Abstract
Aims
To describe glycaemic control and diabetes management in adults with type 1 diabetes (T1DM), in a real‐life global setting.
Materials and Methods
Study of Adults' GlycEmia (SAGE) was a multinational, multicentre, single visit, noninterventional, cross‐sectional study in adult patients with T1DM. Data were collected at a single visit, analysed according to predefined age groups (26–44, 45–64 and ≥65 years) and reported across different regions. The primary endpoint was the proportion of participants achieving HbA1c less than 7.0 % in each age group. Secondary endpoints included incidence of hypoglycaemia, severe hypoglycaemia and severe hyperglycaemia leading to diabetic ketoacidosis (DKA) and therapeutic management of T1DM.
Results
Of 3903 included participants, 3858 (98.8%) were eligible for the study. Overall, 24.3% (95% confidence interval [CI]: 22.9–25.6) of participants achieved the glycaemic target of HbA1c less than 7.0 %, with more participants achieving this target in the 26–44 years group (27.6% [95% CI: 25.5–29.8]). Target achievement was highest in Eastern and Western Europe, and lowest in the Middle East. The incidence of hypoglycaemia and of severe hyperglycaemia leading to DKA tended to decrease with age, and varied across regions. Age and regional differences were observed in therapeutic management, including types of device/insulin usage, frequency of insulin dose adjustment and technology usage.
Conclusions
Glycaemic control remains poor in adults with T1DM globally. Several areas of treatment may be optimised to improve outcomes, including supporting patient self‐management of insulin therapy, increasing use of technologies such as CGM, and greater provision of healthcare support.
Keywords: adults, clinical practice, global, glycaemic control, hypoglycaemia, type 1 diabetes
1. INTRODUCTION
Glycaemic control is a modifiable risk factor for diabetes‐related complications, but many people with type 1 diabetes (T1DM) do not meet appropriate glycaemic targets. The T1D Exchange registry found that only 21% of adults with T1DM achieved the American Diabetes Association (ADA)‐recommended HbA1c target of less than 7.0 %, while 37% achieved HbA1c less than 7.5 %. 1 Additionally, average HbA1c levels had not improved over a 5‐year period despite increased use of insulin pumps and continuous glucose monitoring (CGM). 1 By comparison, a National Diabetes Audit in the UK reported 30% of patients with T1DM achieving HbA1c less than 7.5 %; an increase of 10% over a 6‐year period. 2
In T1DM, poor glycaemic control is associated with long‐term complications and increased mortality. 3 , 4 , 5 The Diabetes Control and Complications Trial found intensive treatment was associated with lower HbA1c levels than conventional treatment (approximately 7 % vs. 9 %), with lower long‐term mortality and incidence of cardiovascular disease according to the Epidemiology of Diabetes Interventions and Complications extension. 6 Although intensive treatment was associated with a threefold increased risk of severe hypoglycaemia, 7 more recent data have not confirmed an association between severe hypoglycaemia and HbA1c levels. 8 Nevertheless, fear of hypoglycaemia represents a major physician‐ and patient‐related barrier to glycaemic control. 9 , 10 Moreover, hypoglycaemia directly and indirectly impacts the cost of diabetes management, through increased need for hospitalisation, clinic visits and absence from work. 11
While global real‐life data have been previously reported for people aged 8–25 years, limited data are available examining regional differences in diabetes management in adults with T1DM, although the International Diabetes Management Practice Study documented treatment practices in developing countries (in Africa, Asia, Eastern Europe, Latin America and the Middle East) and identified factors including self‐monitoring and patient education that impacted glycaemic target achievement. 12
Study of Adults' GlycEmia in T1DM (SAGE) examined socio‐demographics, glycaemic control, therapy and comorbidities, in addition to psychosocial aspects related to the disease in adults with T1DM of different ages across 17 non‐US countries in five regions.
2. MATERIALS AND METHODS
2.1. Study design and participants
SAGE was a multinational, multicentre, single visit, cross‐sectional, observational study to describe glycaemic control and quality of life (QoL) in adults with T1DM. QoL outcomes will not be presented in this report.
Selected sites were expected to see ≥100 people with T1DM per year with ≥1 visit per year for each individual. Endocrinologists, general practitioners and other physicians familiar with the management of people with T1DM participated in the study. In each country, physicians were selected independently and randomly from the pre‐established country specific lists of potential sites.
Included participants were aged ≥26 years with T1DM for ≥1 year, treated with insulin and with an HbA1c value available within 30 days preceding the study visit or planned to be obtained in routine practice within 7 days after the study visit. Exclusion criteria included non‐T1DM, switch from insulin pump to multiple insulin injections regimen, or vice versa, within 3 months preceding the study, treatment with thiazolidinedione, sulfonylurea or dipeptidyl peptidase‐4 inhibitors at any time since T1DM diagnosis and treatment with any investigational drug within the last 3 months. Participants provided written informed consent.
2.2. Data collection
At a single study visit, investigators collected data from the participant's file and interview into an electronic case report form. HbA1c assessments were performed locally in routine practice, using the standard methodology at the laboratory of each site. No investigations for the purpose of the study were performed. The study was performed according to local regulatory requirements including Institutional Review Board and Independent Ethical Committee approvals where required, and conducted in accordance with the Declaration of Helsinki and guidelines for Good Epidemiology Practice. 13 Data were collected between January 2018 and December 2018.
2.3. Endpoints
The primary endpoint was the proportion of participants who achieve the HbA1c target of less than 7 % in predefined age groups (26–44 years; 45–64 years; ≥65 years).
Secondary clinical endpoints were HbA1c, fasting plasma glucose (FPG) and postprandial plasma glucose (PPG) levels, achievement of physician‐established individualised HbA1c targets, hypoglycaemia frequency (documented symptomatic hypoglycaemic episodes during the last 3 months and severe hypoglycaemia during the last 6 months), severe hyperglycaemia leading to DKA during the last 6 months, concomitant diseases and therapeutic management of T1DM (Supporting Information Materials). FPG and PPG were defined as the last available laboratory or self‐monitored glucose value.
2.4. Data analysis and statistics
Assuming that the HbA1c target would be achieved in 20%–27% of participants, and that a lack of sufficient information on HbA1c (nonevaluability criterion) would apply to approximately 5% of the population, the inclusion of 500–1000 participants per country/region (e.g., group of countries) would allow the calculation of a two‐sided 95% confidence interval (CI) with a precision of between 2.5% and 3.9% (all age groups considered). With a recruitment ratio of 40%, 40% and 20% in the different predefined age groups of 26–44, 45–64 and ≥65 years, respectively, the precision would be between 4.0% and 6.3% in the first two age classes and between 5.7% and 9.2% in the last age group of ≥65 years. All analyses were conducted in the eligible population, defined as participants meeting the inclusion criteria. The time window for HbA1c evaluation was extended to within 45 days preceding the study visit or within 15 days following the study visit to define the eligible population. All data were reported and analysed using descriptive statistics.
3. RESULTS
3.1. Participant disposition
Of 3903 included participants, 3858 (98.8%) were eligible for the study. Reasons for noneligibility were: no HbA1c available in the time windows (n = 39), age less than 26 years or missing (n = 17) and no clinical diagnosis of presumed autoimmune T1DM (n = 3). It should be noted that an individual participant could have several reasons for noneligibility. Of the eligible population, 1724 (44.7%), 1512 (39.2%) and 622 (16.1%) comprised the 26–44, 45–64 and ≥65 years age groups, respectively. Participants were recruited from 17 different countries, assessed according to the following regions: Asia (India, Japan, Thailand; n = 780), Eastern Europe (Bulgaria, Croatia, Serbia, Ukraine; n = 996), Latin America (Argentina, Brazil, Chile, Colombia; n = 488), Middle East (Iran, Saudi Arabia; n = 444) and Western Europe (France, Germany, Italy, UK; n = 1150; Table S1).
3.2. Participating centres
Two hundred and thirty centres participated in SAGE with regional differences in the types of participating centres and physician specialities. Overall, participating centres were mostly public hospitals, but also included private clinics/offices, private hospitals and “other” classifications, with some centres categorised as more than one “type”. Most participating physicians were Endocrinologists/Diabetologists (Table 1).
TABLE 1.
