INTERNATIONAL CONSENSUS REPORT

Diabetes Care Volume 42, August 2019 1593

Tadej Battelino,1 Thomas Danne,2

Clinical Targets for Continuous Richard M. Bergenstal,3
Stephanie A. Amiel,4 Roy Beck,5
Glucose Monitoring Data Torben Biester,2 Emanuele Bosi,6
Bruce A. Buckingham,7 William T. Cefalu,8
Interpretation: Recommendations Kelly L. Close,9 Claudio Cobelli,10
Eyal Dassau,11 J. Hans DeVries,12,13
From the International Consensus Kim C. Donaghue,14 Klemen Dovc,1
Francis J. Doyle III,11 Satish Garg,15
on Time in Range George Grunberger,16 Simon Heller,17
Diabetes Care 2019;42:1593–1603 | https://doi.org/10.2337/dci19-0028 Lutz Heinemann,18 Irl B. Hirsch,19
Roman Hovorka,20 Weiping Jia,21
Olga Kordonouri,2 Boris Kovatchev,22
Aaron Kowalski,23 Lori Laffel,24
Brian Levine,9 Alexander Mayorov,25
Chantal Mathieu,26 Helen R. Murphy,27
Revital Nimri,28 Kirsten Nørgaard,29
Christopher G. Parkin,30 Eric Renard,31
David Rodbard,32 Banshi Saboo,33
Improvements in sensor accuracy, greater convenience and ease of use, and Desmond Schatz,34 Keaton Stoner,35
expanding reimbursement have led to growing adoption of continuous glucose Tatsuiko Urakami,36 Stuart A. Weinzimer,37
monitoring (CGM). However, successful utilization of CGM technology in routine and Moshe Phillip28,38
clinical practice remains relatively low. This may be due in part to the lack of clear
and agreed-upon glycemic targets that both diabetes teams and people with
diabetes can work toward. Although unified recommendations for use of key CGM
metrics have been established in three separate peer-reviewed articles, formal This international consensus report has
adoption by diabetes professional organizations and guidance in the practical been endorsed by the American Diabetes
application of these metrics in clinical practice have been lacking. In February 2019, Association, American Association of Clin-
the Advanced Technologies & Treatments for Diabetes (ATTD) Congress convened ical Endocrinologists, American Associa-
tion of Diabetes Educators, European
an international panel of physicians, researchers, and individuals with diabetes
Association for the Study of Diabetes,
who are expert in CGM technologies to address this issue. This article summarizes Foundation of European Nurses in Diabe-
the ATTD consensus recommendations for relevant aspects of CGM data utilization tes, International Society for Pediatric and
and reporting among the various diabetes populations. Adolescent Diabetes, JDRF, and Pediatric
Endocrine Society.
Adoption of continuous glucose monitoring (CGM), which includes both real-time
CGM (rtCGM) and intermittently scanned CGM (isCGM), has grown rapidly over
the past few years as a result of improvements in sensor accuracy, greater
convenience and ease of use, and expanding reimbursement. Numerous studies 1
Department of Pediatric Endocrinology, Diabe-
have demonstrated significant clinical benefits of CGM use in people with tes and Metabolism, University Children’s Hos-
diabetes regardless of insulin delivery method (1–15). In many countries, the pital, University Medical Centre Ljubljana, and
benefits and utility of CGM are now recognized by national and international Faculty of Medicine, University of Ljubljana,
Slovenia
medical organizations for individuals with insulin-requiring diabetes and/or those 2
Diabetes Centre for Children and Adolescents,
at risk for hypoglycemia (16–21). However, despite increased CGM adoption Kinder- und Jugendkrankenhaus Auf der Bult,
(22,23), successful utilization of CGM data in routine clinical practice remains Hannover, Germany
3
relatively low. This may be due in part to the lack of clear and agreed-upon International Diabetes Center at Park Nicollet,
glycemic targets toward which both diabetes teams and people with diabetes can Minneapolis, MN
4
Diabetes Research Group, King’s College London,
work. London, U.K.
In 2012 the Helmsley Charitable Trust sponsored the first expert panel to 5
Jaeb Center for Health Research, Tampa, FL
6
recommend the standardization of CGM metrics and CGM report visualization (24). Diabetes Research Institute, IRCCS San Raffaele
This was followed by a series of CGM consensus statements refining the core CGM Hospital, Vita-Salute San Raffaele University,
metrics, but the conclusions were never in alignment. In 2017, several articles Milan, Italy
7
Division of Endocrinology and Diabetes, Depart-
supported use of systematic approaches to CGM data evaluation (18–20). To date, ment of Pediatrics, Stanford Medical Center,
the key CGM metrics remain as unified recommendations in three separate peer- Stanford, CA
8
reviewed articles, yet formal adoption by diabetes professional organizations and American Diabetes Association, Alexandria, VA
1594 International Consensus Report Diabetes Care Volume 42, August 2019

guidance in the practical application of the published reports assessed here are lack of information about acute glycemic
these metrics in clinical practice have not at the highest evidence level (25). excursions and the acute complications
been lacking (19). However, there is suggestive evidence of hypo- and hyperglycemia. A1C also
In February 2019, the Advanced Tech- from a number of recent studies, including fails to identify the magnitude and fre-
nologies & Treatments for Diabetes a cross-sectional study correlating current quency of intra- and interday glucose
(ATTD) Congress convened an interna- retrospective 3-day time in target range variation (35,36). Moreover, certain con-
tional panel of individuals with diabetes with varying degrees of diabetes retinop- ditions such as anemia (37), hemoglo-
and clinicians and researchers with ex- athy (26) and an analysis of the 7-point binopathies (38), iron deficiency (39),
pertise in CGM. Our objective was to self-monitored blood glucose (SMBG) and pregnancy (40) can confound A1C
develop clinical CGM targets to supple- data from the Diabetes Control and Com- measurements. Importantly, as reported
ment the currently agreed-upon metrics plications Trial (DCCT) (27), showing cor- by Beck et al. (41), the A1C test can fail at
for CGM-derived times in glucose ranges relations of time in target range (70– times to accurately reflect mean glucose
(within target range, below target range, 180 mg/dL [3.9–10.0 mmol/L]) with di- even when none of those conditions are
above target range) in order to pro- abetes complications. Relationships be- present. Despite these limitations, A1C is
vide guidance for clinicians, researchers, tween time in target range and A1C the only prospectively evaluated tool
and individuals with diabetes in using, (26,27) and number of severe and non- for assessing the risk for diabetes com-
interpreting, and reporting CGM data severe hypoglycemic events (28–32) have plications, and its importance in clinical
in routine clinical care and research. also been observed. Recommendations decision making should not be under-
Importantly, in order to make the rec- from each subgroup were presented to the valued. Rather, the utility of A1C is
ommendations generalizable and compre- full panel and voted upon. This article further enhanced when used as a com-
hensive, the consensus panel included summarizes the consensus recommenda- plement to glycemic data measured by
individuals living with diabetes and had tions and represents the panel members’ CGM.
international representation from physi- evaluation of the issues. Unlike A1C measurement, use of CGM
cians and researchers from all geographic allows for the direct observation of gly-
regions. NEED FOR METRICS BEYOND A1C cemic excursions and daily profiles,
The panel was divided into subgroups A1C is currently recognized as the key which can inform on immediate therapy
to review literature and provide recom- surrogate marker for the development of decisions and/or lifestyle modifications.
mendations for relevant aspects of CGM long-term diabetes complications in peo- CGM also provides the ability to assess
data utilization and reporting among the ple with type 1 and type 2 diabetes and glucose variability and identify patterns
various diabetes populations. Long-term has been used as the primary end point of hypo- and hyperglycemia. However,
trials demonstrating how CGM metrics for many CGM studies (1,3,4,6,33,34). potential drawbacks of CGM use include
relate to and/or predict clinical outcomes While A1C reflects average glucose over the need to be actively used in order to be
have not been conducted, and many of the last 2–3 months, its limitation is the effective; that it may induce anxiety;

