrolled diabetes and long-term complica-
tions, health authorities and care providers should increase efforts to ensure widespread
healthcare coverage and access to self-management education to reduce the socioeco-
nomic and humanistic burden of type 1 diabetes.
Ó 2018 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-
NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction ment and patterns of care across time in developing regions,

and is conducted on a yearly basis in real-world settings.
Care for people with diabetes and its chronic complications Using data from these surveys, we have reported multidimen-
represents a substantial social and economic burden for sional factors related to glycemic control including clinical
national healthcare systems and society overall [1]. Health- characteristics, healthcare coverage, care processes and
care systems in Latin America face the challenge of increas- access to diabetes education [11]. Since its initiation in 2005,
ing prevalence of chronic diseases such as diabetes and we have conducted six waves, each within a 12-month period,
hypertension without a parallel growth of their budget [2]. enrolling patients from diverse clinical settings. Data from
In 2013, care for adults with diabetes in South and Central the first wave showed that people with T1DM were more
America accounted for approximately 13% of total healthcare likely to attain HbA1c if they practiced self-monitoring of
expenditure [3]. In order to optimize the use of finite blood glucose (SMBG) and had access to diabetes education
resources, a shift towards preventative medicine is required [11,12]. For the current analysis, we focused on people with
[2]. T1DM in Latin America from four successive waves and
Within the last decade, most Latin American countries explored associations between glycemic control and diabetes
have introduced new schemes of universal health insurance education, self-management and healthcare coverage.
which have substantially improved access to healthcare ser-
vices [2]. Despite these schemes, the quality of diabetes care
in Latin America remains suboptimal. Many patients were 2. Materials and methods
not monitored for risk factors and complications at the rec-
ommended intervals while fasting blood glucose, HbA1c, The design and objectives of the IDMPS study have been
triglyceride and cholesterol levels are generally outside target described previously [11]. Briefly, IDMPS is an observational,
ranges [4]. Furthermore, access to insulin in the region is multinational study to assess the therapeutic management
often limited and, even when available, it is not always pre- of people with diabetes in real-world medical practice. The
scribed and used appropriately, and most patients do not study was conducted in six waves (Wave 1: 2005; Wave 2:
meet glycemic targets [5]. 2006; Wave 3: 2008; Wave 4: 2010; Wave 5: 2011–12; Wave 6:
Care for people with type 1 diabetes mellitus (T1DM) pre- 2013–14), each of which included a cross-sectional survey.
sents a particular challenge since the provision of multidisci- Data from the Latin American cohort, Waves 1–4, are reported
plinary care, essential for management of this complex here.
disease, is scarce in Latin America [5]. Effective control of
T1DM requires patients’ active participation in making day-
2.1. Study implementation
to-day decisions related to the control and treatment of their
disease. In order to do so effectively, they need to understand
The study was coordinated by Sanofi-Diabetes Intercontinen-
how to adjust their insulin doses in response to self-
tal and a steering committee of international diabetologists.
measured blood glucose levels, carbohydrate intake, exercise
Ethics approval was obtained from institutional review boards
load and illness. It has been shown that diabetes education is
in each participating country and the study was conducted in
the most appropriate strategy to promote effective self-
accordance with the Declaration of Helsinki. All participants
management [6,7]. Several reports have demonstrated the
provided written informed consent.
beneficial effect of education for people with type 2 diabetes
mellitus (T2DM) in Latin America [8–10] but data are scarce
for T1DM. 2.2. Selection of centers/physicians and sample size
In order to address this lack of information, we have ana- estimation
lyzed data from the Latin American cohort of the Interna-
tional Diabetes Mellitus Practices Study (IDMPS). Since 2005, Participating investigators included endocrinologists, dia-
the International Diabetes Mellitus Practices Study (IDMPS) betologists and general practitioners with experience in initia-
has been seeking to understand the challenges of managing tion and titration of insulin therapy [11]. Investigators/centers
diabetes in the real world. IDMPS is the largest ever observa- for each study wave were selected independently and investi-
tional study program that describes patient profiles, manage- gators could participate in more than one wave. Sample sizes
 diabetes research and clinical practice 1 4 7 ( 2 0 1 9 ) 4 7 –5 4 49

