Burden of disease and costs associated with type 2 diabetes in emerging and established markets: systematic review analyses

ABSTRACT

Objectives: To estimate the clinical and economic burden of type 2 diabetes (T2D) in established (EST) and emerging markets (EMG).

Methods: Three systematic literature reviews were conducted in MEDLINE and Embase to capture all relevant publications reporting 1) the epidemiology of T2D and complications in T2D and 2) the economic burden of T2D and associated complications.

Results: In total, 294 studies were included in this analysis. Evidence indicates a high and increasing overall prevalence of T2D globally, ranging up to 23% in EMG markets and 14% in EST markets. Undiagnosed cases were higher in EMG versus EST markets (up to 67% vs 38%), potentially due to a lack of education and disease awareness in certain regions, that could lead to important clinical and economic consequences. Poor glycemic control was associated with the development of several complications (e.g. retinopathy, cardiovascular diseases and nephropathy) that increase the risk of morbidity and mortality. Direct costs were up to 9-fold higher in patients with vs without T2D-related complications.

Conclusions: The burden of T2D, related complications and inherent costs are higher in emerging versus established market countries. This review explores potential strategies to reduce costs and enhance outcomes of T2D treatment in developing countries.

Keywords:

• Burden of illness
• Emerging market
• Established market
• Cost of illness
• Systematic literature review
• Type 2 diabetes
• Type 2 diabetes complications
• Type 2 diabetes epidemiology
• Type 2 diabetes care
• Economic burden

1. Introduction

Since 1980, the prevalence of type 2 diabetes (T2D) has increased at a proportionately faster rate in emerging markets (EMG; i.e. low- and middle-income countries) than in established markets (EST; i.e. high-income countries) [1,2]. The number of adults living with T2D is estimated to increase from 463.0 million to 700.2 million between 2019 and 2045. The total annual costs of managing this disease are expected to increase accordingly from 760.3 billion USD to 845.0 billion USD in this period [3]. The high cost of managing T2D is particularly challenging for healthcare systems in low- and middle-income countries that have limited infrastructures for the detection and management of the disease. These countries account for less than 10% of the global health expenditure on diabetes but have to cater for the largest proportion of diabetes patients in the world [4-6].

This study was designed to examine the general characteristics and gaps in clinical and economic burden of T2D in EMG and EST markets. Analysis was confined to data published from 15 representative countries based on per capita GDP, growth rate, and standard of living as defined by international indices. To this end, three independent systematic literature reviews (SLRs) were conducted to describe: (1) the epidemiology of T2D, with a focus on prevalence, incidence, and mortality, (2) the epidemiology of complications and comorbidities associated with T2D, with a focus on prevalence and long-term outcomes, and (3) the economic burden of T2D and associated complications, with a focus on direct and indirect costs of patient management and treatment of complications.

2. Methods

2.1. Search strategy and selection criteria

Independent SLRs were conducted between July and August 2018 in accordance with guidelines from the Centre for Reviews and Dissemination [7]. Details of the search strategies for these are described in Supplementary Appendix A. Briefly, EMBASE, MEDLINE, and MEDLINE In-Process were queried to identify relevant publications based on prespecified search criteria. Hand searches were conducted to identify relevant evidence from conference abstracts and posters, and health technology assessments. The search was limited to English and local language publications reported between January 2000 and August 2018. All titles and abstracts identified in the search were screened by two independent researchers against prespecified inclusion/exclusion criteria. Full text articles were also screened independently by both researchers. Any discrepancies that arose during these processes were discussed and resolved between the two researchers and/or a third reviewer when necessary.

2.2 Data extraction and analysis

The following data fields were extracted from the included studies: publication details, study characteristics, patient characteristics, methods of data collection and analysis, detailed results. Epidemiological outcomes data were stratified by geography and analyzed descriptively. Economic data were converted into Euros using exchange rates from the years of the studies and analyzed descriptively.

For the analysis, the following country grouping was applied based on The World Factbook GDP rank and The Morgan Stanley Capital International Emerging Market Index rankings:

• Emerging (EMG) markets: Brazil, China, Philippines, Taiwan, Russia, Saudi Arabia, Turkey, and UAE.

• Established (EST) markets: Australia, Canada, and EU5 (UK, France, Germany, Italy, and Spain).