Participating centres and physicians by region
| Parameter | Asia | Eastern Europe | Western Europe | Latin America | Middle East | Global |
|---|---|---|---|---|---|---|
| Number of sites | 44 | 44 | 67 | 27 | 48 | 230 |
| Type of centre (multiple answers possible), n (%) | ||||||
| Public hospital | 9 (20.5) | 34 (77.3) | 53 (79.1) | 3 (11.1) | 25 (52.1) | 124 (53.9) |
| Private clinic or office | 18 (40.9) | 7 (15.9) | 8 (11.9) | 24 (88.9) | 31 (64.6) | 88 (38.3) |
| Private hospital | 17 (38.6) | 1 (2.3) | 2 (3.0) | 1 (3.7) | 11 (22.9) | 32 (13.9) |
| Other centre | 1 (2.3) | 4 (9.1) | 4 (6.0) | 1 (3.7) | 6 (12.5) | 16 (7.0) |
| Physician speciality, n (%) | ||||||
| Endocrinology/diabetology | 42 (95.5) | 44 (100.0) | 59 (88.1) | 23 (85.2) | 30 (62.5) | 198 (86.1) |
| Internal medicine | 2 (4.5) | 0 | 4 (6.0) | 3 (11.1) | 12 (25.0) | 21 (9.1) |
| Primary care physician | 0 | 0 | 1 (1.5) | 0 | 2 (4.2) | 3 (1.3) |
| Other | 0 | 0 | 3 (4.5) | 1 (3.7) | 4 (8.3) | 8 (3.5) |
3.3. Participant characteristics
Participants had a mean (standard deviation [SD]) age of 47.44 (14.00) years and a mean body mass index (BMI) of 25.15 (4.48) kg/m2, which was similar across all age groups (Table 2). Mean (SD) duration of diabetes was 20.73 (12.63) years overall, ranging from 15.92 (9.05) years in the youngest age group to 28.79 (15.10) years in the oldest age group. Overall, 20.6% had a family history of T1DM, and this was similar across all age groups.
TABLE 2.
Participant characteristics by age group
| Parameter a | Age groups, years | |||
|---|---|---|---|---|
| ≥26 to <45 N = 1724 | ≥45 to <65 N = 1512 | ≥65 N = 622 | All ages N = 3858 | |
| Proportion of total population, % | 44.7 | 39.2 | 16.1 | 100.0 |
| Mean (SD) age, years | 34.63 (5.44) | 52.74 (5.58) | 70.04 (4.88) | 47.44 (14.00) |
| Gender, % female/male | 58.1/41.9 | 51.1/48.9 | 53.9/46.1 | 54.6/45.4 |
| Mean (SD) body mass index, kg/m2 | 24.49 (4.31) | 25.62 (4.49) | 25.84 (4.64) | 25.15 (4.48) |
| Body mass index, n (%) | ||||
| <25 kg/m2 | 1055 (61.3) | 747 (49.5) | 288 (46.3) | 2090 (54.3) |
| 25–30 kg/m2 | 496 (28.8) | 529 (35.1) | 227 (36.5) | 1252 (32.5) |
| ≥30 kg/m2 | 169 (9.8) | 233 (15.4) | 107 (17.2) | 509 (13.2) |
| <27 kg/m2 | 1330 (77.3) | 1002 (66.4) | 399 (64.1) | 2731 (70.9) |
| ≥27 kg/m2 | 390 (22.7) | 507 (33.6) | 223 (35.9) | 1120 (29.1) |
| Mean (SD) duration of diabetes, years | 15.92 (9.05) | 22.91 (12.73) | 28.79 (15.10) | 20.73 (12.63) |
| <10 Years, n (%) | 514 (29.8) | 262 (17.3) | 79 (12.7) | 855 (22.2) |
| ≥10 Years, n (%) | 1210 (70.2) | 1250 (82.7) | 542 (87.3) | 3002 (77.8) |
| Family history of T1DM, % Yes/No | 20.1/79.9 | 21.7/78.3 | 19.2/80.8 | 20.6/79.4 |
| Education level, n (%) | ||||
| Primary | 69 (4.0) | 145 (9.6) | 97 (15.6) | 311 (8.1) |
| Secondary | 688 (39.9) | 719 (47.6) | 295 (47.4) | 1702 (44.1) |
| University/higher education | 907 (52.6) | 598 (39.6) | 198 (31.8) | 1703 (44.1) |
| Employment status | ||||
| Student | 40 (2.3) | 1 (0.1) | 0 | 41 (1.1) |
| Employed b | 1332 (77.3) | 1000 (66.1) | 89 (14.3) | 2421 (62.8) |
| Unemployed | 257 (14.9) | 212 (14.0) | 61 (9.8) | 530 (13.7) |
| Retired | 19 (1.1) | 187 (12.4) | 421 (67.7) | 627 (16.3) |
| Incapacity | 17 (1.0) | 39 (2.6) | 7 (1.1) | 63 (1.6) |
| Other/unknown | 59 (3.4) | 73 (4.8) | 44 (7.0) | 176 (4.6) |
| Health insurance, n (%) | ||||
| Yes | 1319 (76.5) | 1158 (76.6) | 481 (77.5) | 2958 (76.7) |
| Public | 952 (55.2) | 861 (57.0) | 369 (59.4) | 2182 (56.6) |
| Private | 229 (13.3) | 145 (9.6) | 47 (7.6) | 421 (10.9) |
| Public and private | 138 (8.0) | 152 (10.1) | 65 (10.5) | 355 (9.2) |
| No | 405 (23.5) | 353 (23.4) | 140 (22.5) | 898 (23.3) |
Abbreviations: SD, standard deviation; T1DM, type 1 diabetes.
Data are not complete for all participants. Presented results are for participants with data available for each given parameter.
Employee status is “employee” or “independent”.
Mean BMI was lowest in Asia (23.3 kg/m2) and highest in the Middle East (26. 3 kg/m2). Mean BMI in Eastern Europe, Western Europe and Latin America was 25.4, 25.6 and 25.5 kg/m2, respectively. Duration of diabetes was lowest in Asia and highest in Western Europe, while the proportion of participants with a family history of T1DM was highest in Latin America, Western Europe and Middle East (Table 3).
TABLE 3.
Participant characteristics by region
| Parameter a | Asia (n = 780) | Eastern Europe (n = 996) | Western Europe (n = 1150) | Latin America (n = 488) | Middle East (n = 444) |
|---|---|---|---|---|---|
| Mean (SD) age, years | 49.03 (14.04) | 48.58 (13.91) | 46.72 (14.09) | 45.59 (13.85) | 45.95 (13.59) |
| Gender, % female/male | 59.2/40.8 | 52.3/47.7 | 50.9/49.1 | 61.5/38.5 | 54.1/45.9 |
| Mean (SD) body mass index, kg/m2 | 23.28 (4.01) | 25.37 (4.30) | 25.64 (4.55) | 25.51 (4.23) | 26.27 (4.81) |
| Body mass index, n (%) | |||||
| <25 kg/m2 | 566 (72.7) | 510 (51.2) | 578 (50.5) | 238 (48.8) | 198 (44.6) |
| 25–30 kg/m2 | 177 (22.7) | 344 (34.5) | 399 (34.9) | 180 (36.9) | 152 (34.2) |
| ≥30 kg/m2 | 36 (4.6) | 142 (14.3) | 167 (14.6) | 70 (14.3) | 94 (21.2) |
| <27 kg/m2 | 666 (85.5) | 691 (69.4) | 774 (67.7) | 330 (67.6) | 270 (60.8) |
| ≥27 kg/m2 | 113 (14.5) | 305 (30.6) | 370 (32.3) | 158 (32.4) | 174 (39.2) |
| Mean (SD) duration of diabetes, years | 16.84 (11.57) | 19.80 (12.14) | 22.95 (13.26) | 22.49 (12.40) | 21.98 (12.25) |
| <10 Years, n (%) | 254 (32.6) | 231 (23.2) | 209 (18.2) | 83 (17.0) | 78 (17.6) |
| ≥10 Years, n (%) | 525 (67.4) | 765 (76.8) | 941 (81.8) | 405 (83.0) | 366 (82.4) |
| Family history of T1DM, % Yes/No | 12.0/88.0 | 16.8/83.2 | 25.0/75.0 | 24.3/75.7 | 28.1/71.9 |
| Education level, n (%) | |||||
| Primary | 21 (2.7) | 38 (3.8) | 107 (9.3) | 50 (10.2) | 95 (21.4) |
| Secondary | 366 (46.9) | 496 (49.8) | 544 (47.3) | 153 (31.4) | 143 (32.2) |
| University/higher education | 381 (48.8) | 457 (45.9) | 387 (33.7) | 277 (56.8) | 201 (45.3) |
| Employment status, n (%) | |||||
| Student | 2 (0.3) | 3 (0.3) | 16 (1.4) | 11 (2.3) | 9 (2.0) |
| Employed b | 531 (68.1) | 595 (59.7) | 744 (64.7) | 326 (66.8) | 225 (50.7) |
| Unemployed | 83 (10.6) | 138 (13.9) | 109 (9.5) | 64 (13.1) | 136 (30.6) |
| Retired | 94 (12.1) | 228 (22.9) | 186 (16.2) | 70 (14.3) | 49 (11.0) |
| Incapacity | 3 (0.4) | 28 (2.8) | 24 (2.1) | 4 (0.8) | 4 (0.9) |
| Other/unknown | 67 (8.6) | 4 (0.4) | 71 (6.1) | 13 (2.6) | 21 (4.8) |
| Health insurance, n (%) | |||||
| Yes | 663 (85.0) | 569 (57.1) | 926 (80.7) | 433 (88.7) | 367 (82.7) |
| Public | 561 (71.9) | 553 (55.5) | 651 (56.7) | 105 (21.5) | 312 (70.3) |
| Private | 84 (10.8) | 8 (0.8) | 26 (2.3) | 293 (60.0) | 10 (2.3) |
| Public and private | 18 (2.3) | 8 (0.8) | 249 (21.7) | 35 (7.2) | 45 (10.1) |
| No | 117 (15.0) | 427 (42.9) | 222 (19.3) | 55 (11.3) | 77 (17.3) |
Abbreviations: SD, standard deviation; T1DM, type 1 diabetes.