9 32
Close Concerns and The diaTribe Foundation, Diabetes Institute, Shanghai Key Laboratory of Biomedical Informatics Consultants LLC, Poto-
San Francisco, CA Diabetes Mellitus, Shanghai Jiao Tong University mac, MD
10 33
Department of Information Engineering, Uni- Affiliated Sixth People’s Hospital, Shanghai, China DiaCare, Ahmedabad, Gujarat, India
22 34
versity of Padova, Padua, Italy Center for Diabetes Technology, University of Pediatric Endocrinology, University of Florida,
11
Harvard John A. Paulson School of Engineering Virginia, Charlottesville, VA Gainesville, FL
23 35
and Applied Sciences, Harvard University, Cam- JDRF, New York, NY dQ&A Market Research, Inc., San Francisco, CA
24 36
bridge, MA Pediatric, Adolescent and Young Adult Section Department of Pediatrics, Nihon University
12
Profil, Neuss, Germany and Section on Clinical, Behavioral and Outcomes School of Medicine, Tokyo, Japan
13 37
Academic Medical Center, University of Am- Research, Joslin Diabetes Center, Harvard Med- Department of Pediatrics, Yale University
sterdam, Amsterdam, the Netherlands ical School, Boston, MA School of Medicine, New Haven, CT
14 25 38
Children’s Hospital at Westmead, University of Endocrinology Research Centre, Moscow, Sackler Faculty of Medicine, Tel Aviv University,
Sydney, Sydney, Australia Russia Tel Aviv, Israel
15 26
University of Colorado Denver and Barbara Clinical and Experimental Endocrinology, KU Corresponding author: Tadej Battelino, tadej
Davis Center for Diabetes, Aurora, CO Leuven, Leuven, Belgium [email protected]
16 27
Grunberger Diabetes Institute, Bloomfield Norwich Medical School, University of East
Hills, MI Anglia, Norwich, U.K. This article contains Supplementary Data online
17
Academic Unit of Diabetes, Endocrinology and 28
Jesse Z and Sara Lea Shafer Institute of Endo- at http://care.diabetesjournals.org/lookup/
Metabolism, University of Sheffield, Sheffield, crinology and Diabetes, National Center for suppl/doi:10.2337/dci19-0028/-/DC1.
U.K. Childhood Diabetes, Schneider Children’s Med- © 2019 by the American Diabetes Association.
18
Science Consulting in Diabetes, Neuss, ical Center of Israel, Petah Tikva, Israel Readers may use this article as long as the work
29
Germany Steno Diabetes Center Copenhagen, Gentofte, is properly cited, the use is educational and not
19
Division of Metabolism, Endocrinology and Denmark for profit, and the work is not altered. More infor-
30
Nutrition, Department of Medicine, University CGParkin Communications, Inc., Henderson, mation is available at http://www.diabetesjournals
of Washington School of Medicine, Seattle, WA NV .org/content/license.
20 31
Wellcome Trust-MRC Institute of Metabolic Department of Endocrinology, Diabetes, and
Science, and Department of Paediatrics, Univer- Nutrition, Montpellier University Hospital; Institute
sity of Cambridge, Cambridge, U.K. of Functional Genomics, University of Montpellier;
21
Department of Endocrinology & Metabolism, and INSERM Clinical Investigation Centre, Mont-
Shanghai Clinical Center of Diabetes, Shanghai pellier, France
care.diabetesjournals.org Battelino and Associates 1595

Table 1—Standardized CGM metrics use of CGM over the most recent 14 days streamline data interpretation, the con-
2017 international consensus on CGM
correlates strongly with 3 months of mean sensus panel identified “time in ranges”
metrics (18) glucose, time in ranges, and hyperglyce- as a metric of glycemic control that pro-
mia metrics (42,43). In individuals with vides more actionable information than
1. Number of days CGM worn
type 1 diabetes, correlations are weaker A1C alone. The panel agreed that estab-
2. Percentage of time CGM is active
for hypoglycemia and glycemic variabil- lishing target percentages of time in the
3. Mean glucose
ity; however, these correlations have various glycemic ranges with the ability to
4. Estimated A1C
not been shown to increase with longer adjust the percentage cut points to ad-
5. Glycemic variability (%CV or SD) sampling periods (43). Longer CGM data dress the specific needs of special di-
6. Time .250 mg/dL (.13.9 mmol/L) collection periods may be required for abetes populations (e.g., pregnancy,
7. Time .180 mg/dL (.10.0 mmol/L) individuals with more variable glycemic high-risk) would facilitate safe and ef-
8. Time 70–180 mg/dL (3.9–10.0 mmol/L) control (e.g., 4 weeks of data to in- fective therapeutic decision making
9. Time ,70 mg/dL (,3.9 mmol/L) vestigate hypoglycemia exposure). within the parameters of the established
10. Time ,54 mg/dL (,3.0 mmol/L) glycemic goals.
11. LBGI and HBGI (risk indices) TIME IN RANGES The metric includes three key CGM
12. Episodes (hypoglycemia and The development of blood glucose test- measurements: percentage of readings
hyperglycemia) 15 min ing provided individuals with diabetes and time per day within target glucose
13. Area under the curve the ability to obtain immediate informa- range (TIR), time below target glucose
14. Time blocks (24-h, day, night) tion about their current glucose levels range (TBR), and time above target glu-
Use of Ambulatory Glucose Profile (AGP) and adjust their therapy accordingly. cose range (TAR) (Table 3). The primary
for CGM report Over the past decades, national and in- goal for effective and safe glucose control
CV, coefficient of variation; LBGI, low blood ternational medical organizations have is to increase the TIR while reducing the
glucose index; HBGI, high blood glucose been successful in developing, harmo- TBR. The consensus group agreed that
index. nizing, and disseminating standardized expressing time in the various ranges can
glycemic targets based on risk for acute be done as the percentage (%) of CGM
and chronic complications. CGM tech- readings, average hours and minutes
that it may have accuracy limitations, nology greatly expands the ability to spent in each range per day, or both,
particularly with the delay in registering assess glycemic control throughout the depending on the circumstances.
blood glucose changes in dynamic sit- day, presenting critical data to inform It was agreed that CGM-based glyce-
uations; and that it can provoke aller- daily treatment decisions and quantify- mic targets must be personalized to meet
gies. Another limitation of CGM is that ing time below, within, and above the the needs of each individual with diabe-
this technology is not yet widely avail- established glycemic targets. tes. In addition, the group reached con-
able in several regions of the world. Although each of the core metrics sensus on glycemic cutpoints (a target
Effective use of CGM data to optimize established in the 2017 ATTD consensus range of 70–180 mg/dL [3.9–10.0 mmol/L]
clinical outcomes requires the user to conference (18) provides important in- for individuals with type 1 diabetes and
interpret the collected data and act upon formation about various aspects of gly- type 2 diabetes and 63–140 mg/dL [3.5–
them appropriately. This requires 1) com- cemic status, it is often impractical to 7.8 mmol/L] during pregnancy, along
mon metrics for assessment of CGM gly- assess and fully utilize many of these with a set of targets for the time per
cemic status, 2) graphical visualization of metrics in real-world clinical practices. To day [% of CGM readings or minutes/
the glucose data and CGM daily profile,
and 3) clear clinical targets. Table 2—Standardized CGM metrics for clinical care: 2019
1. Number of days CGM worn (recommend 14 days) (42,43)
STANDARDIZATION OF CGM 2. Percentage of time CGM is active (recommend 70% of
METRICS data from 14 days) (41,42)
In February 2017, the ATTD Congress 3. Mean glucose
convened an international panel of ex- 4. Glucose management indicator (GMI) (75)
pert clinicians and researchers to define 5. Glycemic variability (%CV) target #36% (90)*
core metrics for assessing CGM data (18) 6. Time above range (TAR): % of readings and time .250 mg/dL
(Table 1). (.13.9 mmol/L) Level 2
The list of core CGM metrics has now 7. Timeaboverange(TAR):%ofreadingsandtime181–250mg/dL
(10.1–13.9 mmol/L) Level 1
been streamlined for use in clinical prac-
8. Time in range (TIR): % of readings and time 70–180 mg/dL
tice based on the expert opinion of this
(3.9–10.0 mmol/L) In range
international consensus group (18). Of
9. Time below range (TBR): %ofreadings andtime54–69mg/dL
the 14 core metrics, the panel selected
(3.0–3.8 mmol/L) Level 1
that 10 metrics that may be most useful
10. Time below range (TBR): % of readings and time ,54 mg/dL
in clinical practice (Table 2). (,3.0 mmol/L) Level 2
Fundamental to accurate and mean- Use of Ambulatory Glucose Profile (AGP) for CGM report
ingful interpretation of CGM is ensuring
that adequate glucose data are available CV, coefficient of variation. *Some studies suggest that lower %CV targets (,33%) provide
additional protection against hypoglycemia for those receiving insulin or sulfonylureas (45,90,91).
for evaluation. As shown in studies, .70%
1596 International Consensus Report Diabetes Care Volume 42, August 2019