3. Results

Results of the literature search and subsequent full text review are as follows:

• Epidemiology of T2D: 4,721 unique citations of which 95 publications were included in the narrative synthesis

• Epidemiology of T2D complications: 3,714 unique citations of which 109 were included in the narrative synthesis

• Economic burden of T2D and its complications: 3,990 unique citations of which 90 were included in the narrative synthesis

PRISMA flow diagrams for each of the reviews can be found in Supplementary Appendix B

3.1.1. Epidemiology: prevalence

Overall crude prevalence of T2D was well documented across most countries surveyed, and varied from 5.4% to 23.1% across EMG markets [8,9] and from 4.3% to 13.8% across EST markets (Table 1) [10,11]. Studies have reported increasing prevalence of T2D over time across all countries examined [10,12–16]. Where data were available, regional variations in prevalence were examined and varied between 1.3- and 1.7-fold in EMG markets [17,18] and between 1.4- and 3.0-fold in EST markets [12,19,20]. Large variations within individual countries were reported in Italy (2.4 fold) [20] and France (3.0 fold) [12]. EMG countries reported differences between rural (10.3–25.7%) and urban (14.3–27.2%) regions [21–25]. Details of each country are described in Supplementary Appendix C.

Table 1. Prevalence of Type 2 diabetes in emerging vs established markets

Country	Latest prevalence estimate (%)	Reference year for prevalence estimate	Proportion of undiagnosed (%)	Incidence	Reference year for incidence estimate	Mortality	Reference year for mortality estimate
Emerging markets
Brazil [15,26]	6.9%	2013	46%	Male: 2.45 per 1,000 people Female: 2.46 per 1,000 people	2008	N/A
China [27,28,121]	10.9%	2013	53.6%	Male: 9.2 per 1,000 PY Female: 9.5 per 1,000PY	2010	19.4 per 1,000 PY	1986–2009
Philippines [29–31]	5.4%	2013	66.7%	163 per 1,000 people	1998–2007	0.27 per 1,000 people	2013
Taiwan [43,55]	6.4%	2008	42.8%	9.32 per 1,000 people	2008	Men: 45.9 per 1,000 PY Female: 36.0 per 1,000 PY	2000–2009
Russia [8,56]	5.4%	2013-2015	37%	N/A		0.76 per 1,000 people	2016
Saudi Arabia [9]	23.1%	2011	39.4%	N/A		N/A
Turkey [21,32,51]	18.4%	2007-2012	38.2%	Male: 12.4 per 1000 PY Female: 11.0 per 1000 PY	2003	13.2 per 1,000 PY	1990–2013
UAE [33,54]	10.2%	2006	40.7%	Males: 3.3 per 1000 PY Females: 6.3 per1000 PY	2010	N/A
Established markets
Australia [46,60]	5.7%	2017-2018	35.6%	N/A		Male: 0.19 per 1,000 people Female: 0.13 per 1,000 people	2017
Canada [16,34,35,57]	7.4%	2012	30.5%	Male: 6.8 per 1,000 people Female: 5.7 per 1,000 people	2008/2009	Male: 0.24 per 1,000 people Female: 0,22 per 1,000 people	2004–2008
France [12]	4.8%	2013	37.5%	Male: 1.54 per 1,000 people Female: 1.21 per 1,000 people		N/A
Germany [48,59]	9.9%	2010	34.2%	Male: 16 per 1,000 PY Female: 13 per 1,000 PY	2009–2010	N/A^a
Italy [36,50,61]	5.4%	2015	35.8%	3.8 per 1,000 people	2007	40.3 per 1,000 patients	2007
Spain [11,58]	13.8%	2010	28.4%	N/A		0.21 per 1,000 people	2013
UK [10,13,37]	5.3%	2013	18.5%	Male: 3.99 per 1,000 PY Female: 3.73 per 1,000 PY	2013	Male: 0.09 per 1,000 people Female: 0.13 per 1,000 people	2015

Abbreviations: N/A, not available; PY, person-years

^aon graph

Age distribution of T2D was reported for all countries examined except Spain. Peak prevalence among EMG markets was primarily in the age group 60–69 years, with the exception of Taiwan where peak age was 80 years and older [8,9,15,21,22,38–44]. Among EST markets the peak prevalence was reported in patients aged 75 years and older [12,20,45–50].