Data are not complete for all participants. Presented results are for participants with data available for each given parameter.
Employee status is “employee” or “independent”.
3.4. Education level, employment status, health insurance coverage and lifestyle
Globally, most participants were educated to secondary or university/higher education level and were in employment (Table 2). The majority had health insurance with comparable coverage across age groups with public, private and both public and private insurance. Health insurance coverage was lowest in Eastern Europe (57.1%, with most of the participants lacking health insurance coming from Ukraine; Table 3). Most participants had public health insurance in all regions except Latin America.
Regardless of age group, most participants lived in urban areas (82.7%) and lived with another adult (88.4%). Professional drivers (a job category for which relevant data were available and for whom poorly controlled diabetes may have particularly dangerous consequences) constituted 13.5% of participants globally, ranging between 8.4% and 15.3% of the population across age groups, while in Asia, the proportion of participants who were professional drivers was particularly high (36.5%).
Overall, most participants (71.4%) adhered to dietary advice (recommendation for carbohydrate intake, a balanced healthy diet with appropriate caloric, fat and fibre intake and/or other special dietary advice), with comparable results across all age groups. Only 34.3% of participants reported recommended activity levels (4 or more days/week with at least 30 min of physical activity), which was similar across age groups. Adherence to dietary recommendations was lowest in the Middle East (59.2%) and Asia (45.1%), while recommended activity levels were highest in Eastern Europe (48.4%) and lowest in Latin America (24.4%).
3.5. Primary endpoint
Overall, the proportion (95% CI) of participants who achieved the glycaemic target of HbA1c less than 7.0 % was 24.3% (22.9–25.6). A higher proportion of participants achieved HbA1c less than 7.0 % in the 26–44 years age group (27.6% [25.5–29.8]), compared with the 45–64 years (21.0% [19.0–23.2]) and ≥65 years (22.8% [19.6–26.3]) age groups (Table 4). Achievement of HbA1c less than 7.0 % was highest in Western Europe (27.0% [24.4–29.6]) and Eastern Europe (26.2% [23.5–29.1]) and was lowest in the Middle East (18.9% [15.4–22.9]; Table 5 ).
TABLE 4.
Endpoints by age group
| Parameter a | Age groups, years | |||
|---|---|---|---|---|
| HbA1c | ≥26 to <45 N = 1724 | ≥45 to <65 N = 1512 | ≥65 N = 622 | All ages N = 3858 |
| HbA1c <7 %, n (%) [95% CI] | 476 (27.6)[25.5–29.8] | 318 (21.0) [19.0–23.2] | 142 (22.8) [19.6– 26.3] | 936 (24.3) [22.9–25.6] |
| HbA1c, n (%) | ||||
| 7.0%–7.5 % (53.0–58.5 mmol/mol) | 278 (16.1) | 249 (16.5) | 101 (16.2) | 628 (16.3) |
| 7.5%–8 % (58.5–63.9 mmol/mol) | 258 (15.0) | 254 (16.8) | 109 (17.5) | 621 (16.1) |
| 8.0%–9 % (63.9–74.9 mmol/mol) | 367 (21.3) | 370 (24.5) | 162 (26.0) | 899 (23.3) |
| 9.0%–10 % (74.9‐85.8 mmol/mol) | 182 (10.6) | 194 (12.8) | 67 (10.8) | 443 (11.5) |
| 10.0%–11.0 % (85.8‐96.7 mmol/mol) | 76 (4.4) | 84 (5.6) | 29 (4.7) | 189 (4.9) |
| ≥11.0 % (≥96.7 mmol/mol) | 87 (5.0) | 43 (2.8) | 12 (1.9) | 142 (3.7) |
| Mean (SD) HbA1c | ||||
| % | 7.91 (1.52) | 8.02 (1.37) | 7.91 (1.24) | 7.95 (1.42) |
| mmol/mol | 62.98 (16.66) | 64.17 (14.97) | 62.96 (13.60) | 63.44 (15.56) |
| Individualised HbA1c target value, n (%) | ||||
| <6.5 % (<47.5 mmol/mol) | 97 (5.6) | 59 (3.9) | 17 (2.7) | 173 (4.5) |
| 6.5%–7.0 % (47.5–53.0 mmol/mol) | 522 (30.3) | 288 (19.0) | 78 (12.5) | 888 (23.0) |
| 7.0%–7.5 % (53.0–58.5 mmol/mol) | 959 (55.6) | 909 (60.1) | 289 (46.5) | 2157 (55.9) |
| 7.5%–8 % (58.5–63.9 mmol/mol) | 115 (6.7) | 203 (13.4) | 169 (27.2) | 487 (12.6) |
| 8.0%–9 % (63.9–74.9 mmol/mol) | 27 (1.6) | 50 (3.3) | 65 (10.5) | 142 (3.7) |
| ≥9 % (≥74.9 mmol/mol) | 4 (0.2) | 3 (0.2) | 4 (0.6) | 11 (0.3) |
| Achieved individualised HbA1c target, n (%) [95% CI] | 373 (21.6) [19.7–23.7] | 269 (17.8) [15.9–19.8] | 163 (26.2) [22.8–29.8] | 805 (20.9) [19.6–22.2] |
| Hypoglycemia and hyperglycemia | ||||
| ≥1 Symptomatic hypoglycaemia with BG ≤ 3.9 mmol/L (≤70 mg/dl) in the previous 3 months, n (%) | 1183 (69.6) | 991 (66.3) | 402 (65.7) | 2576 (67.7) |
| Events/participant, median (min, max) | 4 (0, 180) | 3 (0, 180) | 3 (0, 180) | 3 (0, 180) |
| ≥1 Symptomatic hypoglycaemia with BG < 3.0 mmol/L (<54 mg/dl) in the previous 3 months, n (%) | 882 (51.8) | 728 (48.6) | 293 (47.9) | 1903 (49.9) |
| Events/participant, median (min, max) | 1 (0, 90) | 0 (0, 90) | 0 (0, 90) | 0 (0, 90) |
| ≥1 Severe hypoglycaemia in the previous 6 months, n (%) | 197 (11.5) | 185 (12.2) | 78 (12.6) | 460 (11.9) |
| Mean (SD) number of events/participant | 0.46 (2.24) | 0.45 (2.91) | 0.41 (1.62) | 0.45 (2.45) |
| Median (min, max) | 0 (0, 36) | 0 (0, 90) | 0 (0, 20) | 0 (0, 90) |
| ≥1 Hospitalisation/emergency visit linked to severe hypoglycaemia in the previous 6 months, n (%) b | 49 (24.9) | 47 (25.4) | 25 (32.1) | 121 (26.3) |
| ≥1 Severe hyperglycaemia leading to DKA in the previous 6 months, n (%) | 83 (4.8) | 61 (4.0) | 18 (2.9) | 162 (4.2) |
| Events/participant, median (min, max) | 0 (0, 23) | 0 (0, 10) | 0 (0, 3) | 0 (0, 23) |
| Insulin treatment | ||||
| Device, n (%) | ||||
| Pump | 413 (24.0) | 290 (19.2) | 66 (10.6) | 769 (19.9) |
| Injection/pens | 1310 (76.0) | 1218 (80.6) | 553 (88.9) | 3081 (79.9) |