Table 3—Guidance on targets for assessment of glycemic control for adults with type 1 or type 2 diabetes and older/high-risk
individuals
TIR TBR TAR
% of readings; % of readings; % of readings;
Diabetes group time per day Target range time per day Below target level time per day Above target level
Type 1*/type 2 .70%; 70–180 mg/dL ,4%; ,70 mg/dL ,25%; .180 mg/dL
.16 h, 48 min (3.9–10.0 mmol/L) ,1 h (,3.9 mmol/L) ,6 h (.10.0 mmol/L)
,1%; ,54 mg/dL ,5%; .250 mg/dL
,15 min (,3.0 mmol/L) ,1 h, 12 min (.13.9 mmol/L)
Older/high-risk# .50%; 70–180 mg/dL ,1%; ,70 mg/dL ,10%; .250 mg/dL
type 1/type 2 .12 h (3.9–10 mmol/L) ,15 min (,3.9 mmol/L) ,2 h, 24 min (.13.9 mmol/L)
Each incremental 5% increase in TIR is associated with clinically significant benefits for individuals with type 1 or type 2 diabetes (26,27). *For age ,25
years, if the A1C goal is 7.5%, set TIR target to approximately 60%. See the section CLINICAL APPLICATION OF TIME IN RANGES for additional information
regarding target goal setting in pediatric management. #See the section OLDER AND/OR HIGH-RISK INDIVIDUALS WITH DIABETES for additional information
regarding target goal setting.

hours]) individuals with type 1 diabetes reaching these targets with CGM in in- that there are different expectations for
and type 2 diabetes (Table 3) and women dividuals using multiple daily injections the various time in ranges relating to
during pregnancy (Table 4) should strive (6). In type 2 diabetes, there is generally safety concerns and efficacy based on
to achieve. It should be noted that pre- less glycemic variability and hypoglycemia currently available therapies and medical
meal and postprandial SMBG targets than in type 1 diabetes (45). Thus, people practice.
remain for diabetes in pregnancy (44), with type 2 diabetes can often achieve
in addition to the new CGM TIR targets more time in the target range while CLINICAL VALIDITY OF MEASURES
for overall glycemia. minimizing hypoglycemia (4). As demon- To fundamentally change clinical care
Although the metric includes TIR, TBR, strated by Beck et al. (4), individuals with with use of the new metrics, it would
and TAR, achieving the goals for both TBR type 2 diabetes increased their TIR by be important to demonstrate that the
and TIR would result in reduced time 10.3% (from 55.6% to 61.3%) after metrics relate to and predict clinical
spent above range and thereby improve 24 weeks of CGM use with slight reduc- outcomes. In this regard, longer-term
glycemic control. However, some clinicians tions in TBR. Most recently, the beneficial studies relating to time spent within
may choose to target the reduction of the effects of new medications, such as so- specific CGM glycemic ranges, diabetes
high glucose values and minimize hypo- dium–glucose cotransporter 2 agents complications, and other outcomes are
glycemia, thereby arriving at more time in have helped individuals with type 1 di- required. However, there is evidence
the target range. In both approaches, the abetes increase TIR (46–48). Targets for from a number of recent studies that
first priority is to reduce TBR to target type 1 diabetes and type 2 diabetes were have shown correlations of TIR (70–
levels and then address TIR or TAR targets. close enough to combine into one set of 180 mg/dL [3.9–10.0 mmol/L]) with di-
Note that for people with type 1 di- targets, outside of pregnancy. abetes complications (49,50) as well as
abetes, the targets are informed by the Another way to visualize the CGM- a relationship between TIR and A1C
ability to reach the targets with hybrid derived targets for the four categories of (26,27). Although evidence regarding TIR
closed-loop therapy (11), the first exam- diabetes is shown in Fig. 1, which displays for older and/or high-risk individuals is
ple of which is now commercially avail- and compares the targets for TIR (green), lacking, numerous studies have shown
able with several more systems in final TBR (two categories in light and dark the elevated risk for hypoglycemia in
stages of testing. Importantly, recent red), and TAR (two categories in yellow these populations (51–56). Therefore,
studies have shown the potential of and orange). It becomes clear at a glance we have lowered the TIR target from

Table 4—Guidance on targets for assessment of glycemic control during pregnancy

TIR TBR TAR
% of readings; % of readings; Below target % of readings; Above target
Diabetes group time per day Target range time per day level time per day level
Pregnancy, .70%; 63–140 mg/dL† ,4%; ,63 mg/dL† ,25%; .140 mg/dL
type 1§ .16 h, 48 min (3.5–7.8 mmol/L†) ,1 h (,3.5 mmol/L†) ,6 h (.7.8 mmol/L)
,1%; ,54 mg/dL
,15 min (,3.0 mmol/L)
Pregnancy, See PREGNANCY 63–140 mg/dL† See PREGNANCY ,63 mg/dL† See PREGNANCY .140 mg/dL
type 2/GDM§ section (3.5–7.8 mmol/L†) section (,3.5 mmol/L†) section (.7.8 mmol/L)
,54 mg/dL
(,3.0 mmol/L)
Each incremental 5% increase in TIR is associated with clinically significant benefits for pregnancy in women with type 1 diabetes (59,60). †Glucose
levels are physiologically lower during pregnancy. §Percentages of TIR are based on limited evidence. More research is needed.
care.diabetesjournals.org Battelino and Associates 1597

Figure 1—CGM-based targets for different diabetes populations.

.70% to .50% and reduced TBR to that individuals with more advanced DR (58). TBR ,50 mg/dL was reduced
,1% at ,70 mg/dL (,3.9 mmol/L) to spent significantly less time within target from 6% to 4%, although the higher
place greater emphasis on reducing hy- range (70–180 mg/dL [3.9–10.0 mmol/L]) TBR ,70 mg/dL was high (13–15%) us-
poglycemia with less emphasis on main- and that prevalence of DR decreased with ing older-generation sensors. With im-
taining target glucose levels (Table 3). increasing TIR. proved sensor accuracy, recent type 1
Relationship Between TIR and A1C diabetes pregnancy studies report a
Type 1 Diabetes and Type 2 Diabetes Analyses were conducted utilizing data- lower threshold of ,63 mg/dL (,3.5
Association With Complications sets from four randomized trials encom- mmol/L) for TBR and $63 mg/dL
Associations between TIR and progres- passing 545 adults with type 1 diabetes ($3.5 mmol/L) for TIR (59,60). Data
sion of both diabetic retinopathy (DR) who had central laboratory measure- from Sweden, and the Continuous Glu-
and development of microalbuminuria ments of A1C (26). TIR (70–180 mg/dL cose Monitoring in Women With Type 1
were reported by Beck et al. (50), using [3.9–10.0 mmol/L]) of 70% and 50% Diabetes in Pregnancy Trial (CONCEPTT)
7-point blood glucose profiles from the strongly corresponded with an A1C of control group, report 50% TIR in the
DCCT data set to validate the use of TIR approximately 7% (53 mmol/mol) and 8% first trimester, improving to 60% TIR
as an outcome measure for clinical trials. (64 mmol/mol), respectively. An increase in the third trimester, reflecting contem-
Their analysis showed that the hazard in TIR of 10% (2.4 h per day) corre- porary antenatal care. Of note, these
rate for retinopathy progression in- sponded to a decrease in A1C of approx- data confirm that the TBR ,63 mg/dL
creased by 64% for each 10% reduction imately 0.5% (5.0 mmol/mol); similar (,3.5 mmol/L) recommendation of ,4%
in TIR. The hazard rate for microalbumin- associations were seen in an analysis is safely achievable, especially after the
uria development increased by 40% of 18 randomized controlled trials (RCTs) first trimester. Furthermore, 33% of
for each 10% reduction in TIR. A post by Vigersky and McMahon (27) that in- women achieved the recommendation of
hoc analysis of the same DCCT data cluded over 2,500 individuals with type 1 70% TIR 63–140 mg/dL (3.5–7.8 mmol/L)
set showed a link between glucose of diabetes and type 2 diabetes over a wide in the final (.34) weeks of pregnancy.
,70 mg/dL (,3.9 mmol/L) and ,54 range of ages and A1C levels (Table 5). Preliminary data suggest that closed-
mg/dL (,3.0 mmol/L) and an increased loop systems may allow pregnant women
risk for severe hypoglycemia (57). Pregnancy to safely achieve 70% TIR at an earlier
Similar associations between DR and During pregnancy, the goal is to safely (.24 weeks) stage of gestation (61,62).
TIR were reported in a recent study by Lu increase TIR as quickly as possible, while Law et al. (63) analyzed data from two
et al. (49) in which 3,262 individuals with reducing TAR and glycemic variability. early CGM trials (64,65) describing the
type 2 diabetes were evaluated for DR, Data from the first study of longitudinal associations between CGM measures
which was graded as non-DR, mild non- CGM use in pregnancy demonstrated a and risk of large-for-gestational-age (LGA)
proliferative DR (NPDR), moderate NPDR, 13–percentage point increase in TIR (43% infants. Taken together, the Swedish and
or vision-threatening DR. Results showed to 56% TIR 70–140 mg/dL [3.9–7.8 mmol/L]) CONCEPTT data confirm that a 5–7% higher
1598 International Consensus Report Diabetes Care Volume 42, August 2019