The proportion of undiagnosed cases was considerably high among EMG countries (37–66.7%) relative to EST countries (18.5–37.5%) [8,11,18,51,52]. Age distribution of undiagnosed cases was reported for Russia, Turkey, and China, where up to 90% of undiagnosed T2D was reported in the age group of 18–30 years and the lowest prevalence was reported for 65–70 years [8,18,51].

3.1.2. Epidemiology: incidence

Data on the incidence rates of T2D were available for all countries except Australia, Russia, Saudi Arabia, and Spain, with longitudinal data available for Italy, UK, and Taiwan for different periods (Table 1). Incidence rates in the UK were stable from 2008 to 2013, in both sexes (Men, 4.62/1000 PY to 3.99/100 PY; Women, 4.06/1000 PY to 3.73/1000 PY) [10]. In Italy, the crude incidence rate decreased from 381.68/100,000 to 379.21/100,000 between the period 2002–2007 [50]. In contrast, an increase in incidence rates (from 764/100,000 to 932/100,000) was reported in Taiwan between 2000 and 2009 [43]. Only four studies reported age distribution of incident cases: in Taiwan and Germany, the incidence was highest in patients ≥80 years for both sexes [43,53], while in the UAE, peak incidence for T2D was reported at 55–59 years [54].

3.1.3. Epidemiology: mortality due to T2D

Mortality due to T2D in EMG countries varied between 13.2/1,000 PY (Turkey) and 41/1,000 PY (Taiwan) [51,55]. In Russia, mortality was reported at 76.6/100,000 people [56]. Across EST countries, mortality rates varied between 8.9/100,000 (UK) and 23.7/100,000 (Canada) [14,57]. Significant regional variability in mortality has been reported, notably in Spain where the range was between 6.1/100,000 and 82.7/100,000 people in the lowest and highest mortality regions [58]. Age-specific epidemiological data showed an increase in T2D mortality rates with increasing age in Canada (219.6/100,000 for 65–74 years and 1229.9/100,000 for ≥85 years), Germany (in 2010, 16.6% deaths due to diabetes were for 60–69 years and 34.2% for 70–79 years) and Taiwan (184.5/1,000 PY for ages 80–89 and 398.7/1,000 PY for age ≥90 years) [55,57,59]. In contrast, time-trend analyses from Australia, England, and Italy indicate an approximately 5% to 10% overall decrease in mortality rates over the past decade [14,50,60,61]. Despite reductions in mortality rates, evidence from China and France showed that patients with T2D have an elevated risk of mortality compared to the general population [62,63]. In China, individuals with T2D have a two-fold higher risk of all-cause mortality, compared to those without (1373/100,000 vs 646/100,000, respectively) [62]. In France 2007–2012 standardized mortality rate was 1.34 (males) and 1.51 (females), remaining elevated compared to the general population [63].

3.1.4. Epidemiology: complications

Estimates of microvascular and macrovascular complications associated with T2D showed variation within as well as between the countries included in this study (Table 2). Such differences were likely driven by the impact of confounding factors such as age, duration of disease, quality of control observed in the different population cohorts. These differences, however, made it difficult to draw conclusions on differences in patterns of complications between EST and EMG countries. Studies in EST and EMG countries investigated the risk of developing macrovascular and microvascular complications in relation to glycemic control [64–74]. Using a range of HbA1c target levels, (HbA1c >6.5%, >7%, >8%, etc.) these studies demonstrated that poor glycemic control was associated with an increased risk of chronic complications in both EST and EMG populations.

Table 2. Prevalence of Type 2 diabetes complications in emerging and established markets

