Clinical Targets for ContinuousGlucose Monitoring DataInterpretation: RecommendationsFrom the International Consensuson Time in RangeDiabetes Care 2019;42:1593–1603 | https://doi.org/10.2337/dci19-0028

Improvements in sensor accuracy, greater convenience and ease of use, andexpanding reimbursement have led to growing adoption of continuous glucosemonitoring (CGM). However, successful utilization of CGM technology in routineclinical practice remains relatively low. This may be due in part to the lack of clearand agreed-upon glycemic targets that both diabetes teams and people withdiabetes can work toward. Although unified recommendations for use of key CGMmetrics have been established in three separate peer-reviewed articles, formaladoption by diabetes professional organizations and guidance in the practicalapplication of thesemetrics in clinical practice have been lacking. In February 2019,the Advanced Technologies & Treatments for Diabetes (ATTD) Congress convenedan international panel of physicians, researchers, and individuals with diabeteswho are expert in CGM technologies to address this issue. This article summarizesthe ATTD consensus recommendations for relevant aspects of CGM data utilizationand reporting among the various diabetes populations.

Adoption of continuous glucose monitoring (CGM), which includes both real-timeCGM (rtCGM) and intermittently scanned CGM (isCGM), has grown rapidly overthe past few years as a result of improvements in sensor accuracy, greaterconvenience and ease of use, and expanding reimbursement. Numerous studieshave demonstrated significant clinical benefits of CGM use in people withdiabetes regardless of insulin delivery method (1–15). In many countries, thebenefits and utility of CGM are now recognized by national and internationalmedical organizations for individuals with insulin-requiring diabetes and/or thoseat risk for hypoglycemia (16–21). However, despite increased CGM adoption(22,23), successful utilization of CGM data in routine clinical practice remainsrelatively low. This may be due in part to the lack of clear and agreed-uponglycemic targets toward which both diabetes teams and people with diabetes canwork.In 2012 the Helmsley Charitable Trust sponsored the first expert panel to

recommend the standardization of CGMmetrics and CGM report visualization (24).This was followed by a series of CGM consensus statements refining the core CGMmetrics, but the conclusions were never in alignment. In 2017, several articlessupported use of systematic approaches to CGM data evaluation (18–20). To date,the key CGM metrics remain as unified recommendations in three separate peer-reviewed articles, yet formal adoption by diabetes professional organizations and

1Department of Pediatric Endocrinology, Diabe-tes and Metabolism, University Children’s Hos-pital, University Medical Centre Ljubljana, andFaculty of Medicine, University of Ljubljana,Slovenia2Diabetes Centre for Children and Adolescents,Kinder- und Jugendkrankenhaus Auf der Bult,Hannover, Germany3International Diabetes Center at Park Nicollet,Minneapolis, MN4Diabetes Research Group, King’s College London,London, U.K.5Jaeb Center for Health Research, Tampa, FL6Diabetes Research Institute, IRCCS San RaffaeleHospital, Vita-Salute San Raffaele University,Milan, Italy7Division of Endocrinology and Diabetes, Depart-ment of Pediatrics, Stanford Medical Center,Stanford, CA8American Diabetes Association, Alexandria, VA

Tadej Battelino,1 Thomas Danne,2

Richard M. Bergenstal,3

Stephanie A. Amiel,4 Roy Beck,5

Torben Biester,2 Emanuele Bosi,6

Bruce A. Buckingham,7 William T. Cefalu,8

Kelly L. Close,9 Claudio Cobelli,10

Eyal Dassau,11 J. Hans DeVries,12,13

Kim C. Donaghue,14 Klemen Dovc,1

Francis J. Doyle III,11 Satish Garg,15

George Grunberger,16 Simon Heller,17

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

Desmond Schatz,34 Keaton Stoner,35

Tatsuiko Urakami,36 Stuart A.Weinzimer,37

and Moshe Phillip28,38

This international consensus report hasbeen endorsed by the American DiabetesAssociation, American Association of Clin-ical Endocrinologists, American Associa-tion of Diabetes Educators, EuropeanAssociation for the Study of Diabetes,Foundation of European Nurses in Diabe-tes, International Society for Pediatric andAdolescent Diabetes, JDRF, and PediatricEndocrine Society.