were determined for each country to attain the primary study 3. Results
endpoint, which was to establish the percentage of people with
T2DM treated with insulin. Across Waves 1–4, a total of 2693 participants with T1DM were
recruited in Latin America (Table 1). Almost all participants
2.3. Participants (96%) lived in an urban setting, and 70% were recruited by dia-
betes specialists. Most participants (56%) attended clinics that
Physicians enrolled the first five adults (aged  18 years) with cared for a mixture of public and private patients.
T1DM and ten adult patients with T2DM who attended their Of the patients recruited, 44% were male, with a mean age
clinic during the 2-week recruitment period. Patients only of 38 years (standard deviation [SD]: 16 years). Mean time
participated in one wave. Patients who were actively partici- since diagnosis of T1DM was 14.1 years (SD: 10.7 years).
pating in another clinical study, or were receiving temporary Almost half (48%) of the participants received university or
insulin treatment (e.g. for gestational diabetes or pancreatic higher level education. Overall, 83% of participants were cov-
cancer) were excluded. ered by health insurance.
Basal plus prandial insulin was the most frequently used
2.4. Data collection and outcome measures treatment regimen (65%) while 24% used basal insulin alone.
Approximately half (52%) of participants who used a basal
Before each study wave, attributes of the participating inves- plus prandial regimen used analog insulin. Basal plus pran-
tigator and center were recorded, including their specialist dial insulin regimen was more frequently used by partici-
status, years of experience, nature of healthcare organization pants with healthcare coverage than those without
and medical coverage. All patient data were collected on case insurance (68% vs. 53%, p < 0.001).
report forms which included demographics, socio-economic
profile, types of diabetes, disease duration, co-existing com- 3.1. Diabetes education
plications and cardiovascular risk factors, glycemic control,
history of hypoglycemia, frequency of physical activity, cur- The majority of patients (65%) had received diabetes educa-
rent insulin treatment regimen. Patients’ attendance of dia- tion, mainly on an individual basis. Attendance to diabetes
betes education programs (including type and modality of education was more common among participants recruited
education received) and self-care practices (including SMBG by diabetes specialists than those recruited by a general prac-
and insulin dose self-adjustment [ISA]) were also recorded. titioner, (67% vs. 61%, p = 0.009) and among those with health
Self-management was defined as practice of both SMBG and insurance (67% vs. 56% among those without insurance,
ISA. Glycemic control target was defined as HbA1c < 7% p < 0.001). Rates of diabetes education differed between coun-
(53 mmol/mol). tries with the highest rate reported in Chile (82%) and the low-
est in the Dominican Republic (29%, Table 1).
2.5. Statistical analysis
3.2. Diabetes self-management activities
Unless specified, data from all waves were pooled for analysis.
For variables with two modalities, Wilcoxon signed-rank Most participants (82%) performed SMBG, but only 63% prac-
(quantitative variables) or Chi-squared (qualitative variables) ticed ISA while 58% of participants practiced both SMBG and
tests were used; for variables with more than two modalities, ISA (i.e. self-management). Rates of diabetes self-
Kruskal-Wallis (quantitative variables) or Fisher’s exact (qual- management differed between countries with the highest
itative variables) tests were used. Univariate and logistic rate reported in Chile (84%) and the lowest in the Dominican
regression analyses were performed to identify predictive fac- Republic (29%).
tors for: self-management, receipt of diabetes education and Self-management (SMBG + ISA) was more common among
glycemic control. For the logistic regression, age was divided participants who had received diabetes education (65% vs.
into three classes: < 40 years old, 40–64 years old and 46% of participants without diabetes education, p < 0.001),
 65 years old. Continuous variables included in the model with health insurance (62% vs. 40% without insurance,
were: total daily insulin dose, time since diagnosis, time on p < 0.001) and among those recruited by a diabetes specialist
insulin treatment and waist circumference. (68% vs. 35% of participants recruited by a general practi-
All predictors with a p-value < 0.20 in univariate analysis tioner, p < 0.001).
were included in a logistic regression model. Then, a stepwise
procedure was used to select the most relevant model. Start- 3.3. Glycemic control
ing from a full model with all independent variables selected
based on the univariate analysis, all non-significant variables Overall, 25% of participants met HbA1c target (< 7%
were removed one by one until all parameters reached a level [53 mmol/mol]) and 28% had HbA1c > 9% (75 mmol/mol).
of significance of at least 0.05. Interactions between indepen- Rates of HbA1c target attainment were similar in participants
dent variables were not considered. Odds ratios were pro- managed by a specialist or a general practitioner (24.7% vs.
vided with 95% confidence intervals. 24.9%, respectively; p = 0.932). Target attainment was numer-
In all data analyses, participants with missing data were ically higher among participants with health insurance cover-
not considered when reporting proportions of participants age than those without (26% vs. 21%, respectively; p = 0.061;
in categories described. Fig. 1).
50 diabetes research and clinical practice 1 4 7 ( 2 0 1 9 ) 4 7 –5 4

Table 1 – Participant baseline characteristics, overall and according to diabetes education.