	Retinopathy	Nephropathy/CKD	Neuropathy	Any CVD	Congestive heart failure	Cerebrovascular event/Stroke
Emerging markets
China [66,75–83,122]	6.5–31.1%^a	4–10%/1% – 8%	5.4–20.4%	13.5%	NA	NA/3.3–6.6%
Taiwan [65,127]	19.5–27.3%	32.8%/16%	21.3–29.7%	33–49 %	10–13%	20–24.8%/2.2%
Russia [105]	12.5–64.1%^b	NA	NA	NA	NA	NA
Saudi Arabia [67,84,85,106,125]	6.3–21.8%^b	11%/23.8%	71%	22.7%	2%	1.8%/0.8%
Brazil [86,118]	45.2–46.2%	NA/9.9%	NA	NA	NA	8%
UAE [44]	54.2%	41%/15.4%	34.7%	NA	NA	NA
Turkey	NA	NA/NA	NA	NA	NA	NA
Established markets
Australia [110]	16.4–22.4%	NA/17–24%	30.8–58.8%^b	28.4%	NA	8.5–10%/NA
Canada [87,88]	27.2%	NA/2–4%	12.50%	NA	8–14%	NA/1.7%
France [89,107]	21–38.2%	7–39%^b/1%	45%	28.3%	3%	1.7%/2.6%
Germany [64,90–92,113]	3.6–20.1%^b	9–14%/NA	5.70–14.4%^b	NA	9.9%	4.6–9.6%
Italy [93]	18.60%	NA/23.5%	30.6%	NA	NA	NA
Spain [74,94–98,108]	7.2–26%^b	10–19%/21% – 23%	4–8.8%	7–29%	5.5%	4.9–7.5%/5.9–7%
UK & Scotland [99–104,111]	14.7–30.3%	0.5–13%^b/7.7–21.1%	2.4–21.7%	10–25.4%	4–9%	7.5–9%/4.6%–9.8%

^ahigher estimate in study which had as a primary objective to document the prevalence of retinopathy; ^bdifference due to variation in disease duration. NA, not available

Retinopathy was one of the most frequently reported complications in T2D patients, with prevalence rates ranging between 6.3% and 64.1% in EMG [105,106] and 3.6–38.2% in EST countries [64,107]. In EST countries, 3–7% of patients presented with retinopathy at the time of diagnosis and up to 90% had retinopathy after 40 years of follow-up [64,108,109]. Time-trend analyses from Australia, China, Russia, and UK suggest that the prevalence of retinopathy increases with the duration of diabetes [105,110–112]. Long-term data from Hong Kong suggest that not only does the prevalence of retinopathy increases over time but also its severity [112]. A study from the German/Austrian Diabetes Prospective Documentation Initiative reported up to 50% of patients with sight threatening retinopathy [113]. Diabetic retinopathy also increases the risk of all-cause mortality, regardless of the stage of T2D, compared with the diabetic individuals without the complication [114]. Retinopathy is also associated with an increased risk of stroke and heart failure [114].

Renal complications such as clinical nephropathy and chronic kidney disease (CKD) also contribute to the clinical burden in patients with T2D. These complications are affected by disease duration and patient age, with prevalence reported between 1%-25% in EMG [115,116], and 5%-20% in EST markets [116,117]. Variations in those numbers have been observed, mostly due to differences in diagnostic practices as well as diseases durations. Furthermore, prevalence of CKD of more than 50% was reported in Brazil among patients of poor sociodemographic status [118] (Supplementary Appendix C). Due to heterogeneity in patient populations and diagnostic practices, no clear trends were observed between EMG and EST markets with respect to CKD or nephropathy epidemiology. Prevalence of nephropathy is known to be influenced by the other co-morbidities such as obesity and hypertension [116,117], the latter being a common and major contributing factor to the development of renal disease in T2D patients [119,120]. Cross-sectional studies from EST and EMG markets which look into the characteristics of CKD patients show similarities in terms of severity distribution, with 60–90% of CKD patients being classed as stage 1 or 2 [121,122]. Progression to end stage renal disease (ESRD) is generally poorly documented, with evidence from France showing that 7% of patients reaching ESRD within 13y of follow-up [123]. Emerging evidence indicates that not all patients progress to ESRD at the same rate and that some (~10%) do so at an accelerated rate [123]. Renal complications have an impact on mortality with significantly reduced kidney function (eGFR of 15–29) being associated with a 3 fold increase in all-cause mortality [124].