Diabetes Care Volume 42, August 2019 1593

INTER

NATIO

NALCONSEN

SUSREP

ORT

guidance in the practical application ofthese metrics in clinical practice havebeen lacking (19).In February 2019, the Advanced Tech-

nologies & Treatments for Diabetes(ATTD) Congress convened an interna-tional panel of individuals with diabetesand clinicians and researchers with ex-pertise in CGM. Our objective was todevelop clinical CGM targets to supple-ment the currently agreed-upon metricsfor CGM-derived times in glucose ranges(within target range, below target range,above target range) in order to pro-vide guidance for clinicians, researchers,and individuals with diabetes in using,interpreting, and reporting CGM datain routine clinical care and research.Importantly, in order to make the rec-ommendations generalizable and compre-hensive, the consensus panel includedindividuals living with diabetes and hadinternational representation from physi-cians and researchers from all geographicregions.The panel was divided into subgroups

to review literature and provide recom-mendations for relevant aspects of CGMdata utilization and reporting among thevarious diabetes populations. Long-termtrials demonstrating how CGM metricsrelate to and/or predict clinical outcomeshave not been conducted, and many of

the published reports assessed here arenot at the highest evidence level (25).However, there is suggestive evidencefrom a number of recent studies, includinga cross-sectional study correlating currentretrospective 3-day time in target rangewith varying degrees of diabetes retinop-athy (26) and an analysis of the 7-pointself-monitored blood glucose (SMBG)data from the Diabetes Control and Com-plications Trial (DCCT) (27), showing cor-relations of time in target range (70–180 mg/dL [3.9–10.0 mmol/L]) with di-abetes complications. Relationships be-tween time in target range and A1C(26,27) and number of severe and non-severe hypoglycemic events (28–32) havealso been observed. Recommendationsfromeach subgroupwerepresented to thefull panel and voted upon. This articlesummarizes the consensus recommenda-tions and represents the panel members’evaluation of the issues.

NEED FOR METRICS BEYOND A1C

A1C is currently recognized as the keysurrogate marker for the development oflong-termdiabetes complications in peo-ple with type 1 and type 2 diabetes andhas been used as the primary end pointfor many CGM studies (1,3,4,6,33,34).While A1C reflects average glucose overthe last 2–3 months, its limitation is the

lack of information about acute glycemicexcursions and the acute complicationsof hypo- and hyperglycemia. A1C alsofails to identify the magnitude and fre-quency of intra- and interday glucosevariation (35,36). Moreover, certain con-ditions such as anemia (37), hemoglo-binopathies (38), iron deficiency (39),and pregnancy (40) can confound A1Cmeasurements. Importantly, as reportedby Beck et al. (41), the A1C test can fail attimes to accurately reflect mean glucoseeven when none of those conditions arepresent. Despite these limitations, A1C isthe only prospectively evaluated toolfor assessing the risk for diabetes com-plications, and its importance in clinicaldecision making should not be under-valued. Rather, the utility of A1C isfurther enhanced when used as a com-plement to glycemic data measured byCGM.

Unlike A1Cmeasurement, use of CGMallows for the direct observation of gly-cemic excursions and daily profiles,which can inform on immediate therapydecisions and/or lifestyle modifications.CGM also provides the ability to assessglucose variability and identify patternsof hypo- and hyperglycemia. However,potential drawbacks of CGM use includethe need to be actively used in order to beeffective; that it may induce anxiety;