| Pump and injection/pens | 0 (0.0) | 4 (0.3) | 3 (0.5) | 7 (0.2) |
| Sometimes pump and sometimes injection/pen | 1 (0.1) | 0 (0.0) | 0 (0.0) | 1 (<0.1) |
| Mean (SD) total insulin daily dose | ||||
| U/kg/day | 0.74 (0.31) | 0.70 (0.30) | 0.65 (0.28) | 0.71 (0.30) |
| U/day | 51.0 (23.1) | 50.4 (25.2) | 46.4 (23.7) | 50.0 (24.1) |
| Recommended insulin dose adjustment approach n (%) | ||||
| Physician‐driven | 703 (41.4) | 648 (43.3) | 291 (47.0) | 1642 (43.0) |
| Patient‐driven | 996 (58.6) | 850 (56.7) | 328 (53.0) | 2174 (57.0) |
| Insulin type | ||||
| Pump only | 413 (24.0) | 290 (19.2) | 66 (10.6) | 769 (19.9) |
| Basal c | 1213 (70.4) | 1108 (73.3) | 497 (79.9) | 2818 (73.0) |
| Intermediate acting NPH | 155 (9.0) | 159 (10.5) | 69 (11.1) | 383 (9.9) |
| Long acting analogues | 1058 (61.4) | 949 (62.8) | 428 (68.8) | 2435 (63.1) |
| First generation | 669 (38.8) | 554 (36.6) | 243 (39.1) | 1466 (38.0) |
| Second generation | 389 (22.6) | 395 (26.1) | 185 (29.7) | 969 (25.1) |
| Premix c | 78 (4.5) | 85 (5.6) | 52 (8.4) | 215 (5.6) |
| Short acting insulin c | 1196 (69.4) | 1100 (72.8) | 480 (77.2) | 2776 (72.0) |
| Basal insulin dose adjustment frequency, n (%) d | ||||
| More than weekly | 255 (22.1) | 229 (21.9) | 112 (24.1) | 596 (22.4) |
| Weekly | 263 (22.8) | 255 (24.4) | 114 (24.6) | 632 (23.7) |
| Less than weekly but more than every 2 weeks | 73 (6.3) | 69 (6.6) | 31 (6.7) | 173 (6.5) |
| Less than every 2 weeks but more than monthly | 214 (18.5) | 214 (20.5) | 103 (22.2) | 531 (19.9) |
| Less than monthly | 351 (30.4) | 277 (26.5) | 104 (22.4) | 732 (27.5) |
| Short‐acting insulin dose adjustment frequency, n (%) | ||||
| More than weekly | 906 (57.4) | 731 (53.6) | 263 (49.3) | 1900 (54.7) |
| Weekly | 195 (12.4) | 179 (13.1) | 77 (14.4) | 451 (13.0) |
| Less than weekly but more than every 2 weeks | 50 (3.2) | 46 (3.4) | 22 (4.1) | 118 (3.4) |
| Less than every 2 weeks but more than monthly | 171 (10.8) | 181 (13.3) | 88 (16.5) | 440 (12.7) |
| Less than monthly | 256 (16.2) | 227 (16.6) | 83 (15.6) | 566 (16.3) |
Abbreviations: BG, blood glucose; CI, confidence interval; DKA, diabetic ketoacidosis; NPH, neutral protamine Hagedorn; SD, standard deviation.
Data are not complete for all participants. Presented results are for participants with data available for each given parameter.
Percentage calculated among those patients with ≥1 severe hypoglycaemia.
Alone or in combinations, including participants using only injections/pens, or those using pump and injections/pens or those sometimes using pump and sometimes using injections/pens.
Calculated as percentages of those participants on basal insulin.
TABLE 5.
Endpoints by region
| Parameter a HbA1c | Asia (n = 780) | Eastern Europe (n = 996) | Western Europe (n = 1150) | Latin America (n = 488) | Middle East (n = 444) |
|---|---|---|---|---|---|
| Mean (SD) HbA1c | |||||
| % | 7.98 (1.37) | 8.02 (1.48) | 7.70 (1.21) | 8.15 (1.64) | 8.21 (1.55) |
| mmol/mol | 63.68 (14.99) | 64.11 (16.16) | 60.68 (13.22) | 65.62 (17.88) | 66.28 (16.90) |
| HbA1c <7 %, n (%) [95% CI] | 165 (21.2) [18.3–24.2] | 261 (26.2) [23.5–29.1] | 310 (27.0) [24.4– 29.6] | 116 (23.8) [20.1–27.8] | 84 (18.9) [15.4–22.9] |
| HbA1c, n (%) | |||||
| 7.0%–7.5 % (53.0–58.5 mmol/mol) | 127 (16.3) | 136 (13.7) | 228 (19.8) | 71 (14.5) | 66 (14.9) |
| 7.5%–8 % (58.5–63.9 mmol/mol) | 135 (17.3) | 137 (13.8) | 220 (19.1) | 72 (14.8) | 57 (12.8) |
| 8.0%–9 % (63.9–74.9 mmol/mol) | 212 (27.2) | 229 (23.0) | 234 (20.3) | 100 (20.5) | 124 (27.9) |
| 9.0%–10 % (74.9–85.8 mmol/mol) | 84 (10.8) | 139 (14.0) | 98 (8.5) | 67 (13.7) | 55 (12.4) |
| 10.0%–11.0 % (85.8–96.7 mmol/mol) | 32 (4.1) | 49 (4.9) | 41 (3.6) | 33 (6.8) | 34 (7.7) |
| ≥11.0 % (≥96.7 mmol/mol) | 25 (3.2) | 45 (4.5) | 19 (1.7) | 29 (5.9) | 24 (5.4) |
| Individualised HbA1c target value, n (%) | |||||
| <6.5 % (<47.5 mmol/mol) | 20 (2.6) | 40 (4.0) | 77 (6.7) | 16 (3.3) | 20 (4.5) |
| 6.5%–7.0 % (47.5–53.0 mmol/mol) | 64 (8.2) | 281 (28.2) | 374 (32.5) | 79 (16.2) | 90 (20.3) |
| 7.0%–7.5 % (53.0–58.5 mmol/mol) | 574 (73.6) | 428 (43.0) | 543 (47.2) | 350 (71.7) | 262 (59.0) |
| 7.5%–8 % (58.5–63.9 mmol/mol) | 64 (8.2) | 211 (21.2) | 127 (11.0) | 34 (7.0) | 51 (11.5) |
| 8.0%–9 % (63.9–74.9 mmol/mol) | 53 (6.8) | 36 (3.6) | 23 (2.0) | 9 (1.8) | 21 (4.7) |
| ≥9 % (≥74.9 mmol/mol) | 5 (0.6) | – | 6 (0.5) | – | – |
| Achieved individualised HbA1c target, n (%) [95% CI] | 169 (21.7) [18.8–24.7] | 201 (20.2) [17.7–22.8] | 275 (23.9) [21.5–26.5] | 96 (19.7) [16.2–23.5] | 64 (14.4) [11.3–18.0] |
| Hypoglycaemia and hyperglycaemia | |||||
| ≥1 Symptomatic hypoglycaemia BG ≤ 3.9 mmol/L (≤70 mg/dl) in the previous 3 months, n (%) | 458 (59.2) | 746 (74.9) | 851 (76.7) | 356 (73.4) | 165 (37.2) |
| Events/participant, median (min, max) | 2 (0, 180) | 3 (0, 100) | 8 (0, 180) | 5 (0, 180) | 0 (0, 90) |
| ≥1 Symptomatic hypoglycaemia BG < 3.0 mmol/L (<54 mg/dl) in the previous 3 months, n (%) | 319 (41.3) | 536 (53.8) | 663 (59.6) | 278 (57.2) | 107 (24.1) |
| Events/participant, median (min, max) | 0 (0, 80) | 1 (0, 90) | 2 (0, 90) | 1 (0, 54) | 0 (0, 50) |
| ≥1 Severe hypoglycaemia in the previous 6 months, n (%) | 72 (9.2) | 116 (11.6) | 142 (12.4) | 68 (14.0) | 62 (14.0) |
| Mean (SD) number of events/participant | 0.29 (1.23) | 0.38 (1.63) | 0.59 (2.79) | 0.48 (4.23) | 0.50 (1.90) |
| Median (min, max) | 0 (0, 12) | 0 (0, 20) | 0 (0, 36) | 0 (0, 90) | 0 (0, 20) |