The overall prevalence of cardiovascular diseases (CVD) was observed in approximately 20–30% of patients with T2D although outlier figures were noted and attributed to differences in the populations studied [125,126]. Similarly, coronary artery disease (CAD) has been reported in 20%-40% of patients [68,109]. Chronic heart failure (CHF) and myocardial infarction (MI) have been reported in 5%-10% EST and EMG countries with little difference between the two groups. Cerebrovascular and stroke events are also common in T2D patients with reported prevalence in EST and EMG markets of 2%–10%, with variability related to the disease duration [124,125,127,128]. Long-term trend evidence from EST markets indicates that the incidence of stroke among T2D is stable with small year-on-year variation. In the recent years, the incidence rates of stroke have marginally decreased, moving closer to what is observed in the general non-T2D population. Data on peripheral arterial disease (PAD) were reported for China, Saudi Arabia, and Taiwan [129–132], with the latter two having highest estimates of 23%.

Epidemiological data on other complications such as foot ulcers and depression have been relatively well documented in EST countries either through focused follow-up studies or large cross-sectional ones. Detailed data on those have been presented in the supplementary materials (Supplementary Appendix D to Appendix M).

3.2. Costs associated with T2D

This study identified 92 publications (EMG, n=41; EST, n=51) reporting costs in T2D for all countries in our scope except Australia, Philippines, Russia, and UAE. Total annual direct costs per patient were higher in EST markets (€1,279 [Italy] – €6,506 [France])[133,134] compared with EMG markets (€396 [Brazil] to €2,906 [China]) (Tables 3 and Table 4) [135,136]. Within-country variability in reported costs was high in both EST and EMG markets (up to 3.25-fold [Italy] and 4.04-fold [China]) [133,137–139]. Furthermore, studies reported significant within country differences by region, patient complications, glycemic control, and treatment adherence. The variability was primarily due to differences in costing methods and years when the data were collected. Median hospitalization costs (n=17) and medication costs (n=20) represented 42% and 35% of total direct annual costs [135,140], respectively, and were comparable between EST and EMG markets: 36% vs 34% and 32% vs 39%, respectively [140,141]. Lack of glycemic control was reported in both EST and EMG markets and associated with a 1.16- to 2.24-fold increase in total annual direct costs and 1.28- to 3.41-fold increase in drug costs [133,135,142–144].

Table 3. Economic burden of T2D in emerging markets

Country	Direct costs^a	Indirect costs^a	Drug/Medication costs^a	Inpatient costs^a
Brazil[135,145,146,169]	€396 to €1,136	€655	€871	€181 to €232
China[136,138,139,147–153,170,175,178–180,185]	€719 to €2,906	€537 to €5,881	NA	€185 to €3,401
Philippines	NA	NA	NA	NA
Taiwan^a[154,181]	€1,388 to €2,849	NA	€75 to €152	€110 to €328
Russia[155]	NA	NA	€3,506 to €4,919	NA
Saudi Arabia[143]	€1,937	NA	€733 to €1,840	€337 to €918
Turkey[156]	€399 to €466	NA	NA	NA
UAE	NA	NA	NA	NA

^aCosts are presented per patient unless specified otherwise

Table 4. Economic burden of T2D in established markets

Country	Direct costs^a	Indirect costs^a	Drug/Medication costs^a	Inpatient costs^a
Australia	NA	NA	NA	NA
Canada[157]	€4,880	NA	NA	NA
France[134,158]	€2,003 to €6,506	NA	NA	€1,216 to €3,866
Germany[144,159–161,171]	€2,197 to €5,146	€4,103	€498 to €869	€848 to €2,057
Italy[133,137,162,163,176]	€1,279 to €3,532	NA	€385 to €974	NA
Spain[140,142,164,165]	€2,331 to €3,432	NA	NA	€436 to €1,467
UK[166–168]	€2,882 to €3,605	NA	€1,015	€1,430

^aCosts are presented per patient unless specified otherwise

Limited data on indirect costs were obtained from the publications identified by our search strategy. In EMG markets, the indirect costs varied widely (€655 in Brazil and €537 to €5,881 in China) [139,169,170]. In EST markets, one study was identified reporting indirect costs at €4,103 in Germany [171].

This study also identified 43 publications (EMG, n=23; EST, n=20) reporting costs of T2D-associated complications for all countries in our scope except Australia, Philippines, Russia, Saudi Arabia, and UAE.