9Close Concerns and The diaTribe Foundation,San Francisco, CA10Department of Information Engineering, Uni-versity of Padova, Padua, Italy11Harvard John A. Paulson School of Engineeringand Applied Sciences, Harvard University, Cam-bridge, MA12Profil, Neuss, Germany13Academic Medical Center, University of Am-sterdam, Amsterdam, the Netherlands14Children’s Hospital at Westmead, University ofSydney, Sydney, Australia15University of Colorado Denver and BarbaraDavis Center for Diabetes, Aurora, CO16Grunberger Diabetes Institute, BloomfieldHills, MI17Academic Unit of Diabetes, Endocrinology andMetabolism, University of Sheffield, Sheffield,U.K.18Science Consulting in Diabetes, Neuss,Germany19Division of Metabolism, Endocrinology andNutrition, Department of Medicine, Universityof Washington School of Medicine, Seattle, WA20Wellcome Trust-MRC Institute of MetabolicScience, and Department of Paediatrics, Univer-sity of Cambridge, Cambridge, U.K.21Department of Endocrinology & Metabolism,Shanghai Clinical Center of Diabetes, Shanghai

Diabetes Institute, Shanghai Key Laboratory ofDiabetes Mellitus, Shanghai Jiao Tong UniversityAffiliated Sixth People’s Hospital, Shanghai, China22Center for Diabetes Technology, University ofVirginia, Charlottesville, VA23JDRF, New York, NY24Pediatric, Adolescent and Young Adult Sectionand Section on Clinical, Behavioral and OutcomesResearch, Joslin Diabetes Center, Harvard Med-ical School, Boston, MA25Endocrinology Research Centre, Moscow,Russia26Clinical and Experimental Endocrinology, KULeuven, Leuven, Belgium27Norwich Medical School, University of EastAnglia, Norwich, U.K.28Jesse Z and Sara Lea Shafer Institute of Endo-crinology and Diabetes, National Center forChildhood Diabetes, Schneider Children’s Med-ical Center of Israel, Petah Tikva, Israel29Steno Diabetes Center Copenhagen, Gentofte,Denmark30CGParkin Communications, Inc., Henderson,NV31Department of Endocrinology, Diabetes, andNutrition,Montpellier University Hospital; Instituteof Functional Genomics, University of Montpellier;and INSERM Clinical Investigation Centre, Mont-pellier, France

32Biomedical Informatics Consultants LLC, Poto-mac, MD33DiaCare, Ahmedabad, Gujarat, India34Pediatric Endocrinology, University of Florida,Gainesville, FL35dQ&AMarket Research, Inc., San Francisco, CA36Department of Pediatrics, Nihon UniversitySchool of Medicine, Tokyo, Japan37Department of Pediatrics, Yale UniversitySchool of Medicine, New Haven, CT38Sackler Faculty ofMedicine, Tel Aviv University,Tel Aviv, Israel

Corresponding author: Tadej Battelino, tadej.battelino@mf.uni-lj.si

This article contains Supplementary Data onlineat http://care.diabetesjournals.org/lookup/suppl/doi:10.2337/dci19-0028/-/DC1.

© 2019 by the American Diabetes Association.Readers may use this article as long as the workis properly cited, the use is educational and notfor profit, and the work is not altered. More infor-mation is available at http://www.diabetesjournals.org/content/license.

1594 International Consensus Report Diabetes Care Volume 42, August 2019

that it may have accuracy limitations,particularly with the delay in registeringblood glucose changes in dynamic sit-uations; and that it can provoke aller-gies. Another limitation of CGM is thatthis technology is not yet widely avail-able in several regions of the world.Effective use of CGM data to optimize

clinical outcomes requires the user tointerpret the collected data and act uponthem appropriately. This requires 1) com-mon metrics for assessment of CGM gly-cemic status, 2) graphical visualization ofthe glucose data and CGM daily profile,and 3) clear clinical targets.

STANDARDIZATION OF CGMMETRICS

In February 2017, the ATTD Congressconvened an international panel of ex-pert clinicians and researchers to definecore metrics for assessing CGM data (18)(Table 1).The list of core CGM metrics has now

been streamlined for use in clinical prac-tice based on the expert opinion of thisinternational consensus group (18). Ofthe 14 core metrics, the panel selectedthat 10 metrics that may be most usefulin clinical practice (Table 2).Fundamental to accurate and mean-

ingful interpretation of CGM is ensuringthat adequate glucose data are availableforevaluation.As shown in studies,.70%

use of CGM over the most recent 14 dayscorrelates strongly with 3 months of meanglucose, time in ranges, and hyperglyce-mia metrics (42,43). In individuals withtype 1 diabetes, correlations are weakerfor hypoglycemia and glycemic variabil-ity; however, these correlations havenot been shown to increase with longersampling periods (43). Longer CGM datacollection periods may be required forindividuals with more variable glycemiccontrol (e.g., 4 weeks of data to in-vestigate hypoglycemia exposure).