| ≥1 Hospitalisation/emergency visit linked to severe hypoglycaemia in the previous 6 months, n (%) b | 30 (41.7) | 24 (20.7) | 13 (9.2) | 26 (38.2) | 28 (45.2) |
| ≥1 Severe hyperglycaemia leading to DKA in the previous 6 months, n (%) | 18 (2.3) | 30 (3.0) | 77 (6.7) | 22 (4.5) | 15 (3.4) |
| Events/participant, median (min, max) | 0 (0, 5) | 0 (0, 5) | 0 (0, 23) | 0 (0, 10) | 0 (0, 3) |
| Insulin treatment | |||||
| Device, n (%) | |||||
| Pump | 138 (17.7) | 45 (4.5) | 498 (43.3) | 74 (15.2) | 14 (3.2) |
| Injection/pens | 638 (81.8) | 950 (95.4) | 650 (56.5) | 413 (84.6) | 430 (96.8) |
| Pump and injection/pens | 4 (0.5) | 0 | 2 (0.2) | 1 (0.2) | 0 |
| Sometimes pump and sometimes injection/pens | 0 | 1 (0.1) | 0 | 0 | 0 |
| Mean (SD) total insulin daily dose | |||||
| U/kg/day | 0.7 (0.3) | 0.8 (0.3) | 0.6 (0.3) | 0.7 (0.3) | 0.8 (0.4) |
| U/day | 41.4 (21.1) | 57.4 (21.5) | 45.9 (22.9) | 49.5 (25.5) | 59.3 (28.1) |
| Recommended insulin dose adjustment approach | |||||
| Physician‐driven | 436 (56.3) | 286 (28.7) | 355 (31.0) | 242 (53.1) | 323 (72.7) |
| Patient‐driven | 338 (43.7) | 710 (71.3) | 791 (69.0) | 214 (46.9) | 121 (27.3) |
| Insulin type | |||||
| Pump only | 138 (17.7) | 45 (4.5) | 498 (43.3) | 74 (15.2) | 14 (3.2) |
| Basal c | 545 (69.9) | 899 (90.3) | 621 (54.0) | 397 (81.4) | 356 (80.2) |
| Intermediate acting NPH | 29 (3.7) | 217 (21.8) | 15 (1.3) | 75 (15.4) | 47 (10.6) |
| Long acting analogues | 516 (66.2) | 682 (68.5) | 606 (52.7) | 322 (66.0) | 309 (69.6) |
| First generation | 181 (23.2) | 431 (43.3) | 323 (28.1) | 223 (45.7) | 308 (69.4) |
| Second generation | 335 (42.9) | 251 (25.2) | 283 (24.6) | 99 (20.3) | 1 (0.2) |
| Premix c | 77 (9.9) | 50 (5.0) | 11 (1.0) | 3 (0.6) | 74 (16.7) |
| Short acting insulin c | 570 (73.1) | 898 (90.2) | 623 (54.2) | 351 (71.9) | 334 (75.2) |
| Basal insulin dose adjustment frequency, n (%) d | |||||
| More than weekly | 58 (11.4) | 247 (28.7) | 165 (29.0) | 75 (20.3) | 51 (14.3) |
| Weekly | 70 (13.8) | 287 (33.3) | 140 (24.6) | 54 (14.6) | 81 (22.8) |
| Less than weekly but more than every 2 weeks | 12 (2.4) | 82 (9.5) | 24 (4.2) | 9 (2.4) | 46 (12.9) |
| Less than every 2 weeks but more than monthly | 210 (41.3) | 154 (17.9) | 72 (12.7) | 37 (10.0) | 58 (16.3) |
| Less than monthly | 158 (31.1) | 91 (10.6) | 168 (29.5) | 195 (52.7) | 120 (33.7) |
| Short‐acting insulin dose adjustment frequency, n (%) | |||||
| More than weekly | 233 (33.1) | 677 (71.9) | 730 (67.5) | 179 (44.9) | 81 (23.3) |
| Weekly | 100 (14.2) | 116 (12.3) | 121 (11.2) | 51 (12.8) | 63 (18.1) |
| Less than weekly but more than every 2 weeks | 14 (2.0) | 30 (3.2) | 33 (3.0) | 7 (1.8) | 34 (9.8) |
| Less than every 2 weeks but more than monthly | 208 (29.5) | 75 (8.0) | 70 (6.5) | 31 (7.8) | 56 (16.1) |
| Less than monthly | 149 (21.2) | 44 (4.7) | 128 (11.8) | 131 (32.8) | 114 (32.8) |
Abbreviations: BG, blood glucose; CI, confidence interval; DKA, diabetic ketoacidosis; NPH, Neutral Protamine Hagedorn; SD, standard deviation.
Data are not complete for all participants. Presented results are for participants with data available for each given parameter.
Percentage calculated among those patients with ≥1 severe hypoglycaemia.
Alone or in combinations, including participants using only injections/pens, or those using pump and injections/pens or those sometimes using pump and sometimes using injections/pens.
Calculated as percentages of those participants on basal insulin.
4. SECONDARY CLINICAL ENDPOINTS
4.1. Mean HbA1c
Mean (SD) HbA1c was 7.95 (1.42) % and was similar across age groups (Table 4). Mean (SD) HbA1c was lowest in Western Europe (7.70 [1.21] %) and highest in the Middle East (8.21 [1.55] %; Table 5).
4.2. Individualised HbA1c targets
Overall, 20.9% of participants (95% CI: 19.6–22.2) reached their physician‐defined individualised HbA1c target, while 26.2% reached these targets in the oldest subgroup (22.8–29.8). Notably, individualised targets were between 7.0 and 7.5 % for most participants (56% overall), but were generally higher in the older subgroup (Table 4). Individualised target achievement was greatest in Western Europe (23.9% [21.5–26.5]) and lowest in the Middle East (14.4% [11.3–18.0]; Table 5).
4.3. FPG and PPG
Mean (SD) FPG and PPG were 144.8 (62.92) mg/dl and 171.3 (67.00) mg/dl and did not appear to differ between age groups, although FPG and PPG data were missing for approximately 20% and 30% of participants, respectively. Mean (SD) FPG and PPG were lowest in Eastern Europe 137.2 (49.89) and 159.5 (50.92) mg/dl, respectively, and were highest in the Middle East 153.0 (73.18) and 190.1 (78.78) mg/dl, respectively.
4.4. Hypoglycaemia
Symptomatic hypoglycaemia (≤3.9 and <3.0 mmol/L) incidence within the last 3 months was similar across all age groups. The proportion of participants who experienced ≥1 severe hypoglycaemic event in the previous 6 months increased modestly with age (Table 4). Of these participants, the incidence of at least one hospitalisation or emergency room visit linked to severe hypoglycaemia was higher in the ≥65‐year age group. The incidence of symptomatic hypoglycaemia (≤3.9 and <3.0 mmol/L) was lowest in the Middle East and Asia compared with the other regions (Table 5). The proportion who experienced ≥1 severe hypoglycaemic event was highest in the Middle East and Latin America and lowest in Asia; rates of hospitalisation or emergency visits varied considerably (Table 5).