Overall (N = 2693)a Diabetes education status Significance

N = 2659b (test used)

Educated Not educated

n = 1735 (65%) n = 924 (35%)

Country, n (%) N = 2693 N = 1735 N = 924 < 0.001 (F)

Argentina 830 (31) 519 (64) 295 (36)
Chile 119 (4) 98 (82) 21 (18)
Colombia 450 (17) 345 (78) 98 (22)
Dominican Republic 49 (2) 14 (29) 34 (71)
Ecuador 30 (1) 24 (80) 6 (20)
Guatemala 55 (2) 30 (57) 23 (43)
Mexico 733 (27) 431 (59) 300 (41)
Panama 13 (1) 8 (62) 5 (38)
Venezuela 414 (15) 266 (65) 142 (35)
Ethnicity, n (%) N = 2287 N = 1473 N = 791 0.179 (F)
Caucasian 728 (32) 459 (31) 257 (33)
Native Latin 1492 (65) 961 (65) 520 (66)
Black 8 (< 1) 5 (< 1) 3 (< 1)
Japanese 2 (< 1) 2 (< 1) 0
Oriental, Arab, Persian 1 (< 1) 1 (< 1) 0
Other Asian 3 (< 1) 2 (< 1) 1 (< 1)
Other 53 (2) 43 (3) 10 (1)
Gender, n (%) N = 2625 N = 1697 N = 894
Male 1143 (44) 743 (44) 387 (43) 0.809 (C)
Mean age, years (SD) 37.7 (15.8) 36.7 (15.4) 39.4 (16.3) < 0.001 (W)
Mean time since diabetes diagnosis, years (SD) 14.1 (10.7) 14.5 (10.7) 13.3 (10.6) 0.002 (W)
Mean body mass index, kg/m2 (SD) 24.7 (4.5) 24.4 (4.1) 25.1 (5.1) 0.027 (W)
Physician specialty, n (%) N = 2536 N = 1647 N = 889 0.009 (C)
Diabetes specialist 1782 (70) 1186 (72) 596 (67)
General practitioner 754 (30) 461 (28) 293 (33)
Covered by health insurance, n (%) N = 2644 N = 1705 N = 907 < 0.001 (C)
2207 (83) 1461 (86) 719 (79)
Glycemic control, n (%) N = 2188 N = 1495 N = 688 < 0.001 (C)
HbA1c < 7% 546 (25) 412 (28) 128 (19)
Insulin regimen, n (%) N = 2602 N = 1670 N = 898 < 0.001 (C)
Basal alone 618 (24) 324 (19) 285 (32)
Basal + prandial 1701 (65) 1181 (71) 498 (54)
Prandial alone 39 (2) 31 (2) 8 (1)
Premix alone 172 (7) 94 (6) 76 (8)
Others 72 (3) 40 (2) 31 (3)
Mean number of daily injections by insulin regimen (SD) < 0.001 (C)
Basal alone 1.57 (0.55) 1.62 (0.52) 1.52 (0.57) 0.016 (W)
Basal + prandial 4.18 (0.97) 4.21 (0.97) 4.09 (1.00) 0.031 (W)
Prandial alone 2.42 (0.90) 2.46 (0.78) 2.33 (1.21) 0.853 (W)
Premix alone 2.10 (0.46) 2.12 (0.42) 2.08 (0.51) 0.610 (W)
Diabetes management strategy used, n (%)
No self-management N = 2584 N = 1688 N = 896 < 0.001 (C)
333 (13) 150 (9) 183 (20)
Self-monitoring blood glucose (SMBG)c N = 2655 N = 1715 N = 910 < 0.001 (C)
2186 (82) 1499 (87) 667 (73)
Self-management (practices both SMBG and ISA) N = 2648 N = 1706 N = 912 < 0.001 (C)
1538 (58) 1109 (65) 416 (46)
C, Chi-squared test; F, Fisher exact test; W, Wilcoxon test.
a
Participants with missing data were not considered when reporting proportions of participants in categories listed.
b
Data on diabetes education status were not available for 34 participants.
c
Includes all patients who practice SMBG, some of whom are included among those who practice self-management.
 diabetes research and clinical practice 1 4 7 ( 2 0 1 9 ) 4 7 –5 4 51