TIME IN RANGES

The development of blood glucose test-ing provided individuals with diabetesthe ability to obtain immediate informa-tion about their current glucose levelsand adjust their therapy accordingly.Over the past decades, national and in-ternational medical organizations havebeen successful in developing, harmo-nizing, and disseminating standardizedglycemic targets based on risk for acuteand chronic complications. CGM tech-nology greatly expands the ability toassess glycemic control throughout theday, presenting critical data to informdaily treatment decisions and quantify-ing time below, within, and above theestablished glycemic targets.

Although each of the core metricsestablished in the 2017 ATTD consensusconference (18) provides important in-formation about various aspects of gly-cemic status, it is often impractical toassess and fully utilize many of thesemetrics in real-world clinical practices. To

streamline data interpretation, the con-sensus panel identified “time in ranges”as a metric of glycemic control that pro-vides more actionable information thanA1C alone. The panel agreed that estab-lishing target percentages of time in thevarious glycemic rangeswith the ability toadjust the percentage cut points to ad-dress the specific needs of special di-abetes populations (e.g., pregnancy,high-risk) would facilitate safe and ef-fective therapeutic decision makingwithin the parameters of the establishedglycemic goals.

The metric includes three key CGMmeasurements: percentage of readingsand time per day within target glucoserange (TIR), time below target glucoserange (TBR), and time above target glu-cose range (TAR) (Table 3). The primarygoal for effective and safe glucose controlis to increase the TIR while reducing theTBR. The consensus group agreed thatexpressing time in the various ranges canbe done as the percentage (%) of CGMreadings, average hours and minutesspent in each range per day, or both,depending on the circumstances.

It was agreed that CGM-based glyce-mic targets must be personalized to meetthe needs of each individual with diabe-tes. In addition, the group reached con-sensus on glycemic cutpoints (a targetrange of 70–180 mg/dL [3.9–10.0 mmol/L]for individuals with type 1 diabetes andtype 2 diabetes and 63–140 mg/dL [3.5–7.8 mmol/L] during pregnancy, alongwith a set of targets for the time perday [% of CGM readings or minutes/

Table 1—Standardized CGM metrics

2017 international consensus on CGMmetrics (18)

1. Number of days CGM worn

2. Percentage of time CGM is active

3. Mean glucose

4. Estimated A1C

5. Glycemic variability (%CV or SD)

6. Time .250 mg/dL (.13.9 mmol/L)

7. Time .180 mg/dL (.10.0 mmol/L)

8. Time 70–180 mg/dL (3.9–10.0 mmol/L)

9. Time ,70 mg/dL (,3.9 mmol/L)

10. Time ,54 mg/dL (,3.0 mmol/L)

11. LBGI and HBGI (risk indices)

12. Episodes (hypoglycemia andhyperglycemia) 15 min

13. Area under the curve

14. Time blocks (24-h, day, night)

Use of Ambulatory Glucose Profile (AGP)for CGM report

CV, coefficient of variation; LBGI, low bloodglucose index; HBGI, high blood glucoseindex.

Table 2—Standardized CGM metrics for clinical care: 20191. Number of days CGM worn (recommend 14 days) (42,43)