4.5. Hyperglycaemia
The incidence of severe hyperglycaemia leading to DKA within the last 6 months was 4.2% overall and was lower in the ≥65‐year age group compared with the other age groups (Table 4). In participants who experienced DKA, predisposing factors present in more than 20% of participants overall were infection (21.6%) and missing insulin dose (23.5%), with similar proportions across all age groups. Pump malfunction was the predisposing factor in 13.6% of participants with DKA and decreased with age. Of those participants who had experienced severe hyperglycaemia leading to DKA within the last 6 months, 46.9% had an associated hospitalisation.
Rates of severe hyperglycaemia leading to DKA ranged from 2.3% in Asia to 6.7% in Western Europe (Table 4). The rate of being hospitalised at least once in relation to severe hyperglycaemia leading to DKA ranged from 18.2% in Western Europe to 83.3% in Eastern Europe.
4.6. Complications and comorbidities
Overall, 46.7% of participants reported at least one microvascular diabetes complication. The incidence of diabetic neuropathy, diabetic retinopathy and diabetes‐related renal function impairment was 32.5%, 33.2% and 15.9% of participants overall, respectively and increased with age (Table S2). The incidence of each of these comorbidities was particularly high in Eastern Europe compared with other regions (Table S3).
Rates of macrovascular complications (such as coronary heart disease and peripheral vascular disease), other complications and hypertension were highest in the ≥65‐year age group and in Eastern Europe compared with the other regions (Tables S2 and S3).
5. THERAPEUTIC MANAGEMENT
5.1. Insulin device and formulation use
Injections/pens, pumps or a combination of the two were used by 79.9%, 19.9% and 0.2% of participants, respectively (Table 4). Insulin pump use decreased as age increased, while injection/pen use increased with increasing age. Neutral Protamine Hagedorn (NPH) use tended to increase with age. Long‐acting basal insulin (BI) analogues (first‐generation [insulin glargine 100 U/ml; insulin detemir] or second‐generation [insulin glargine 300 U/ml; insulin degludec]) were used by 63.1% of participants and increased with age. Second‐generation long‐acting BI analogues were used by more participants in the older age groups (Table 4).
Insulin pump use ranged from 43.3% in Western Europe to 3.2% in the Middle East (Table 5). NPH was used most frequently in Eastern Europe and Latin America. Long acting BI analogue use was highest in the Middle East (mostly first‐generation analogues), while most participants from Asia used second‐generation long‐acting BI analogues (Table 5).
5.2. Insulin dose and dose adjustment
Mean (SD) total insulin doses were 50.0 (24.1) U/day or 0.71 (0.30) U/kg/day (Table 4). The mean daily dose decreased from 51.0 U/day or 0.74 U/kg/day in the youngest subgroup to 46.4 U/day or 0.65 U/kg in the oldest subgroup. Overall, 2660 (68.9%) participants were receiving a basal‐bolus regimen, the majority (78.8%) were injecting one daily basal dose, administered in the evening in most cases (70.1%).
Most participants followed a patient‐driven protocol for insulin dose adjustments (insulin titration; 57.0%), but as age progressed more participants followed physician‐driven insulin adjustments (Table 4). Physician‐driven insulin dose adjustment was most common in Asia, Latin America and the Middle East (Table 5).
Overall, 27.5% of participants adjusted their BI dose less than every month, and this was more common in the younger age groups (Table 4). Approximately 20% of participants adjusted their BI dose more than every month but less than every 2 weeks (Table 4). By comparison, most participants adjusted short‐acting insulin doses more than once a week (54.7% overall; Table 4). Over 50% of participants in Latin America adjusted their BI dose less than once a month, while BI dose adjustment more than every month but less than every 2 weeks, was most common in Asia. Frequency of short‐acting insulin dose adjustment also varied by region, with particularly high proportions of participants in Eastern (71.9%) and Western Europe (67.5%) adjusting doses more than once a week (Table 5).
5.3. Concomitant glucose‐lowering therapies
Globally, 11.1% of participants used ≥1 glucose‐lowering drug in addition to insulin, most commonly metformin (9.3%). Metformin use increased with age (6.9% in the 26–44 years group; 11.7% in the ≥65 years group) and was most common in Latin America and the Middle East (13.7% and 19.4%) and lowest in Eastern Europe (3.6%). Use of sodium‐glucose linked transporter‐2, alpha‐glucosidases inhibitors and glucagon‐like peptide‐1 was low; 1.2%, 1.1% and 0.3%, respectively. It should be noted that thiazolidinedione, sulfonylurea or dipeptidyl peptidase‐4 inhibitor use at any time since T1DM diagnosis was an exclusion criterion.
5.4. Structure of medical care
Globally, the healthcare team of most people with T1DM included a diabetes specialist (54.6%) and/or an endocrinologist (52.1%), which was consistent across age groups. For each participant, multiple healthcare professionals could be involved. Healthcare teams also included an ophthalmologist for 29.7%, a diabetes‐specialist nurse for 22.0% and a dietician/nutrition specialist for 17.8% of the global population. Only 3.0% of participants had access to a psychologist.
5.5. Technology use
Based on responses to the technology use questionnaire, finger‐stick blood glucose meters were used by most participants worldwide (92.0%) and did not vary by age group, with use being lowest in Western Europe (83.1%) and Asia (90.8%) and highest in Eastern Europe (98.4%).
The proportion who used CGM (23.2% globally) tended to decrease with age: 24.4% and 20.1% in the 26–44 and ≥ 65 years age group, respectively. The proportion who used CGM also varied from 46.4% in Western Europe to 2.5% in the Middle East. Insulin pump use (19.5% overall) decreased with age (23.4% in the youngest and 10.9% in the oldest age groups) and ranged from 42.3% in Western Europe to 2.7% in the Middle East. Blood ketone meters were used by only 11.1% of participants globally. The use of apps to monitor diet/provide carbohydrate counting (11.3%), to remind users to take their diabetes medication (4.2%), to assist with insulin dose adjustment (4.6%), to manage weight (4.7%) or to store personal health information (5.4%) was low and was mostly used when recommended by the HCP. In general, younger participants were more likely to use apps than older participants (e.g., 14.3% of the youngest and 5.5% of the oldest subgroups used a diet/carbohydrate counting app).
6. DISCUSSION
SAGE adds further evidence that glycaemic control remains poor in adults with T1DM in real‐life. Approximately a quarter of participants achieved HbA1c less than 7.0 % across age groups, while target achievement declined with increasing age. Despite being low, this proportion is higher than 18% of the young adult cohort (aged 19–25 years) reported in the Global Teens Registry and 21% of adults in the T1D Exchange registry. 1 , 14 Suboptimal glycaemic control was identified across every region analysed, notably in the Middle East and Asia where only approximately one‐fifth of participants achieved HbA1c less than 7.0 %.
While guidelines provide generalised HbA1c targets, they also recommend individualised targets according to factors such as age, diabetes duration, pregnancy, lifestyle, hypoglycaemia risk, life expectancy, diabetes complications, comorbidities and occupation. 15 , 16 , 17 Few participants (20.9%) achieved individualised HbA1c targets overall but rates of achievement were highest in the ≥65 years age group, potentially reflecting more relaxed HbA1c targets of <7.5 to ≤8.5 % (depending on coexisting illnesses, cognitive function and functional status) in this age group. 18
Previous studies have shown that technology usage by patients with T1DM has increased in recent years; during 2016–2018 in T1D Exchange study in the US, pumps and CGM were used by 63% and 30% of participants, respectively. By comparison, overall rates of pump and CGM use were lower in SAGE (approximately 20% and 23% globally). However, usage varied considerably across the regions analysed, with the highest use of pumps (42%) and CGM (46%) observed in Western Europe. Therefore, technology usage potentially reflects differences in insurance coverage or physician training for CGM.
Interestingly, HbA1c target achievement rates seem to be higher in regions where patient‐led insulin dose adjustment was more common (Eastern and Western Europe), with BI dose adjustment at least once a week also more frequent in these regions.