Table 5—Estimate of A1C for a given TIR level based on type 1 diabetes and type 2 diabetes studies
Vigersky and McMahon (27) (n = 1,137
Beck et al. (26) (n = 545 participants with type 1 diabetes) participants with type 1 or type 2 diabetes)
TIR 70–180 mg/dL A1C, % 95% CI for predicted TIR 70–180 mg/dL A1C, %
(3.9–10.0 mmol/L) (mmol/mol) A1C values, % (3.9–10.0 mmol/L) (mmol/mol)
20% 9.4 (79) (8.0, 10.7) 20% 10.6 (92)
30% 8.9 (74) (7.6, 10.2) 30% 9.8 (84)
40% 8.4 (68) (7.1, 9.7) 40% 9.0 (75)
50% 7.9 (63) (6.6, 9.2) 50% 8.3 (67)
60% 7.4 (57) (6.1, 8.8) 60% 7.5 (59)
70% 7.0 (53) (5.6, 8.3) 70% 6.7 (50)
80% 6.5 (48) (5.2, 7.8) 80% 5.9 (42)
90% 6.0 (42) (4.7, 7.3) 90% 5.1 (32)
Every 10% increase in TIR = ;0.5% (5.5 mmol/mol) A1C reduction Every 10% increase in TIR = ;0.8%
(8.7 mmol/mol) A1C reduction
The difference between findings from the two studies likely stems from differences in number of studies analyzed and subjects included (RCTs
with subjects with type 1 diabetes vs. RCTs with subjects with type 1 or type 2 diabetes with CGM and SMBG).

TIR during the second and third trimesters assisted care, which can complicate report, glucose ranges are defined as
is associated with decreased risk of LGA treatment regimens (56). Therefore, “Very High” (Level 2), “High” (Level 1),
and neonatal outcomes including macro- when setting glycemic targets for high- “Low” (Level 1), and “Very Low” (Level 2).
somia, shoulder dystocia, neonatal hy- risk and/or elderly people, it is important An “mmol/L” version is provided in
poglycemia, and neonatal intensive care to individualize and be conservative, Supplementary Fig. 1.
admissions. More data are needed to with a strong focus on reducing the per- There is a general consensus that a
define the clinical CGM targets for preg- centage of time spent ,70 mg/dL (,3.9 useful CGM report is one that can be
nant women with type 2 diabetes, who mmol/L) and preventing excessive hy- understood by clinicians and people with
spend one-third less time hyperglycemic perglycemia. diabetes. While there may be some terms
than women with type 1 diabetes and (e.g., glucose variability) that are less
achieve TIR of 90% (58). Because of the STANDARDIZATION OF CGM DATA familiar to many people with diabetes, a
lack of evidence on CGM targets for PRESENTATION single-page report that the medical team
women with gestational diabetes mellitus As noted above, in 2013 a panel of can review and file in the electronic medical
(GDM) or type 2 diabetes in pregnancy, clinicians with expertise in CGM pub- record and that can be used as a shared
percentages of time spent in range, below lished recommendations for use of the decision-making tool with people with di-
range, and above range have not been Ambulatory Glucose Profile (AGP) as a abetes was considered to be of value
included in this report. Recent data sug- template for data presentation and vi- (69–72). More detailed reports (e.g., ad-
gest that even more stringent targets sualization. Originally created by Mazze justable data ranges, detailed daily reports)
(66) and greater attention to overnight et al. (67), the standardized AGP report should remain available for individualized
glucose profiles may be required to was further developed by the Interna- review by or with people with diabetes.
normalize outcomes in pregnant women tional Diabetes Center and now incor-
with GDM (63). porates all the core CGM metrics and Clinical Application of Time in Ranges
targets along with a 14-day composite Despite its demonstrated value, clinical
Older and/or High-Risk Individuals glucose profile as an integral component utilization of CGM data has remained
With Diabetes of clinical decision making (24). This suboptimal. Although time constraints
Older and/or high-risk individuals with recommendation was later endorsed and reimbursement issues are clearly
diabetes are at notably higher risk for at the aforementioned international obstacles, clinician inexperience in data
severe hypoglycemia due to age, dura- consensus conference on CGM metrics interpretation and lack of standardiza-
tion of diabetes, duration of insulin (18) and is referenced as an example tion software for visualization of CGM
therapy, and greater prevalence of hy- in the American Diabetes Association data have also played a role (73). The
poglycemia unawareness (51–55). The 2019 “Standards of Medical Care in Di- proposed standardized report enables
increased risk of severe hypoglycemia abetes” (16) and in an update to the clinicians to readily identify important
is compounded by cognitive and physical American Association of Clinical Endo- metrics such as the percentage of time
impairments and other comorbidities crinologists consensus on use of CGM spent within, below, and above each
(53,56). High-risk individuals include (68). The AGP report, in slightly modified individual’s target range, allowing for
those with a higher risk of complica- formats, has been adopted by most of greater personalization of therapy through
tions, comorbid conditions (e.g., cogni- the CGM device manufacturers in their shared decision making.
tive deficits, renal disease, joint disease, download software. An example of the Using the standardized report, the
osteoporosis, fracture, and/or cardio- AGP report, updated to incorporate tar- clinician can also address glucose vari-
vascular disease), and those requiring gets, is presented in Fig. 2. In the AGP ability (e.g., the coefficient of variation
care.diabetesjournals.org Battelino and Associates 1599

Figure 2—Ambulatory Glucose Profile.