2. Percentage of time CGM is active (recommend 70% ofdata from 14 days) (41,42)

3. Mean glucose

4. Glucose management indicator (GMI) (75)

5. Glycemic variability (%CV) target #36% (90)*

6. Timeaboverange(TAR):%ofreadingsandtime.250mg/dL(.13.9 mmol/L) Level 2

7.Timeaboverange(TAR):%ofreadingsandtime181–250mg/dL(10.1–13.9 mmol/L) Level 1

8. Time in range (TIR): %of readings and time 70–180mg/dL(3.9–10.0 mmol/L) In range

9.Timebelowrange(TBR):%ofreadingsandtime54–69mg/dL(3.0–3.8 mmol/L) Level 1

10. Timebelowrange (TBR):%of readingsandtime,54mg/dL(,3.0 mmol/L) Level 2

Use of Ambulatory Glucose Profile (AGP) for CGM report

CV, coefficient of variation. *Some studies suggest that lower %CV targets (,33%) provideadditional protection against hypoglycemia for those receiving insulin or sulfonylureas (45,90,91).

care.diabetesjournals.org Battelino and Associates 1595

hours]) individuals with type 1 diabetesand type 2 diabetes (Table 3) and womenduring pregnancy (Table 4) should striveto achieve. It should be noted that pre-meal and postprandial SMBG targetsremain for diabetes in pregnancy (44),in addition to the new CGM TIR targetsfor overall glycemia.Although the metric includes TIR, TBR,

and TAR, achieving the goals for both TBRand TIR would result in reduced timespent above range and thereby improveglycemic control. However, some cliniciansmay choose to target the reduction of thehigh glucose values and minimize hypo-glycemia, thereby arriving at more time inthe target range. In both approaches, thefirst priority is to reduce TBR to targetlevels and thenaddress TIRorTAR targets.Note that for people with type 1 di-

abetes, the targets are informed by theability to reach the targets with hybridclosed-loop therapy (11), the first exam-ple of which is now commercially avail-able with several more systems in finalstages of testing. Importantly, recentstudies have shown the potential of

reaching these targets with CGM in in-dividuals using multiple daily injections(6). In type 2 diabetes, there is generallyless glycemic variability andhypoglycemiathan in type 1 diabetes (45). Thus, peoplewith type 2 diabetes can often achievemore time in the target range whileminimizing hypoglycemia (4). As demon-strated by Beck et al. (4), individuals withtype 2 diabetes increased their TIR by10.3% (from 55.6% to 61.3%) after24 weeks of CGM use with slight reduc-tions in TBR.Most recently, the beneficialeffects of new medications, such as so-dium–glucose cotransporter 2 agentshave helped individuals with type 1 di-abetes increase TIR (46–48). Targets fortype 1 diabetes and type 2 diabetes wereclose enough to combine into one set oftargets, outside of pregnancy.

Another way to visualize the CGM-derived targets for the four categories ofdiabetes is shown in Fig. 1,which displaysand compares the targets for TIR (green),TBR (two categories in light and darkred), and TAR (two categories in yellowand orange). It becomes clear at a glance

that there are different expectations forthe various time in ranges relating tosafety concerns and efficacy based oncurrently available therapies andmedicalpractice.

CLINICAL VALIDITY OF MEASURES

To fundamentally change clinical carewith use of the new metrics, it wouldbe important to demonstrate that themetrics relate to and predict clinicaloutcomes. In this regard, longer-termstudies relating to time spent withinspecific CGM glycemic ranges, diabetescomplications, and other outcomes arerequired. However, there is evidencefrom a number of recent studies thathave shown correlations of TIR (70–180 mg/dL [3.9–10.0 mmol/L]) with di-abetes complications (49,50) as well asa relationship between TIR and A1C(26,27). Although evidence regarding TIRfor older and/or high-risk individuals islacking, numerous studies have shownthe elevated risk for hypoglycemia inthese populations (51–56). Therefore,we have lowered the TIR target from

Table 3—Guidance on targets for assessment of glycemic control for adults with type 1 or type 2 diabetes and older/high-riskindividuals

Diabetes group

TIR TBR TAR

% of readings;time per day Target range

% of readings;time per day Below target level

% of readings;time per day Above target level

Type 1*/type 2 .70%;.16 h, 48 min

70–180 mg/dL(3.9–10.0mmol/L)

,4%;,1 h

,70 mg/dL(,3.9 mmol/L)

,25%;,6 h

.180 mg/dL(.10.0 mmol/L)

,1%;,15 min

,54 mg/dL(,3.0 mmol/L)