Of note, rates of microvascular and macrovascular complications were highest in Eastern Europe, while there were no major differences in factors such as current glycaemic control, age, BMI or duration of diabetes between Eastern Europe and the other regions. However, complications result from long‐term diabetes control, 19 on which no data were available in this study and the current HbA1c levels captured may not represent historical glycaemic control. Such a discrepancy between historic and current glycaemic control may be a particularly relevant consideration for Eastern Europe (comprising Bulgaria, Croatia, Serbia and Ukraine in SAGE) because of evolving healthcare systems associated with the dramatic political, economic and social changes in this region since the early 1990s 20 Additionally, it is important to highlight that health insurance coverage was lowest in Eastern Europe, with most of the participants who lacked insurance coming from Ukraine. Differences in access to healthcare must be considered as many patients in some countries included in SAGE (e.g., India) face considerable challenges accessing treatment (including insulin), screening for complications and support. 21 Ethnical, cultural and healthcare system‐related factors and type of centres that recruited the participants may also impact outcomes. Further analyses on the SAGE data, including patient‐reported outcomes, may help elucidate the association between these factors and glycaemic control.
Regions in which a higher percentage of participants achieved their HbA1c target, also recorded a higher incidence of ≥1 symptomatic hypoglycaemia event. The poor glycaemic control observed overall, may be due to suboptimal insulin management. Use of structured education programmes can support reduction in HbA1c levels and variability 22 and may help enforce the benefits of self‐management tools.
As may be expected given the HbA1c data, rates of DKA reported in SAGE were similar to those reported in the literature. 23 , 24 , 25 For example, in a UK retrospective study approximately 1%–2% of participants experienced DKA episodes leading to hospital visits over a 6‐month period, 25 while in SAGE the incidence of hospitalisation/emergency visits linked to severe hyperglycaemia leading to DKA was approximately 2%. In SAGE, this rate of hospitalisation/emergency visits corresponded to approximately half of the participants who experienced severe hyperglycaemia leading to DKA, which appears lower than expected given the clinical consequences of DKA. Of note, as DKA and hospitalisations were reported by physicians who may not have had direct contact with patients in the emergency room, it is possible the incidence of hospitalisation/emergency visits linked to severe hyperglycaemia leading to DKA may have been under reported.
It is of interest that approximately 21% of participants had a family history of T1DM. This is somewhat higher than previous reports that suggest that approximately 15% of people with T1DM have a family history of T1DM. 26 However, our data are comparable to a Finnish study, which reported approximately 24% of children with T1DM has a first‐ or second‐degree relative with the disease. 27
The strengths of this study include the insights provided by the assessment of a large population of adults with T1DM across five geographic regions and comprehensive collection of data on various aspects of diabetes management. These data are of interest as there are limited data describing real‐life outcomes in this population. The status of T1DM management in the United States was recently reported in the T1D Exchange study, and SAGE expands on this evidence to non‐US regions. However, limitations of the SAGE study include the lack of North American and African participants. While participants were representative of each region (although selection bias cannot be completely ruled out), not all countries from each region were included. Furthermore, country‐specific analyses were not performed, so conclusions about the impact of factors such as health care systems on T1DM outcomes cannot be made, since healthcare systems may differ between countries within a region. Data were collected retrospectively, and in accordance with local clinical practice, so it is possible that there were some variations between sites in areas such as how T1DM was diagnosed and how endpoints such as DKA were defined. Additionally, while a longer study period may present a challenge in terms of maintaining data collection from patients, 6 months represent a relatively short period over which to collect severe hypoglycaemia and DKA data, so this may have influenced results.
In conclusion, results from this global study confirm that glycaemic control is suboptimal in adults with T1DM. SAGE identified several areas where treatment can be optimised in order to improve outcomes, including better supporting patients to self‐manage their insulin therapy, increasing the use of technologies such as CGM and the provision of greater healthcare support. The results of this study will be of importance when taking into account the current incremental incidence in T1DM over the globe. 28
CONFLICT OF INTERESTS
Eric Renard is a consultant/advisor for Abbott, Air Liquide, Cellnovo, Eli Lilly, Insulet, Johnson & Johnson (Animas, LifeScan), Medtronic, Novo Nordisk, Roche Diagnostics and Sanofi‐Aventis and research grant/material support from Abbott, Dexcom, Insulet, Roche Diagnostics and Tandem Diabetes Care. Hiroshi Ikegami receives honoraria for lectures: Astellas, MSD, Terumo, Eli Lilly Japan, Novartis, Novo Nordisk, Research Support: Sumitomo Dainippon Pharma, Otsuka Pharmaceutical, Takeda, Mitsubishi Tanabe Pharma, Novo Nordisk, Abott, Johnson & Johnson, Astellas, Ono Pharmaceutical, Kyowa Kirin, Daiichi Sankyo, Boehringer Ingelhaim Japan and Bayer. André Gustavo Daher Vianna is on advisory board and received speaker fees from Sanofi, Abbott Diabetes Care, Medtronic, Novo Nordisk, Lilly, Servier and Astra‐Zeneca and research grant support from Sanofi, Lilly, Astra‐Zeneca and Novo Nordisk. Paolo Pozzilli is a consultant for Astra Zeneca, Sanofi, Lilly and Abbott. Sandrine Brette is an Aixial employee, mandated by Sanofi. Zsolt Bosnyak, Felipe Lauand and Valerie Pilorget are Sanofi employees and shareholders. Anne Peters received advisory board fees from Abbott diabetes Care, Bigfoot, BI, Eli Lilly, Livongo, MannKind, Novo, Sanofi, Whole Biome and research support: Dexcom, vTvTherapeutics. Stock Options: Mellitus Health, Omada Health, Stability Health, Pendulum Therapeutics. Dubravka Jurišić‐Eržen received consultancy fees from Astra Zeneca, Boehringer Ingelheim, Eli Lily, Novo Nordisk, Sanofi Aventis and Takeda. Jothydev Kesavadev received advisory board and speaker fees from Sanofi, Novo Nordisk, MSD, AstraZeneca, Biocon, Abbott Diabetes Care, Medtronic, Boehringer Ingelheim and Johnson & Johnson. Jochen Seufert is an advisory board member of Abbott, AstraZeneca, Boehringer Ingelheim, GI Dynamics, Janssen, LifeScan Mundipharma Novartis, Novo Nordisk, Sanofi Speaker Abbott, AstraZeneca, Bayer, Berlin Chemie Boehringer Ingelheim, Bristol Myers Squibb, Janssen, Eli Lilly, Merck Sharp Dohme (MSD) MedScape Novartis, Novo Nordisk, Omniamed Sanofi; research support from AstraZeneca, Boehringer Ingelheim, Bristol Myers Squibb, GI Dynamics, Intarcia Ipsen, Janssen, Novartis, Novo Nordisk, Sanofi, Ypsomed. Emma G. Wilmot received personal fees from Abbott Diabetes Care, Dexcom, Eli Lilly, Medtronic, Novo Nordisk and Sanofi Aventis.
AUTHOR CONTRIBUTIONS
Eric Renard, André Gustavo Daher Vianna, Paolo Pozzilli, Dubravka Jurišić‐Eržen, Jothydev Kesavadev and Emma G. Wilmot participated as an investigator, provided comments and input to all drafts of the manuscript and interpretation of the results. Hiroshi Ikegami, Felipe Lauand Anne Peters and Jochen Seufert provided comments and input to all drafts of the manuscript and interpretation of the results. Sandrine Brette contributed to acquisition of data and analysis, provided comments and input to all drafts of the manuscript and interpretation of the results. Zsolt Bosnyak contributed to original design of the trial, provided comments and input to all drafts of the manuscript and interpretation of the results. Valerie Pilorget contributed to original design, acquisition of data and analysis, provided comments and input to all drafts of the manuscript and interpretation of the results.
Supporting information
Supplementary Material
ACKNOWLEDGEMENTS
The authors would like to thank the study participants, trial staff and investigators for their participation. This work was supported by Sanofi, Paris, France. Editorial assistance was provided by Fishawack Communications Ltd., and funded by Sanofi. All authors had full access to all of the data in this study and take complete responsibility for the integrity of the data and accuracy of the data analysis.
Renard E, Ikegami H, Daher Vianna AG, et al. The SAGE study: global observational analysis of glycaemic control, hypoglycaemia and diabetes management in T1DM. Diabetes Metab Res Rev. 2021;37(7):e3430. doi: 10.1002/dmrr.3430
Data from this manuscript has been published in abstract form (Renard E et al. Abstract 13. Diabetologia [2019] 62[Suppl 1]: 1. https://doi.org/10.1007/s00125‐019‐4946‐6; Seufert J et al. Abstract 14. Diabetologia [2019] 62[Suppl 1]: 1. https://doi.org/10.1007/s00125‐019‐4946‐6; Wilmot EG et al. P015. Pediatric Diabetes. 2019;20 [Suppl. 28]:41–231; Seufert J et al. P101 Pediatric Diabetes. 2019;20 [Suppl. 28]:41–231; Renard E et al. P179. Pediatric Diabetes. 2019;20 [Suppl. 28]:41–231) and presented as oral presentations at the European Association for the Study of Diabetes (EASD) 2019 congress, and as posters at the International Society for Pediatric and Adolescent Diabetes (ISPAD) 2019 congress.