1600 International Consensus Report Diabetes Care Volume 42, August 2019

[%CV] metric) (74) or use the glucose of working to reduce the time spent CONCLUSIONS
management indicator (GMI) metric (75) ,70 mg/dL (,3.9 mmol/L) to less Use of CGM continues to expand in
to discuss the possible discrepancies than 1 h per day and time spent clinical practice. As a component of di-
noted in glucose exposure derived from ,54 mg/dL (,3.0 mmol/L) to less abetes self-management, daily use of
CGM data versus the individual’s laboratory- than 15 min per day, rather than us- CGM provides the ability to obtain im-
measured A1C (41,76). With appropriate ing ,4% and ,1%, respectively, as the mediate feedback on current glucose
educational materials, time, and experi- goal. However, as discussed earlier, targets levels as well as direction and rate of
ence, clinicians will develop a systematic must be personalized to meet the needs change in glucose levels. This information
approach to CGM data analysis and the and capabilities of each person, focusing allows people with diabetes to optimize
most effective ways to discuss the data on small steps and small successes. Indi- dietary intake and exercise, make in-
with patients in person or remotely. viduals with diabetes should work with formed therapy decisions regarding
their provider and/or educator to develop mealtime and correction of insulin dos-
Goal Setting a SMART goal to reduce TBR. ing, and, importantly, react immediately
Numerous studies have demonstrated Individualized goals are particularly and appropriately to mitigate or prevent
the clinical benefits of early achievement important for pediatric and young adult acute glycemic events (87–89). Retro-
of near-normal glycemic control in indi- populations. The International Society spective analysis of CGM data, using
viduals with type 1 diabetes and type 2 for Pediatric and Adolescent Diabetes standardized data management tools
diabetes (77–83). However, when advis- recommends that targets for individuals such as the AGP, enables clinicians
ing people with diabetes, goal-setting #25 years of age aim for the lowest and people with diabetes to work col-
must be collaborative and take into ac- achievable A1C without undue exposure laboratively in identifying problem areas
count the individual needs/capabilities of to severe hypoglycemia or negative ef- and then set achievable goals (70–72).
each patient and start with the goals that fects on quality of life and burden of care We conclude that, in clinical practice,
are most achievable. An early study by (86). An A1C target of 7.0% (53 mmol/ time in ranges (within target range, be-
DeWalt et al. (84) found that setting small, mol) can be used in children, adolescents, low range, above range) are both appro-
achievable goals not only enhances peo- and adults #25 years old who have priate and useful as clinical targets and
ple’s ability to cope with their diabetes, but access to comprehensive care (86). outcome measurements that comple-
that people with diabetes who set and However, a higher A1C goal (e.g., ,7.5% ment A1C for a wide range of people
achieved their goals often initiated addi- [,58 mmol/mol]) may be more appro- with diabetes and that the target values
tional behavioral changes on their own. priate in the following situations: inability specified in this article should be con-
One approach to consider is the SMART to articulate hypoglycemia symptoms, sidered an integral component of CGM
goal (Specific, Measurable, Achievable, hypoglycemia unawareness, history of data analysis and day-to-day treatment
Relevant, Time-bound) intervention, which severe hypoglycemia, lack of access to decision making.
is directly applicable to setting targets for analog insulins and/or advanced insulin
time in ranges. First described by Lawlor delivery technology, or inability to regu-
and Hornyak in 2012 (85), this approach larly check glucose (86). This would Acknowledgments. The consensus group par-
incorporates four key components of equate to a TIR target of ;60% (Table 4). ticipants wish to thank the ATTD Congress for
behavioral change relevant to goal set- The consensus group recognized that organizing and coordinating the meeting and
ting: 1) the goal is specific and defines achieving the targets for the various time Rachel Naveh (The Jesse Z and Sara Lea Shafer
Institute for Endocrinology and Diabetes, Na-
exactly what is to be achieved, 2) the goal in ranges is aspirational in some situa- tional Center for Childhood Diabetes, Schneider
is measurable and there is tangible evi- tions, and many individuals will require Children’s Medical Center of Israel) for assistance
dence when it has been achieved, 3) the ongoing support, both educational and in organizing the meeting. They also thank
goal is achievable but stretches the pa- technological, from their health care Courtney Lias from the U.S. Food and Drug
tient slightly so that he/she feels chal- team. Importantly, as demonstrated by Administration for her participation as an ob-
server at the consensus conference.
lenged, and 4) the goal should be Beck et al. (26), Vigersky and McMahon Funding and Duality of Interest. Support for
attainable over a short period of time. (27), and Feig et al. (59), even small, the CGM consensus conference and develop-
Effective goals should utilize CGM data incremental improvements yield signifi- ment of this consensus report was provided by
to identify specific instances for the cant glycemic benefits. Therefore, when the ATTD Congress. Abbott Diabetes Care, Astra
Zeneca, Dexcom Inc., Eli Lilly and Company,
patient to take measurable action to advising individuals with diabetes (par-
Insulet Corporation, Medtronic, Novo Nordisk,
prevent hypoglycemia. Although analysis ticularly children, adolescents, and high- Roche Diabetes Care, and Sanofi provided fund-
of the AGP reports provides an oppor- risk individuals) about their glycemic goals, ing to ATTD to support the consensus meeting.
tunity for meaningful discussion, individ- it is important to take a stepwise ap- Consensus participants were reimbursed for
uals should be counseled to look at proach, emphasizing that what may ap- travel to the ATTD conference and one night
of lodging; no honoraria were provided. ATTD
patterns throughout the day to see pear to be small, incremental successes provided funding to Christopher G. Parkin,
when low glucose events are occurring (e.g., 5% increase in TIR) are, in fact, CGParkin Communications, Inc., for his medical
and make adjustments in their therapy clinically significant in improving their writing and editorial support. T.Ba. has received
to reduce these events. glycemia (26,27,59). However, when coun- honoraria for participation on advisory boards
When applying the CGM metrics in seling women planning pregnancy and for Novo Nordisk, Sanofi, Eli Lilly and Company,
Boehringer, Medtronic, and Bayer Health Care and
clinical practice, it may be more mean- pregnant women, greater emphasis should as a speaker for AstraZeneca, Eli Lilly and Com-
ingful and motivating to communicate be placed on getting to goal as soon as pany, Bayer, Novo Nordisk, Medtronic, Sanofi,
to people with diabetes the importance possible (59,60). and Roche. T.Ba. owns stocks of DreaMed
care.diabetesjournals.org Battelino and Associates 1601