Many studies have confirmed the utility of diabetes educa-

tion in empowering patients with T1DM to take effective con-
trol of their disease with clinical, metabolic and economic
benefits [19,20]. Based on this large body of evidence, wide-
spread implementation of diabetes education programs is
recommended by a number of national organizations
[20,21]. Supporting these recommendations, our data showed
that participants who had received diabetes education were
more likely to practice self-management and to attain HbA1c
values < 7% (53 mmol/mol), than those without education. In
the Middle East population of IDMPS, self-management was a
significant independent predictor of glycemic control [22].
This was not the case in the current study, though glucometer
availability was a predictor of glycemic control, and this was
Fig. 1 – HbA1c percentage distribution according to health
also strongly linked with self-monitoring of blood glucose:
insurance. NS, not significant.
98% of patients who owned a glucometer practiced SMBG
compared with 1% of those without a glucometer. Participants
HbA1c target attainment was significantly higher among who had received diabetes education were also more likely to
those with diabetes education (28% vs. 19% without diabetes receive basal and prandial insulin, than those without who
education, p < 0.001), and those who practiced self- most frequently received basal alone; the use of prandial
management (27% vs. 21% no self-management, p = 0.001). insulin only will likely be a barrier for the achievement of
Diabetes education was also important for effective self- HbA1c target.
management practice. In those who practiced self- In this Latin American population, HbA1c target attain-
management, the rate of HbA1c target attainment was ment was higher among participants with health insurance
significantly higher among those who had also received coverage than in those without. Furthermore, participant
diabetes education than those who had not (30% vs. 22%, health insurance coverage was a significant independent pre-
respectively; p = 0.003). Conversely, HbA1c values > 9% dictor of self-management suggesting that costs associated
(75 mmol/mol) were significantly more frequently recorded with diabetes care may deter people without insurance from
in participants without diabetes education than in those managing their diabetes and thus attaining appropriate gly-
who had received it (36% vs. 25%, respectively; p < 0.001). cometabolic control. In fact, test strips for SMBG account for
a substantial proportion of overall diabetes care costs in Latin
3.4. Multivariate analysis America and attainment of HbA1c target was associated with
greater strip use [23]. Although there is strong evidence that
After controlling for confounders, lack of complications, intensive treatment for people with T1DM is cost-effective
lower insulin dosage (< 1 unit/kg), receipt of diabetes educa- overall [24], self-management without appropriate diabetes
tion and having a glucometer were independent predictors education can incur substantial costs (e.g. through increased
for attaining HbA1c target (Fig. 2A). Predictors for diabetes use of SMBG strips), without any improvement in metabolic
education included having a glucometer and seeing a special- outcomes. This is illustrated by our finding that the rate of
ist (Fig. 2B). Independent predictors for self-management target attainment in participants who practiced self-
included younger age, higher education level, treatment with management without diabetes education was similarly low
basal-bolus regimen, long time since diagnosis, having pri- to the rate in participants who did not practice any self-
vate medical insurance and access to diabetes education management (22% vs. 21%, respectively). Thus, diabetes edu-
and seeing a specialist (Fig. 2C). Odds ratios for these effect cation may help patients to optimize their use of SMBG strips
sizes ranged from 1.1 to 1.82 for positive associations and and learn how to carry out ISA in order to maximize the cost-
from 0.28 to 0.92 for negative associations (Fig. 2A–C). effectiveness of self-management.
Participants recruited by diabetes specialists were more
4. Discussion likely to have received diabetes education than those
recruited by general practitioners. This indicates one of the
In this analysis of a large cohort of people with T1DM from benefits of specialist care, and suggests that efforts may be
Latin America, we confirm the multidimensional nature of needed in the region to promote the importance of diabetes
factors determining attainment of HbA1c target; factors iden- education to general practitioners in the region.
tified included insulin dosage, lack of complications, access to In summary, these findings from Waves 1–4 in Latin Amer-
diabetes education and to a glucometer: the latter being ica, and those reported previously from the IDMPS study for
essential tools for SMBG and ISA. Our analysis also highlights other waves and regions, have proved the strong relationships
a need to improve glycometabolic control for people with between diabetes education, self-management and attain-
T1DM in Latin America, since only 25% of participants ment of HbA1c target [11,22]. Our results suggest that health
attained HbA1c target values (< 7.0% [53 mmol/mol]). Poor authorities, policymakers, insurers, healthcare administra-
metabolic control is associated with increased risk for the tors and providers should increase efforts to ensure wide-
development and progression of chronic complications of spread healthcare coverage and access to education about
T1DM [13–16] and a decrease in quality of life [17,18]. diabetes self-management in order to decrease the heavy
52 diabetes research and clinical practice 1 4 7 ( 2 0 1 9 ) 4 7 –5 4