DATA AVAILABILITY STATEMENT
Qualified researchers may request access to patient level data and related study documents including the clinical study report, study protocol with any amendments, blank case report form, statistical analysis plan and dataset specifications. Patient level data will be anonymised, and study documents will be redacted to protect the privacy of trial participants. Further details on Sanofi's data sharing criteria, eligible studies and process for requesting access can be found at: https://www.clinicalstudydatarequest.com.
References
- 1. Foster NC, Beck RW, Miller KM, et al. State of type 1 diabetes management and outcomes from the T1D Exchange in 2016‐2018. Diabetes Technol Ther. 2019;21(2):66‐72. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. National Diabetes Audit, 2016‐17. 2018. https://digital.nhs.uk/data-and-information/publications/statistical/national-diabetes-audit/national-diabetes-audit-report-1-findings-and-recommendations-2016-2017. Accessed July 2019. [Google Scholar]
- 3. Hainsworth DP, Bebu I, Aiello LP, et al. Risk factors for retinopathy in type 1 diabetes: the DCCT/EDIC study. Diabetes Care. 2019;42(5):875‐882. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Gubitosi‐Klug RA, DCCT/EDIC Research Group . The diabetes control and complications trial/epidemiology of diabetes interventions and complications study at 30 years: summary and future directions. Diabetes Care. 2014;37(1):44‐49. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Herman WH, Braffett BH, Kuo S, et al. What are the clinical, quality‐of‐life, and cost consequences of 30 years of excellent vs. poor glycemic control in type 1 diabetes? J Diabetes Complications. 2018;32(10):911‐915. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. The Diabetes Control and Complications Trial (DCCT)/Epidemiology of Diabetes Interventions and Complications (EDIC) Study Research Group . Intensive diabetes treatment and cardiovascular outcomes in type 1 diabetes: the DCCT/EDIC study 30‐year follow‐up. Diabetes Care. 2016;39:686‐693.26861924 [Google Scholar]
- 7. Nathan DM. The diabetes control and complications trial/epidemiology of diabetes interventions and complications study at 30 Years: overview. Diabetes Care. 2014;37(1):9‐16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Weinstock RS, DuBose SN, Bergenstal RM, et al. Risk factors associated with severe hypoglycemia in older adults with type 1 diabetes. Diabetes Care. 2016;39(4):603‐610. [DOI] [PubMed] [Google Scholar]
- 9. Aschner P, Gagliardino JJ, Ilkova HM, et al. Nonachievement of glycemic target—results from the International diabetes management practices study (IDMPS). Diabetes. 2018;67(suppl 1):1030.29487114 [Google Scholar]
- 10. Aschner P, Gagliardino JJ, Ilkova HM, et al. Reasons for discontinuation of insulin therapy—results from the International diabetes management practices study (IDMPS). Diabetes. 2018;67(suppl 1):1026. [Google Scholar]
- 11. Aronson R, Galstyan G, Goldfracht M, Al Sifri S, Elliott L, Khunti K. Direct and indirect health economic impact of hypoglycaemia in a global population of patients with insulin‐treated diabetes. Diabetes Res Clin Pract. 2018;138:35‐43. [DOI] [PubMed] [Google Scholar]
- 12. Chan JCN, Gagliardino JJ, Baik SH, et al. Multifaceted determinants for achieving glycemic control. Int Diabetes Manag Pract Study. 2009;32(2):227‐233. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. International Society for Pharmocoepidemiology . Guidelines for Good Pharmacoepidemiology Practices; 2007. https://www.pharmacoepi.org/resources/policies/guidelines-08027/. [Google Scholar]
- 14. Anderson BJ, Laffel LM, Domenger C, et al. Factors associated with diabetes‐specific health‐related quality of life in youth with type 1 diabetes: the global TEENs study. Diabetes Care. 2017;40(8):1002‐1009. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. American Diabetes Association . 6. Glycemic targets: standards of medical care in diabetes‐2019. Diabetes Care. 2019;42(suppl 1):S61‐S70. [DOI] [PubMed] [Google Scholar]
- 16. National Clinical Guideline Centre . Type 1 Diabetes in Adults: Diagnosis and Management. London; 2015. [Google Scholar]
- 17. Authors/Task Force M, Ryden L, Grant PJ, et al. ESC guidelines on diabetes, pre‐diabetes, and cardiovascular diseases developed in collaboration with the EASD: the Task Force on diabetes, pre‐diabetes, and cardiovascular diseases of the European Society of Cardiology (ESC) and developed in collaboration with the European Association for the Study of Diabetes (EASD). Eur Heart J. 2013;34(39):3035‐3087. [DOI] [PubMed] [Google Scholar]
- 18. American Diabetes A. 12. Older adults: standards of medical care in diabetes‐2019. Diabetes Care. 2019;42(suppl 1):S139‐S147. [DOI] [PubMed] [Google Scholar]
- 19. Laiteerapong N, Ham SA, Gao Y, et al. The legacy effect in type 2 diabetes: impact of early glycemic control on future complications (the diabetes & aging study). Diabetes Care. 2019;42(3):416‐426. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Romaniuk P, Szromek AR. The evolution of the health system outcomes in Central and Eastern Europe and their association with social, economic and political factors: an analysis of 25 years of transition. BMC Health Serv Res. 2016;16:95. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Kesavadev J, Sadikot SM, Saboo B, et al. Challenges in type 1 diabetes management in south East Asia: descriptive situational assessment. Indian J Endocrinol Metabol. 2014;18(5):600‐607. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Walker GS, Chen JY, Hopkinson H, Sainsbury CAR, Jones GC. Structured education using dose adjustment for normal eating (DAFNE) reduces long‐term HbA1c and HbA1c variability. Diabet Med. 2018;35(6):745‐749. [DOI] [PubMed] [Google Scholar]
- 23. Fazeli Farsani S, Brodovicz K, Soleymanlou N, Marquard J, Wissinger E, Maiese BA. Incidence and prevalence of diabetic ketoacidosis (DKA) among adults with type 1 diabetes mellitus (T1D): a systematic literature review. BMJ Open. 2017;7(7):e016587. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Thomas NJ, Jones SE, Weedon MN, Shields BM, Oram RA, Hattersley AT. Frequency and phenotype of type 1 diabetes in the first six decades of life: a cross‐sectional, genetically stratified survival analysis from UK Biobank. Lancet Diabetes Endocrinol. 2018;6(2):122‐129. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Pang T, Bain SC, Black RNA, et al. A multicentre, UK, retrospective, observational study to assess the effectiveness of insulin glargine 300 units/ml in treating people with Type 1 diabetes mellitus in routine clinical practice (SPARTA). Diabet Med. 2019;36(1):110‐119. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Michels A, Zhang L, Khadra A, Kushner JA, Redondo MJ, Pietropaolo M. Prediction and prevention of type 1 diabetes: update on success of prediction and struggles at prevention. Pediatr Diabetes. 2015;16(7):465‐484. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Parkkola A, Harkonen T, Ryhanen SJ, Ilonen J, Knip M, Finnish Pediatric Diabetes Register . Extended family history of type 1 diabetes and phenotype and genotype of newly diagnosed children. Diabetes Care. 2013;36(2):348‐354. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Xia Y, Xie Z, Huang G, Zhou Z. Incidence and trend of type 1 diabetes and the underlying environmental determinants. Diabetes Metab Res Rev. 2019;35(1):e3075. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplementary Material
Data Availability Statement
Qualified researchers may request access to patient level data and related study documents including the clinical study report, study protocol with any amendments, blank case report form, statistical analysis plan and dataset specifications. Patient level data will be anonymised, and study documents will be redacted to protect the privacy of trial participants. Further details on Sanofi's data sharing criteria, eligible studies and process for requesting access can be found at: https://www.clinicalstudydatarequest.com.