The total annual direct cost per patient increased with the number of complications (Figure 1), primarily driven by inpatient and medication costs. The mean direct costs per patient associated with incident, acute macrovascular, microvascular, and metabolic complications ranged from €11,548 (acute coronary syndrome, France) [172] to €26,090 (incident cardiovascular disease [CVD], UK[173]) in EST markets and from €1,504 (ischemic heart disease, China [174]) to €8,786 (acute myocardial infarction, China [175]) in EMG markets. The mean direct costs per episode per patient associated with the management of chronic comorbidities ranged from €1,323 (retinopathy, Italy[176]) to €9,497 (established CVD, UK [173]) in EST markets and from €769 (microvascular disease, China [177]) to €4,865 (chronic kidney disease (CKD), China [178]) in EMG markets (Figure 2). An increase in the number and severity of any complications resulted in 1.5- and 5-fold higher direct costs across EST and EMG markets, largely due to higher number of hospitalizations (up to 45% of costs) [176,179–184]. In matched studies, the presence of chronic complications resulted in 2- to 9-fold increase in direct costs compared to patients without complications [181,183]. Specifically, costs were between 1.5- to 3-fold higher due to cardiovascular complications (n=9) and 1.5- to 6.3-fold higher due to renal complications (n=6) [138,173,176–178,184–187].

Figure 1. Mean total annual direct cost per patient in patients with and without the presence of any complication

Figure 2. Mean total annual direct cost per patient in patients with different types of complications Abbreviations: CARD: coronary artery disease; CeVD: cerebrovascular disease; CHF: congestive heart failure; CKD: chronic kidney disease; CVD: cardiovascular disease; MIC: microvascular events; MAC: macrovascular events; NEPH: nephropathy; RETI: retinopathy; PVD: peripheral vascular disease

Cost of hypoglycemic episodes was reported in studies from Spain, Italy, Canada, and China [188–193]. In Spain, the mean annual direct cost of hypoglycemia leading to hospitalization was approximately €11,000 among those who experienced such an event [188]. This was approximately 3 times higher compared to patients who did not experience hypoglycemia. In Italy, hospital episode cost of hypoglycemia varied between €547 and €3,145, depending on the patient age and comorbidity profile [189,190]. In Canada, patients who experienced hypoglycemic events incurred additional annual direct costs of €3,859 per patient [191]. In China, the mean direct costs for moderate and severe hypoglycemic event were estimated at €189 and €695 per episode [193]. Although insulin is the likely cause responsible for most cases of hypoglycemia, details on the type of insulin or other diabetes therapies used were unavailable.

4. Discussion

This review highlights similarities and differences in outcomes and costs of managing T2D in two groups of countries with different economic status. While the prevalence of T2D is increasing worldwide, the overall burden of disease is proportionately higher in EMG countries as these account for the largest number of T2D patients and projected to increase even further [3]. Projections for 2045 suggest that disease burden will be greatest in EMG countries with prevalence estimates of 7.3% - 18.5% (versus 4.8% - 12.6% in EST countries) [3]. Our findings suggest that while prevalence increases with time, incidence and mortality rates tend to be stable or even decrease in the same period, especially in EST markets. The frequency and costs of complications associated with T2D were found to increase with disease duration and patient age.

A striking feature of T2D in EMG markets is the high proportion of undiagnosed cases in younger individuals in contrast to EST countries where it is characteristic of older patients. Large regional variations in the proportions of undiagnosed cases in individual countries may be associated with lack of education and disease awareness. EMG countries are, therefore, well advised to raise public awareness and initiate active screening programs for diabetes at a lower age compared to EST countries [194]. It is often not realized that youth-onset T2D represents prognostically the diabetes form with the worst clinical outcome. An adolescent with T2D has twice the mortality risk and higher rate of complications than someone with T1D or with T2D diagnosed later in life [195]. The high rate of undiagnosed T2D amongst young people uncovered in this study along with the relatively young age distribution observed in EMG populations calls for the institution of urgent measures for the detection, control, and prevention of T2D targeting specifically young people in EMG communities.

Our results on the burden of T2D-associated complications are consistent with findings previously reported. Harding et al. showed that retinopathy is one of the most frequent complications reported in patients with T2D across all geographies with overall estimates of approximately 30%. Our study also found large decreases over time in rates of all-cause mortality among EST markets and marginal, if any, decreases in EMG markets (e.g. China and Taiwan), indicative of a large and growing gap between the outcomes of patients from these countries. While multiple factors are likely to be responsible for these trends, clarity of treatment guidelines, and the quality of patient care are undoubtedly important factors for consideration. To this end, Owolabi et al. found that the majority of national guidelines from low- and middle-income countries were inadequate in terms of applicability, clarity, and dissemination plan and not adequately contextualized for socioeconomic status as compared to high-income countries [196]. The introduction of practical and enforceable national guidelines would likely improve clinical outcomes and narrow the mortality gaps observed between countries of different economic status.