Diabetes, and his institution has received research Boehringer Ingelheim, MannKind, Sanofi, Zea- by M.P. received research support from Med-
grant support and travel expenses from Abbott land Pharma, and UNEEG. L.H. is a consultant tronic, Novo Nordisk, Eli Lilly and Company,
Diabetes Care, Medtronic, Novo Nordisk, Glu- for companies developing novel diagnostic and Dexcom, Sanofi, Insulet, OPKO Health, DreaMed
Sense, Sanofi, Sandoz, and Diamyd. T.D. has therapeutic options for diabetes. He is a share- Diabetes, Bristol-Myers Squibb, and Merck. M.P.
received speaker honoraria, research support, holder of Profil Institut für Stoffwechselforschung is a stockholder/shareholder of DreaMed Dia-
and consulting fees from Abbott Diabetes Care, GmbH and ProSciento. I.B.H. receives research betes, NG Solutions, and Nutriteen Professionals
Bayer, Bristol-Myers Squibb, AstraZeneca, Boeh- funding from Medtronic Diabetes and has re- and reports two patent applications. No other
ringer Ingelheim, Dexcom, Eli Lilly and Company, ceived consulting fees from Abbott Diabetes potential conflicts of interest relevant to this
Medtronic, Novo Nordisk, Sanofi, and Roche Care, Bigfoot, Roche, and Becton Dickinson. article were reported.
Diabetes Care; and he is a shareholder of R.H. reports having received speaker honoraria
DreaMed Diabetes. S.A.A. has received honoraria from Eli Lilly and Company, Novo Nordisk, and References
for participation on advisory boards for Roche AstraZeneca, serving on an advisory panel for Eli 1. Lind M, Polonsky W, Hirsch IB, et al. Contin-
and Medtronic and has given a lecture funded by Lilly and Company and Novo Nordisk, and re- uous glucose monitoring vs conventional therapy
Sanofi. R.B. is an employee the Jaeb Center for ceiving license fees from B. Braun and Medtronic. for glycemic control in adults with type 1 diabetes
Health Research, which has received grant sup- O.K. received honoraria from Amring, Eli Lilly and treated with multiple daily insulin injections: the
port from Dexcom, Animas, Bigfoot, and Tan- Company, Novo Nordisk, and Sanofi and owns GOLD randomized clinical trial. JAMA 2017;317:
dem; nonfinancial study support from Dexcom, shares from DreaMed Diabetes. B.K. declares 379–387
Abbott Diabetes Care, and Roche Diabetes Care; research support handled by the University of 2. Aleppo G, Ruedy KJ, Riddlesworth TD, et al.;
and consulting fees from Eli Lilly and Company Virginia from Dexcom, Roche, Sanofi, and Tan- REPLACE-BG Study Group. REPLACE-BG: A ran-
and Insulet. He has no personal financial arrange- dem; patent royalties handled by the University domized trial comparing continuous glucose
ments with any company. R.M.B. has received of Virginia from Johnson & Johnson, Sanofi, and monitoring with and without routine blood
research funding and served as a consultant and Dexcom; and has served as a consultant for Sanofi glucose monitoring in adults with well-controlled
on advisory boards for Abbott Diabetes Care, and Tandem and on a speakers’ bureau for type 1 diabetes. Diabetes Care 2017;40:538–545
Becton Dickinson, Dexcom, Eli Lilly and Company, Dexcom. B.L. is an employee of Close Concerns 3. Beck RW, Riddlesworth T, Ruedy K, et al.;
Glooko, Helmsley Charitable Trust, Hygieia, and The diaTribe Foundation, which receive DIAMOND Study Group. Effect of continuous
Johnson&Johnson,Medtronic,Merck,NovoNordisk, funding from CGM manufacturers, including glucose monitoring on glycemic control in adults
Roche, Sanofi, and Senseonics. His employer, Medtronic, Dexcom, and Abbott Diabetes Care. with type 1 diabetes using insulin injections: The
nonprofit HealthPartners Institute, contracts C.M. serves or has served on the advisory panel DIAMOND randomized clinical trial. JAMA 2017;
for his services and no personal income goes to or speakers’ bureau for Novo Nordisk, Sanofi, 317:371–378
R.M.B. E.B. received honoraria for participation Merck Sharp & Dohme, Eli Lilly and Company, 4. Beck RW, Riddlesworth TD, Ruedy K, et al.;
on advisory boards and speakers’ bureaus from Novartis, AstraZeneca, Boehringer Ingelheim, DIAMOND Study Group. Continuous glucose
Abbott Diabetes Care, AstraZeneca, Medtronic, Hanmi Pharmaceuticals, Roche, Medtronic, monitoring versus usual care in patients with
Novartis, Roche, and Sanofi. B.A.B. is on medical ActoBio Therapeutics, Pfizer, Dianax, and UCB. type 2 diabetes receiving multiple daily insulin
advisory boards for Medtronic and Convatec and Financial compensation for these activities has injections: a randomized trial. Ann Intern Med
has received research funding from the National been received by KU Leuven. H.R.M. received 2017;167:365–374
Institutes of Health, JDRF, the Leona M. and Harry honoraria from participation on advisory boards 5. Polonsky WH, Hessler D, Ruedy KJ, Beck RW;
B. Helmsley Charitable Trust, Medtronic Diabe- for Medtronic and research support from Dex- DIAMOND Study Group. The impact of contin-
tes, ConvaTec, Dexcom, Tandem, and Insulet. com, Medtronic, Abbott Diabetes Care, and uous glucose monitoring on markers of quality
K.L.C. is an employee of Close Concerns and The Johnson & Johnson. R.N. received honoraria of life in adults with type 1 diabetes: further
diaTribe Foundation, which receive funding from for participation on the speakers’ bureau of findings from the DIAMOND randomized clinical
CGM manufacturers, including Medtronic, Dex- Novo Nordisk, Pfizer, Eli Lilly and Company, trial. Diabetes Care 2017;40:736–741
com, and Abbott Diabetes Care. C.C. reports and Sanofi. R.N. owns DreaMed stock and reports 6. Šoupal J, Petruželková L, Flekač M, et al.
10 patents and patent applications related to two patent applications. K.N. owns shares in Comparison of different treatment modalities
continuous glucose sensors and artificial pan- Novo Nordisk and has received consulting fees for type 1 diabetes, including sensor-augmented
creas. E.D. has received consulting fees and from Medtronic, Abbott Diabetes Care, and Novo insulin regimens, in 52 weeks of follow-up:
honoraria for participation on advisory boards Nordisk; speaker honoraria from Medtronic, a COMISAIR study. Diabetes Technol Ther
from Roche, Insulet, and Eli Lilly and Company Roche Diabetes Care, Rubin Medical, Sanofi, 2016;18:532–538
and research support from Dexcom, Insulet, Ani- Novo Nordisk, Zealand Pharma, and Bayer; 7. van Beers CA, DeVries JH, Kleijer SJ, et al.
mas, Xeris, and Roche. J.H.D. has received speaker and research support from Novo Nordisk, Zea- Continuous glucose monitoring for patients with
honoraria and research support from and has land Pharma, Medtronic, and Roche Diabetes type 1 diabetes and impaired awareness of
consulted for Abbott Diabetes Care, Dexcom, Care. C.G.P. has received consulting fees from hypoglycaemia (IN CONTROL): a randomised,
Medtronic, Merck Sharp & Dohme, Novo Nor- Dexcom, Diasome, Onduo, Proteus, Roche Di- open-label, crossover trial. Lancet Diabetes En-
disk, Sanofi, Roche, Senseonics, and Zealand. abetes Care, and Senseonics. E.R. has received docrinol 2016;4:893–902
F.J.D. has received consulting fees from Mod- consulting fees from A. Menarini Diagnostics, 8. Bolinder J, Antuna R, Geelhoed-Duijvestijn P,
eAGC and research support from Dexcom, In- Abbott Diabetes Care, Air Liquide SI, Becton Kröger J, Weitgasser R. Novel glucose-sensing
sulet, Animas, and Xeris. S.G. has received Dickinson, Cellnovo, Dexcom, Eli Lilly and Com- technology and hypoglycaemia in type 1 diabe-
consulting fees and honoraria for participation pany, Insulet, Johnson & Johnson, Medtronic, tes: a multicentre, non-masked, randomised
on advisory boards for Medtronic, Roche Di- Novo Nordisk, Roche, and Sanofi and research controlled trial. Lancet 2016;388:2254–2263
abetes Care, Merck, Lexicon, Novo Nordisk, support from Abbott Diabetes Care, Dexcom, 9. Haak T, Hanaire H, Ajjan R, Hermanns N,
Sanofi, MannKind, Senseonics, Zealand, and Eli Insulet, Tandem, and Roche. D.R. has received Riveline JP, Rayman G. Flash glucose-sensing
Lilly and Company and research grants from Eli consulting fees from Eli Lilly and Company and technology as a replacement for blood glucose
Lilly and Company, Novo Nordisk, Merck, Lexi- Better Therapeutics. K.S. is an employee of dQ&A monitoring for the management of insulin-treated
con, Medtronic, Dario, the National Cancer In- Market Research, Inc., whose clients include type 2 diabetes: a multicenter, open-label ran-
stitute, T1D Exchange, the National Institute of several device and pharmaceutical companies domized controlled trial. Diabetes Ther 2017;8:
Diabetes and Digestive and Kidney Diseases, in the diabetes field. S.A.W. has received con- 55–73
JDRF, Animas, Dexcom, and Sanofi. G.G. has sulting fees from Eli Lillyand Company, Sanofi, 10. Choudhary P, Olsen BS, Conget I, Welsh JB,
received research support from Novo Nordisk and Zealand and speaker honoraria from Med- Vorrink L, Shin JJ. Hypoglycemia prevention and
and Medtronic and honoraria for participation tronic, Insulet, and Tandem. M.P. is a member of user acceptance of an insulin pump system with
on speakers’ bureaus from Novo Nordisk, Eli Lilly the advisory board of AstraZeneca, Sanofi, Med- predictive low glucose management. Diabetes
and Company, Boehringer Ingelheim, and Sanofi. tronic, Eli Lilly and Company, Novo Nordisk, and Technol Ther 2016;18:288–291
S.H. has served as a consultant or speaker for Insulet and is a consultant to RSP Systems A/S, 11. Bergenstal RM, Garg S, Weinzimer SA, et al.
Eli Lilly and Company, Novo Nordisk, Takeda, Qulab Medical, and Pfizer. The institute headed Safety of a hybrid closed-loop insulin delivery
1602 International Consensus Report Diabetes Care Volume 42, August 2019