,5%;,1 h, 12 min

.250 mg/dL(.13.9 mmol/L)

Older/high-risk#type 1/type 2

.50%;

.12 h70–180 mg/dL(3.9–10 mmol/L)

,1%;,15 min

,70 mg/dL(,3.9 mmol/L)

,10%;,2 h, 24 min

.250 mg/dL(.13.9 mmol/L)

Each incremental 5% increase in TIR is associatedwith clinically significant benefits for individuals with type 1 or type 2 diabetes (26,27). *For age,25years, if the A1C goal is 7.5%, set TIR target to approximately 60%. See the section CLINICAL APPLICATION OF TIME IN RANGES for additional informationregarding target goal setting in pediatric management. #See the section OLDER AND/OR HIGH-RISK INDIVIDUALS WITH DIABETES for additional informationregarding target goal setting.

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

Diabetes group

TIR TBR TAR

% of readings;time per day Target range

% of readings;time per day

Below targetlevel

% of readings;time per day

Above targetlevel

Pregnancy,type 1§

.70%;.16 h, 48 min

63–140 mg/dL†(3.5–7.8mmol/L†)

,4%;,1 h

,63 mg/dL†(,3.5 mmol/L†)

,25%;,6 h

.140 mg/dL(.7.8 mmol/L)

,1%;,15 min

,54 mg/dL(,3.0 mmol/L)

Pregnancy,type 2/GDM§

See PREGNANCY

section63–140 mg/dL†(3.5–7.8mmol/L†)

See PREGNANCY

section,63 mg/dL†

(,3.5 mmol/L†)See PREGNANCY

section.140 mg/dL(.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). †Glucoselevels are physiologically lower during pregnancy. §Percentages of TIR are based on limited evidence. More research is needed.

1596 International Consensus Report Diabetes Care Volume 42, August 2019

.70% to .50% and reduced TBR to,1% at ,70 mg/dL (,3.9 mmol/L) toplace greater emphasis on reducing hy-poglycemia with less emphasis on main-taining target glucose levels (Table 3).

Type 1 Diabetes and Type 2 Diabetes

Association With Complications

Associations between TIR and progres-sion of both diabetic retinopathy (DR)and development of microalbuminuriawere reported by Beck et al. (50), using7-point blood glucose profiles from theDCCT data set to validate the use of TIRas an outcomemeasure for clinical trials.Their analysis showed that the hazardrate for retinopathy progression in-creased by 64% for each 10% reductionin TIR. The hazard rate for microalbumin-uria development increased by 40%for each 10% reduction in TIR. A posthoc analysis of the same DCCT dataset showed a link between glucose of,70 mg/dL (,3.9 mmol/L) and ,54mg/dL (,3.0 mmol/L) and an increasedrisk for severe hypoglycemia (57).Similar associations between DR and

TIRwere reported in a recent study by Luet al. (49) in which 3,262 individuals withtype 2 diabetes were evaluated for DR,which was graded as non-DR, mild non-proliferative DR (NPDR), moderate NPDR,or vision-threatening DR. Results showed

that individuals with more advanced DRspent significantly less time within targetrange (70–180 mg/dL [3.9–10.0 mmol/L])and that prevalence of DR decreased withincreasing TIR.

Relationship Between TIR and A1C

Analyses were conducted utilizing data-sets from four randomized trials encom-passing 545 adults with type 1 diabeteswho had central laboratory measure-ments of A1C (26). TIR (70–180 mg/dL[3.9–10.0 mmol/L]) of 70% and 50%strongly corresponded with an A1C ofapproximately 7% (53 mmol/mol) and 8%(64 mmol/mol), respectively. An increasein TIR of 10% (2.4 h per day) corre-sponded to a decrease in A1C of approx-imately 0.5% (5.0 mmol/mol); similarassociations were seen in an analysisof 18 randomized controlled trials (RCTs)by Vigersky and McMahon (27) that in-cluded over 2,500 individuals with type 1diabetes and type 2 diabetes over a widerange of ages and A1C levels (Table 5).