Our analysis also highlights the impact of poor glycemic control on diabetes-associated complications. Control of hyperglycemia constitutes a mandatory objective in clinical practice irrespective of the socio-economic background of the country and underscores the need for effective glucose-lowering therapies in the treatment of people with T2D. As some of the most effective therapies are also the most affordable in the market, effective glycemic control is within the reach of most people with T2D including those in EMG countries.

A review of the economic burden revealed large variations in total annual direct costs per patient between EST (€1,279-€6,506) and EMG markets (€396-€2,906) similar to previous reports [197], and may be attributed to differences in healthcare organization, quality and availability of data as well as methodological heterogeneity. The primary cost drivers in both groups were hospitalization, acute care and long-term care for complications. Hospitalization costs represented approximately 40% of total direct costs. Annual medication costs showed little variation in both groups regardless of the type of therapy.

Our findings are in line with and contribute further to previously published work on the economic burden of T2D. Moucheraud et al. showed significant variations in the total direct cost per patient within and between low- and middle-income countries [198], with inpatient care being the main driver of costs (€18-€900 per patient). Furthermore, the care for complications was associated with a substantially higher cost, which varied widely based on the country and type of complication. Einarson et al. demonstrated that mean annual direct cost per patient with T2D is more than 3-fold higher (€3,097 to €8,796) in those with comorbid CVD vs those without [199]. Long-term care resulted in 2- to 9-fold higher direct costs for patients with T2D with diagnosed complications versus those without [181,183].

Our analyses highlight the substantial cost associated with hypoglycemia similar to other reports [188–191]. The risk of hypoglycemia was noted to be highest amongst elderly patients with T2D especially those with mild elevation of HbA1c levels. The benefits achieved with aggressive glycemic control in elderly people with T2D are offset by the consequences associated with severe hypoglycemia and therefore, de-intensification of treatment is recommended in elderly patients with T2D to ensure clinical safety and reduce hospital costs associated with severe hypoglycemia [192,193].

Our review highlights the complex realities faced by policymakers in EMG markets of striving to contain the relentless increase in number of T2D cases and associated costs with the limited healthcare and economic resources available. Of relevance, the rising cost of diabetes observed in EST markets has not been perturbed by the introduction of new diabetes therapies and technologies [194], calling into question the benefits of any routine adoption of expensive diabetes therapies into EMG markets. Affordable and cost effective medications such as the sulfonylureas provide a rational approach to treatment of T2D in EMG countries, deterring cost without compromising outcome [195]. Additionally, innovative digital healthcare and approaches advocating predictive, preventive, and personalized medicine should be explored as future tools to harmonize diabetes care and reduce the cost burden in all markets considered [196].

This study has several limitations. Differences between EMG and EST markets were difficult to quantify due to heterogeneity in study design, patient populations, and quality of reporting. The estimates of complications associated with T2D were dependent on the duration of disease, which varied across studies. The quality of identified data limits the extent to which valid comparisons can be made or interpreted. Regarding the analysis of economic data, national currencies across 15 countries were converted to describe Euro values, however, no additional adjustments were made to account for variations in local purchasing power parity. This may serve to highlight a wider general limitation in this field, namely the lack of standardized methodology for estimation of cost in T2D [197]. Efforts should be made to address this issue and specific guidelines should be developed in the future. This review focused on a pre-defined list of countries; therefore, the generalizability of our findings may not be extended to other countries or alternative economic regions. Finally, the recommendations proposed here should not be viewed in isolation but addressed within the context of a wider plan to improve living conditions, socio-economic status and environmental wellbeing of the population at large in developing countries.

This review has identified several gaps in our knowledge of T2D and areas for further research which could provide clarity around the overall burden of disease and differences between EMG and EST markets. High-quality evidence is needed on the incidence and economic burden of complications across additional EMG markets. Additional research is required on the clinical and economic impact of undiagnosed disease to better anticipate the future burden of T2D especially in EMG markets where this appears to be of real concern. Finally, there is a lack of data on the economic impact of non-adherence to medical therapy. Non-adherence to medication per se is known to be an independent factor contributing to cost and poor outcome in patients with T2D [198].