system in patients with type 1 diabetes. JAMA 24. Bergenstal RM, Ahmann AJ, Bailey T, et al. 38. Bry L, Chen PC, Sacks DB. Effects of hemo-
2016;316:1407–1408 Recommendations for standardizing glucose re- globin variants and chemically modified deriva-
12. Heinemann L, Freckmann G, Erdmann D, porting and analysis to optimize clinical decision tives on assays for glycohemoglobin. Clin Chem
et al. Real-time continuous glucose monitoring making in diabetes: the Ambulatory Glucose 2001;47:153–163
use in adults with type 1 diabetes and impaired Profile (AGP). Diabetes Technol Ther 2013;15: 39. Ford ES, Cowie CC, Li C, Handelsman Y,
hypoglycaemia awareness or severe hypoglycae- 198–211 Bloomgarden ZT. Iron-deficiency anemia, non-
mia treated with multiple daily insulin injections 25. American Diabetes Association. Introduc- iron-deficiency anemia and HbA1c among adults
(HypoDE): a multicentre, randomised controlled tion: Standards of Medical Care in Diabetesd in the US. J Diabetes 2011;3:67–73
trial. Lancet 2018;391:1367–1377 2017. Diabetes Care 2017;41(Suppl. 1):S1–S2 40. Nielsen LR, Ekbom P, Damm P, et al. HbA1c
13. Reddy M, Jugnee N, El Laboudi A, 26. Beck RW, Bergenstal RM, Cheng P, et al. The levels are significantly lower in early and late
Spanudakis E, Anantharaja S, Oliver N. A ran- relationships between time in range, hypergly- pregnancy. Diabetes Care 2004;27:1200–1201
domised controlled pilot study of continuous cemia metrics, and HbA1c. J Diabetes Sci Technol. 41. Beck RW, Connor CG, Mullen DM, Wesley
glucose monitoring and flash glucose monitoring 13 January 2019 [Epub ahead of print]. DOI: DM, Bergenstal RM. The fallacy of average: how
in people with type 1 diabetes and impaired 10.1177/1932296818822496 using HbA1c alone to assess glycemic control can
awareness of hypoglycaemia. Diabet Med 2018; 27. Vigersky RA, McMahon C. The relationship of be misleading. Diabetes Care 2017;40:994–999
35:483–490 hemoglobin A1C to time-in-range in patients 42. Xing D, Kollman C, Beck RW, et al.; Juvenile
14. Battelino T, Nimri R, Dovc K, Phillip M, with diabetes. Diabetes Technol Ther 2019;21: Diabetes Research Foundation Continuous Glu-
Bratina N. Prevention of hypoglycemia with 81–85 cose Monitoring Study Group. Optimal sampling
predictive low glucose insulin suspension in 28. Brod M, Christensen T, Thomsen TL, intervals to assess long-term glycemic control
children with type 1 diabetes: a randomized Bushnell DM. The impact of non-severe hy- using continuous glucose monitoring. Diabetes
controlled trial. Diabetes Care 2017;40:764–770 poglycemic events on work productivity and Technol Ther 2011;13:351–358
15. Dovc K, Cargnelutti K, Sturm A, Selb J, Bratina diabetes management. Value Health 2011;14: 43. Riddlesworth TD, Beck RW, Gal RL, et al.
N, Battelino T. Continuous glucose monitoring 665–671 Optimal sampling duration for continuous
use and glucose variability in pre-school children 29. Brod M, Rana A, Barnett AH. Impact of self- glucose monitoring to determine long-term gly-
with type 1 diabetes. Diabetes Res Clin Pract treated hypoglycaemia in type 2 diabetes: a mul- cemic control. Diabetes Technol Ther 2018;20:
2019;147:76–80 tinational survey in patients and physicians. Curr 314–316
16. American Diabetes Association. 7. Diabe- Med Res Opin 2012;28:1947–1958 44. American Diabetes Association. 14. Manage-
tes technology: Standards of Medical Care in 30. Seaquist ER, Anderson J, Childs B, et al. ment of diabetes in pregnancy. Standards of
Diabetesd2019. Diabetes Care 2019;42(Suppl. Hypoglycemia and diabetes: a report of a work- Medical Care in Diabetesd2019. Diabetes Care
1):S71–S80 group of the American Diabetes Association and 2019;42(Suppl. 1):S165–S172
17. Fonseca VA, Grunberger G, Anhalt H, et al.; the Endocrine Society. Diabetes Care 2013;36: 45. Rama Chandran S, Tay WL, Lye WK, et al.
Consensus Conference Writing Committee. Con- 1384–1395 Beyond HbA1c: comparing glycemic variability
tinuous glucose monitoring: a consensus con- 31. International Hypoglycaemia Study Group. and glycemic indices in predicting hypoglycemia
ference of the American Association of Clinical Glucose concentrations of less than 3.0 mmol/L in type 1 and type 2 diabetes. Diabetes Technol
Endocrinologists and American College of Endo- (54 mg/dL) should be reported in clinical trials: Ther 2018;20:353–362
crinology. Endocr Pract 2016;22:1008–1021 a joint position statement of the American Di- 46. Famulla S, Pieber TR, Eilbracht J, et al.
18. Danne T, Nimri R, Battelino T, et al. In- abetes Association and the European Association Glucose exposure and variability with empagli-
ternational consensus on use of continuous for the Study of Diabetes. Diabetes Care 2017; flozin as adjunct to insulin in patients with type 1
glucose monitoring. Diabetes Care 2017;40: 40:155–157 diabetes: continuous glucose monitoring data
1631–1640 32. Novodvorsky P, Bernjak A, Chow E, et al. from a 4-week, randomized, placebo-controlled
19. Petrie JR, Peters AL, Bergenstal RM, Holl RW, Diurnal differences in risk of cardiac arrhythmias trial (EASE-1). Diabetes Technol Ther 2017;19:
Fleming GA, Heinemann L. Improving the clinical during spontaneous hypoglycemia in young 49–60
value and utility of CGM Systems: issues and people with type 1 diabetes. Diabetes Care 47. Dandona P, Mathieu C, Phillip M, et al.;
recommendations: a joint statement of the 2017;40:655–662 DEPICT-1 Investigators. Efficacy and safety of
European Association for the Study of Diabetes 33. Battelino T, Conget I, Olsen B, et al.; SWITCH dapagliflozin in patients with inadequately con-
and the American Diabetes Association Diabe- Study Group. The use and efficacy of continu- trolled type 1 diabetes: the DEPICT-1 52-week
tes Technology Working Group. Diabetes Care ous glucose monitoring in type 1 diabetes study. Diabetes Care 2018;41:2552–2559
2017;40:1614–1621 treated with insulin pump therapy: a rando- 48. Mathieu C, Dandona P, Phillip M, et al.;
20. Agiostratidou G, Anhalt H, Ball D, et al. mised controlled trial. Diabetologia 2012;55: DEPICT-1 and DEPICT-2 Investigators. Glucose
Standardizing clinically meaningful outcome 3155–3162 variables in type 1 diabetes studies with dapagli-
measures beyond HbA1c for type 1 diabetes: 34. Bergenstal RM, Tamborlane WV, Ahmann A, flozin: pooled analysis of continuous glucose
a consensus report of the American Association et al.; STAR 3 Study Group. Sensor-augmented monitoring data from DEPICT-1 and -2. Diabetes
of Clinical Endocrinologists, the American Asso- pump therapy for A1C reduction (STAR 3) study: Care 2019;42:1081–1087
ciation of Diabetes Educators, the American results from the 6-month continuation phase. 49. Lu J, Ma X, Zhou J, et al. Association of time
Diabetes Association, the Endocrine Society, Diabetes Care 2011;34:2403–2405 in range, as assessed by continuous glucose
JDRF International, The Leona M. and Harry B. 35. Cox DJ, Kovatchev BP, Julian DM, et al. monitoring, with diabetic retinopathy in type
Helmsley Charitable Trust, the Pediatric Endo- Frequency of severe hypoglycemia in insulin- 2 diabetes. Diabetes Care 2018;41:2370–2376
crine Society, and the T1D Exchange. Diabetes dependent diabetes mellitus can be predicted 50. Beck RW, Bergenstal RM, Riddlesworth TD,
Care 2017;40:1622–1630 from self-monitoring blood glucose data. J Clin et al. Validation of time in range as an outcome
21. Sherr JL, Tauschmann M, Battelino T, et al. Endocrinol Metab 1994;79:1659–1662 measure for diabetes clinical trials. Diabetes
ISPAD Clinical Practice Consensus Guidelines 36. Qu Y, Jacober SJ, Zhang Q, Wolka LL, DeVries Care 2019;42:400–405
2018: diabetes technologies. Pediatr Diabetes JH. Rate of hypoglycemia in insulin-treated pa- 51. Weinstock RS, DuBose SN, Bergenstal RM,
2018;19(Suppl. 27):302–325 tients with type 2 diabetes can be predicted from et al.; T1D Exchange Severe Hypoglycemia in
22. Foster NC, Beck RW, Miller KM, et al. State of glycemic variability data. Diabetes Technol Ther Older Adults With Type 1 Diabetes Study Group.
type 1 diabetes management and outcomes from 2012;14:1008–1012 Risk factors associated with severe hypoglycemia
the T1D Exchange in 2016–2018. Diabetes Technol 37. National Institute of Diabetes and Digestive in older adults with type 1 diabetes. Diabetes
Ther 2019;21:66–72 and Kidney Diseases Health Information Center. Care 2016;39:603–610
23. Foster NC, Miller K, Dimeglio L, et al. Marked Sickle cell trait & other hemoglobinopathies & 52. Bremer JP, Jauch-Chara K, Hallschmid M,
increases in CGM use has not prevented in- diabetes (for providers) [Internet]. Available from Schmid S, Schultes B. Hypoglycemia unaware-
creases in HbA1c levels in participants in the https://www.niddk.nih.gov/health-information/ ness in older compared with middle-aged pa-
T1D Exchange (T1DX) clinic network (Abstract). diagnostic-tests/sickle-cell-trait-hemoglobinopathies- tients with type 2 diabetes. Diabetes Care 2009;
Diabetes 2018;67(Suppl. 1):A451 diabetes. Accessed 12 January 2018 32:1513–1517
care.diabetesjournals.org Battelino and Associates 1603