Fig. 2 – (A) Predictive factors for HbA1c target attainment. (B) Predictive factors for receipt of diabetes education. (C) Predictive
factors for diabetes self-management. *‘‘Others’’ includes any regimen other than Basal + Prandial, Basal alone, Prandial
alone or Premix alone. CI, confidence interval; OR, odds ratio.
 diabetes research and clinical practice 1 4 7 ( 2 0 1 9 ) 4 7 –5 4 53

burden of T1DM for patients, the healthcare budget and soci- emergency treatment lead to reduced costs after structured
ety overall. education in adults with Type 1 diabetes. Diabet Med
2014;31:847–53. https://doi.org/10.1111/dme.12441.
[7] Lemozy-Cadroy S, Crognier S, Gourdy P, Chauchard MC, Chale
Acknowledgments JP, Tauber Dagger JP, et al. Intensified treatment of type 1
diabetes: prospective evaluation at one year of a therapeutic
The authors thank the physicians and patients who partici- patient education programme. Diabetes Metab
pated in the study. 2002;28:287–94.
[8] Gagliardino JJ, Arrechea V, Assad D, Gagliardino GG, Gonzalez
L, Lucero S, et al. Type 2 diabetes patients educated by other
Funding patients perform at least as well as patients trained by
professionals. Diabetes Metab Res Rev 2013;29:152–60.
This study was funded by Sanofi. The sponsor was supported https://doi.org/10.1002/dmrr.2368.
by the steering committee regarding study design and registry [9] Gagliardino JJ, Etchegoyen G. A model educational program
structure, and proposed decisions regarding protocol amend- for people with type 2 diabetes: a cooperative Latin American
ments, analyses and publications The authors acknowledge implementation study (PEDNID-LA). Diabetes Care
medical writing and editorial assistance provided by 2001;24:1001–7.
[10] Gagliardino JJ, Lapertosa S, Pfirter G, Villagra M, Caporale JE,
Paul O’Regan, PhD of Fishawack Communications Ltd, whose
Gonzalez CD, et al. Clinical, metabolic and psychological
service was funded by Sanofi. outcomes and treatment costs of a prospective randomized
trial based on different educational strategies to improve
Disclosures diabetes care (PRODIACOR). Diabet Med 2013;30:1102–11.
https://doi.org/10.1111/dme.12230.
CD and JMC are employees of Sanofi. [11] Chan JC, Gagliardino JJ, Baik SH, Chantelot JM, Ferreira SR,
Hancu N, et al. Multifaceted determinants for achieving
All of the other authors are members of the IDMPS Steer-
glycemic control: the International Diabetes Management
ing Committee and have received honoraria and traveling Practice Study (IDMPS). Diabetes Care 2009;32:227–33. https://
sponsorships in relation to the IDMPS. No other potential con- doi.org/10.2337/dc08-0435.
flicts of interest relevant to this article were reported. [12] Gagliardino JJ, Aschner P, Baik SH, Chan J, Chantelot JM,
Ilkova H, et al. Patients’ education, and its impact on care
outcomes, resource consumption and working conditions:
Author contributions
data from the International Diabetes Management Practices
Study (IDMPS). Diabetes Metab 2012;38:128–34. https://doi.
All authors interpreted the results, revised the manuscript, org/10.1016/j.diabet.2011.09.002.
and approved the final version of the manuscript. JJG is the [13] Group TDCaCTR. The effect of intensive treatment of
guarantor of this work and, as such, had full access to all diabetes on the development and progression of long-term
the data in the study and takes responsibility for the integrity complications in insulin-dependent diabetes mellitus. N Engl
of the data and the accuracy of the data analysis. J Med 1993;329:977-86. http://dx.doi.org/10.1056/
NEJM199309303291401.
[14] Nathan DM. The diabetes control and complications trial/
epidemiology of diabetes interventions and complications
R E F E R E N C E S study at 30 years: overview. Diabetes Care 2014;37:9–16.
https://doi.org/10.2337/dc13-2112.
[15] Orchard TJ, Nathan DM, Zinman B, Cleary P, Brillon D,
[1] Barcelo A, Aedo C, Rajpathak S, Robles S. The cost of diabetes Backlund JY, et al. Association between 7 years of intensive
in Latin America and the Caribbean. Bull World Health Organ treatment of type 1 diabetes and long-term mortality. JAMA
2003;81:19–27. 2015;313:45–53. https://doi.org/10.1001/jama.2014.16107.
[2] Arredondo A. Type 2 diabetes and health care costs in Latin [16] Schnell O, Cappuccio F, Genovese S, Standl E, Valensi P,
America: exploring the need for greater preventive medicine. Ceriello A. Type 1 diabetes and cardiovascular disease.
BMC Med 2014;12:136. https://doi.org/10.1186/s12916-014- Cardiovasc Diabetol 2013;12:156. https://doi.org/10.1186/
0136-z. 1475-2840-12-156.
[3] International Diabetes Federation. Diabetes atlas, 6th ed., [17] Fisher L, Polonsky WH, Hessler DM, Masharani U, Blumer I,
Regional Overviews, http://www.idf.org/sites/default/files/ Peters AL, et al. Understanding the sources of diabetes
EN_6E_Ch3_Regional_Overviews.pdf; 2013 [accessed 24 April distress in adults with type 1 diabetes. J Diabetes
2014]. Complications 2015;29:572–7. https://doi.org/10.1016/j.
[4] Commendatore V, Dieuzeide G, Faingold C, Fuente G, Lujan D, jdiacomp.2015.01.012.
Aschner P, et al. Registry of people with diabetes in three [18] Joensen LE, Almdal TP, Willaing I. Associations between
Latin American countries: a suitable approach to evaluate the patient characteristics, social relations, diabetes
quality of health care provided to people with type 2 diabetes. management, quality of life, glycaemic control and
Int J Clin Pract 2013;67:1261–6. https://doi.org/10.1111/ emotional burden in type 1 diabetes. Prim Care Diabetes
ijcp.12208. 2015. https://doi.org/10.1016/j.pcd.2015.06.007.
[5] Aschner P, Aguilar-Salinas C, Aguirre L, Franco L, Gagliardino [19] Boren SA, Fitzner KA, Panhalkar PS, Specker JE. Costs and
JJ, de Lapertosa SG, et al. Diabetes in South and Central benefits associated with diabetes education: a review of the
America: an update. Diabetes Res Clin Pract 2014;103:238–43. literature. Diabetes Educ 2009;35:72–96. https://doi.org/
https://doi.org/10.1016/j.diabres.2013.11.010. 10.1177/0145721708326774.
[6] Elliott J, Jacques RM, Kruger J, Campbell MJ, Amiel SA, Mansell [20] Heller S, Lawton J, Amiel S, Cooke D, Mansell P, Brennan A,
P, et al. Substantial reductions in the number of diabetic et al. Improving management of type 1 diabetes in the UK:
ketoacidosis and severe hypoglycaemia episodes requiring the Dose Adjustment For Normal Eating (DAFNE) programme
54 diabetes research and clinical practice 1 4 7 ( 2 0 1 9 ) 4 7 –5 4