5. Conclusions

Our findings highlight existing gaps in the field and alert the reader to new observations that can potentially be utilized to curb future costs and improve health outcomes especially for people with T2D living in EMG countries. The heavy burden of complications and existing trends in epidemiology point to a potentially higher future burden, especially in EMG markets, where the high rates of undiagnosed disease among young people represent a major concern. Our review identified large variations in costs associated with the management of T2D and its complications which may to some extend be addressed through improved quality and methods for collecting data as well as higher quality of reporting. Identifying patients at risk, improving detection and monitoring of T2D patients, and implementing effective healthcare programs to enhance awareness and promote prevention of T2D represent key elements to decreasing the future burden of this disease.

6. Expert opinion

Our study provides an important perspective on the overall burden of T2D and the way it differs between select EST and EMG countries. The combined clinical and economic analysis is critical to understanding the true magnitude of the disease on personal and health system levels. This combined view indicates an especially high burden in EMG countries, driven by increasing prevalence, high levels of underdiagnosis, poor long-term outcomes as well as a cost. Published studies from EST countries suggest that even after the adoption of newer T2D treatments, the cost of care has remained high or even increasing. This brings important questions to policymakers and challenges whether the routine adoption of newer and often expensive diabetes therapies across EMG markets a sustainable and appropriate strategy is to reduce the disease burden in these countries. If the objective is to balance the epidemiological and clinical outcomes against the overall health system economic sustainability, perhaps the focus could be on ensuring higher level of adoption of clinical guidelines that cover prevention, diagnosis, treatment, and monitoring. Furthermore, affordable and established cost-effective medications such as the sulfonylureas are still a rational choice in the treatment of T2D, especially in EMG countries, as they deter cost without compromising clinical outcomes.

Our review on the cost burden of diabetes has exposed two outstanding limitations in the field. First, lack of standardized costing methodologies resulting in considerable variations in outcome between studies and partial estimates of cost (e.g. absence of indirect cost). Second relates to the clinical nature of diabetes as a disease that seldom comes in isolation and is nearly always accompanied by other conditions such as obesity, hypertension, and hyperlipidemia. Perhaps, cost addressing the general metabolic burden might be one approach worth exploring in the future.

Future research into the combined clinical and economic disparity between EMG and EST countries is needed to better quantify the gap between the two. Large scale, single protocol-driven studies covering a bigger variety of geographies could provide a way to control methodology and provide a true outcome comparability. Such studies could also provide over time data on the clinical but also economic impact of the adoption of novel treatments into clinical practice in EMG countries. There is also a need for further investigations into the clinical and economic impact of underdiagnosis as well as non-adherence to medical therapies –know drivers of overall T2D burden. Finally, our review has identified significant gaps in research output from countries such as Russia and Turkey, where diabetes prevalence is rising. Further encouragement is needed for some of these EMG countries to adopt research as a tool of development and to actively conduct local diabetes studies.

This review provides an objective and thorough overview of the T2D burden and highlights the need for further actions targeted at patient care and use of cost-effective interventions to ensure health systems sustainability. Disease management programs and public initiatives focusing on these could improve T2D outcomes, especially in countries where this is most needed.

Reviewers Disclosure

Peer reviewers on this manuscript have no relevant financial relationships or otherwise to disclose.

Declaration of interest

All authors report funding from Servier (analysis sponsored by Servier), during the conduct of the study. MB and VLM are employees of Servier. No authors report personal fees from Servier, during the conduct of the study. No authors report grants from Servier (analysis sponsored by Servier), during the conduct of the study. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Acknowledgments

The authors would like to thank Nathalie Gemayel and Ricardo Lopes for assistance with the SLRs. Medical writing support was provided by Khalid Siddiqui and Irene Asensio Gudina of Amaris Consulting and sponsored by Les Laboratoires Servier and Servier Affaires Medicales.

Supplementary material

Supplemental data for this article can be accessed here.

Additional information

Funding

This study was funded by Les Laboratoires Servier and Servier Affaires Medicales.