53. Punthakee Z, Miller ME, Launer LJ, et al.; diabetes: a randomized controlled trial. Diabetes diabetes on the development and progression
ACCORD Group of Investigators; ACCORD-MIND Care 2013;36:1877–1883 of long-term complications in insulin-dependent
Investigators. Poor cognitive function and risk of 66. Paramasivam SS, Chinna K, Singh AKK, et al. diabetes mellitus. N Engl J Med 1993;329:977–
severe hypoglycemia in type 2 diabetes: post hoc Continuous glucose monitoring results in lower 986
epidemiologic analysis of the ACCORD trial. Di- HbA1c in Malaysian women with insulin-treated 80. UK Prospective Diabetes Study (UKPDS)
abetes Care 2012;35:787–793 gestational diabetes: a randomized controlled Group. Intensive blood-glucose control with
54. Giorda CB, Ozzello A, Gentile S, et al.; HYPOS- trial. Diabet Med 2018;35:1118–1129 sulphonylureas or insulin compared with con-
1 Study Group of AMD. Incidence and risk fac- 67. Mazze RS, Lucido D, Langer O, Hartmann K, ventional treatment and risk of complications in
tors for severe and symptomatic hypoglycemia Rodbard D. Ambulatory glucose profile: repre- patients with type 2 diabetes (UKPDS 33). Lancet
in type 1 diabetes. Results of the HYPOS-1 study. sentation of verified self-monitored blood glu- 1998;352:837–853
Acta Diabetol 2015;52:845–853 cose data. Diabetes Care 1987;10:111–117 81. Holman RR, Paul SK, Bethel MA, Matthews
55. Cariou B, Fontaine P, Eschwege E, et al. 68. Fonseca V, Grunberger G. Letter to the DR, Neil HAW. 10-year follow-up of intensive
Frequency and predictors of confirmed hypogly- editor: standard glucose reporting: follow-up glucose control in type 2 diabetes. N Engl J Med
caemia in type 1 and insulin-treated type 2 to the February 2016 AACE CGM Consensus 2008;359:1577–1589
diabetes mellitus patients in a real-life setting: Conference. Endocr Pract 2017;23:629–632 82. Ismail-Beigi F, Craven T, Banerji MA, et al.;
results from the DIALOG study. Diabetes Metab 69. Mullen DM, Bergenstal R, Criego A, Arnold ACCORD trial group. Effect of intensive treat-
2015;41:116–125 KC, Goland R, Richter S. Time savings using a ment of hyperglycaemia on microvascular out-
56. Abdelhafiz AH, Rodrı́guez-Ma~ nas L, Morley standardized glucose reporting system and am- comes in type 2 diabetes: an analysis of the
JE, Sinclair AJ. Hypoglycemia in older people – a bulatory glucose profile. J Diabetes Sci Technol ACCORD randomised trial. Lancet 2010;376:
less well recognized risk factor for frailty. Aging 2018;12:614–621 419–430
Dis 2015;6:156–167 70. Carlson AL, Mullen DM, Bergenstal RM. 83. Hayward RA, Reaven PD, Wiitala WL, et al.;
57. Beck RW, Bergenstal RM, Riddlesworth TD, Clinical use of continuous glucose monitoring VADT Investigators. Follow-up of glycemic con-
Kollman C. The association of biochemical hy- in adults with type 2 diabetes. Diabetes Technol trol and cardiovascular outcomes in type 2 di-
poglycemia with the subsequent risk of a severe Ther 2017;19(Suppl. 2):S4–S11 abetes. N Engl J Med 2015;372:2197–2206
hypoglycemic event: analysis of the DCCT data 71. Hirsch IB, Verderese CA. Professional flash 84. DeWalt DA, Davis TC, Wallace AS, et al. Goal
set. Diabetes Technol Ther 2019;21:1–5 continuous glucose monitoring with ambulatory setting in diabetes self-management: taking the
58. Murphy HR, Rayman G, Duffield K, et al. glucose profile reporting to supplement A1C: baby steps to success. Patient Educ Couns 2009;
Changes in the glycemic profiles of women with rationale and practical implementation. Endocr 77:218–223
type 1 and type 2 diabetes during pregnancy. Pract 2017;23:1333–1344 85. Lawlor KB, Hornyak MJ. SMART goals: how
Diabetes Care 2007;30:2785–2791 72. Kruger DF, Edelman SV, Hinnen DA, Parkin the application of SMART goals can contribute to
59. Feig DS, Donovan LE, Corcoy R, et al.; CG. Reference guide for integrating continuous achievement of student learning outcomes. De-
CONCEPTT Collaborative Group. Continuous glu- glucose monitoring into clinical practice. Diabe- velopments in Business Simulation and Experi-
cose monitoring in pregnant women with type 1 tes Educ 2019;45(Suppl. 1):3S–20S ential Learning: Proceedings of the Annual ABSEL
diabetes (CONCEPTT): a multicentre interna- 73. Rodbard D. Continuous glucose monitoring: Conference 2012;39:259–267
tional randomised controlled trial. Lancet 2017; a review of successes, challenges, and opportu- 86. DiMeglio LA, Acerini C, Codner E, et al. ISPAD
390:2347–2359 nities. Diabetes Technol Ther 2016;18(Suppl. 2): Clinical Practice Consensus Guidelines 2018:
60. Kristensen K, Ögge LE, Sengpiel V, et al. S3–S13 glycemic control targets and glucose monitoring
Continuous glucose monitoring in pregnant 74. Rodbard D. Glucose variability: a review of for children, adolescents, and young adults with
women with type 1 diabetes: an observational clinical applications and research developments. diabetes. Pediatr Diabetes 2018;19:105–114
cohort study of 186 pregnancies. Diabetologia. Diabetes Technol Ther 2018;20(Suppl. 2):S25–S215 87. Aleppo G, Laffel LM, Ahmann AJ, et al. A
23 March 2019 [Epub ahead of print]. DOI: 75. Bergenstal RM, Beck RW, Close KL, et al. practical approach to using trend arrows on the
10.1007/s00125-019-4850-0 Glucose management indicator (GMI): a new term Dexcom G5 CGM system for the management of
61. Stewart ZA, Wilinska ME, Hartnell S, et al. for estimating A1C from continuous glucose adults with diabetes. J Endocr Soc 2017;1:1445–
Closed-loop insulin delivery during pregnancy in monitoring. Diabetes Care 2018;41:2275–2280 1460
women with type 1 diabetes. N Engl J Med 2016; 76. Cohen RM, Franco RS, Smith EP, Higgins JM. 88. Laffel LM, Aleppo G, Buckingham BA, et al. A
375:644–654 When HbA1c and blood glucose do not match: practical approach to using trend arrows on the
62. Stewart ZA, Wilinska ME, Hartnell S, et al. how much is determined by race, by genetics, by Dexcom G5 CGM system to manage children and
Day-and-night closed-loop insulin delivery in a differences in mean red blood cell age? J Clin adolescents with diabetes. J Endocr Soc 2017;1:
broad population of pregnant women with type 1 Endocrinol Metab 2019;104:707–710 1461–1476
diabetes: a randomized controlled crossover 77. Svensson E, Baggesen LM, Johnsen SP, et al. 89. Kudva YC, Ahmann AJ, Bergenstal RM, et al.
trial. Diabetes Care 2018;41:1391–1399 Early glycemic control and magnitude of HbA1c Approach to using trend arrows in the FreeStyle
63. Law GR, Alnaji A, Alrefaii L, et al. Suboptimal reduction predict cardiovascular events and Libre Flash Glucose Monitoring Systems in adults.
nocturnal glucose control is associated with large mortality: population-based cohort study of J Endocr Soc 2018;2:1320–1337
for gestational age in treated gestational diabe- 24,752 metformin initiators. Diabetes Care 90. Monnier L, Colette C, Wojtusciszyn A, et al.
tes mellitus. Diabetes Care 2019;42:810–815 2017;40:800–807 Toward defining the threshold between low and
64. Murphy HR, Rayman G, Lewis K, et al. Ef- 78. Safford MM, Shewchuk R, Qu H, et al. high glucose variability in diabetes. Diabetes
fectiveness of continuous glucose monitoring in Reasons for not intensifying medications: dif- Care 2017;40:832–838
pregnant women with diabetes: randomised ferentiating “clinical inertia” from appropriate 91. Rodbard D. Hypo- and hyperglycemia in
clinical trial. BMJ 2008;337:a1680 care. J Gen Intern Med 2007;22:1648–1655 relation to the mean, standard deviation, co-
65. Secher AL, Ringholm L, Andersen HU, Damm P, 79. Nathan DM, Genuth S, Lachin J, et al.; Di- efficient of variation, and nature of the glucose
Mathiesen ER. The effect of real-time continuous abetes Control and Complications Trial Research distribution. Diabetes Technol Ther 2012;14:
glucose monitoring in pregnant women with Group. The effect of intensive treatment of 868–876