as a research test-bed A mixed-method analysis of the management on the quality of care for people with type 1
barriers to and facilitators of successful diabetes self- diabetes in the Middle East (The International Diabetes
management, a health economic analysis, a cluster Mellitus Practices Study, IDMPS); 2016 [in preparation].
randomised controlled trial of different models of delivery of [23] Elgart JF, Gonzalez L, Prestes M, Rucci E, Gagliardino JJ.
an educational intervention and the potential of insulin Frequency of self-monitoring blood glucose and attainment
pumps and additional educator input to improve outcomes. of HbA1c target values. Acta Diabetol 2015. https://doi.org/
Southampton (UK); 2014. 10.1007/s00592-015-0745-9.
[21] Funnell MM, Brown TL, Childs BP, Haas LB, Hosey GM, Jensen [24] Li R, Zhang P, Barker LE, Chowdhury FM, Zhang X. Cost-
B, et al. National standards for diabetes self-management effectiveness of interventions to prevent and control diabetes
education. Diabetes Care 2012;35(Suppl 1):S101–8. https://doi. mellitus: a systematic review. Diabetes Care 2010;33:1872–94.
org/10.2337/dc12-s101. https://doi.org/10.2337/dc10-0843.
[22] Gagliardino JJ, Chantelot JM, Domenger C, Aschner P, Chan J,
Ilkova H, et al. Impact of diabetes education and self-