PregnancyDuring pregnancy, the goal is to safelyincrease TIR as quickly as possible, whilereducing TAR and glycemic variability.Data from the first study of longitudinalCGM use in pregnancy demonstrated a13–percentage point increase in TIR (43%to 56% TIR 70–140mg/dL [3.9–7.8mmol/L])

(58). TBR ,50 mg/dL was reducedfrom 6% to 4%, although the higherTBR ,70 mg/dL was high (13–15%) us-ing older-generation sensors. With im-proved sensor accuracy, recent type 1diabetes pregnancy studies report alower threshold of ,63 mg/dL (,3.5mmol/L) for TBR and $63 mg/dL($3.5 mmol/L) for TIR (59,60). Datafrom Sweden, and the Continuous Glu-cose Monitoring in Women With Type 1Diabetes in Pregnancy Trial (CONCEPTT)control group, report 50% TIR in thefirst trimester, improving to 60% TIRin the third trimester, reflecting contem-porary antenatal care. Of note, thesedata confirm that the TBR ,63 mg/dL(,3.5 mmol/L) recommendation of ,4%is safely achievable, especially after thefirst trimester. Furthermore, 33% ofwomen achieved the recommendation of70% TIR 63–140mg/dL (3.5–7.8mmol/L)in the final (.34) weeks of pregnancy.Preliminary data suggest that closed-loop systemsmay allow pregnant womento safely achieve 70% TIR at an earlier(.24 weeks) stage of gestation (61,62).Law et al. (63) analyzed data from twoearly CGM trials (64,65) describing theassociations between CGM measuresand risk of large-for-gestational-age (LGA)infants. Taken together, the Swedish andCONCEPTTdata confirmthata5–7%higher

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

care.diabetesjournals.org Battelino and Associates 1597

TIR during the second and third trimestersis associated with decreased risk of LGAand neonatal outcomes includingmacro-somia, shoulder dystocia, neonatal hy-poglycemia, and neonatal intensive careadmissions. More data are needed todefine the clinical CGM targets for preg-nant women with type 2 diabetes, whospend one-third less time hyperglycemicthan women with type 1 diabetes andachieve TIR of 90% (58). Because of thelack of evidence on CGM targets forwomen with gestational diabetes mellitus(GDM) or type 2 diabetes in pregnancy,percentages of time spent in range, belowrange, and above range have not beenincluded in this report. Recent data sug-gest that even more stringent targets(66) and greater attention to overnightglucose profiles may be required tonormalize outcomes in pregnant womenwith GDM (63).

Older and/or High-Risk IndividualsWith DiabetesOlder and/or high-risk individuals withdiabetes are at notably higher risk forsevere hypoglycemia due to age, dura-tion of diabetes, duration of insulintherapy, and greater prevalence of hy-poglycemia unawareness (51–55). Theincreased risk of severe hypoglycemiais compounded by cognitive and physicalimpairments and other comorbidities(53,56). High-risk individuals includethose with a higher risk of complica-tions, comorbid conditions (e.g., cogni-tive deficits, renal disease, joint disease,osteoporosis, fracture, and/or cardio-vascular disease), and those requiring

assisted care, which can complicatetreatment regimens (56). Therefore,when setting glycemic targets for high-risk and/or elderly people, it is importantto individualize and be conservative,with a strong focus on reducing the per-centage of time spent ,70 mg/dL (,3.9mmol/L) and preventing excessive hy-perglycemia.

STANDARDIZATION OF CGM DATAPRESENTATION

As noted above, in 2013 a panel ofclinicians with expertise in CGM pub-lished recommendations for use of theAmbulatory Glucose Profile (AGP) as atemplate for data presentation and vi-sualization. Originally created by Mazzeet al. (67), the standardized AGP reportwas further developed by the Interna-tional Diabetes Center and now incor-porates all the core CGM metrics andtargets along with a 14-day compositeglucose profile as an integral componentof clinical decision making (24). Thisrecommendation was later endorsedat the aforementioned internationalconsensus conference on CGM metrics(18) and is referenced as an examplein the American Diabetes Association2019 “Standards of Medical Care in Di-abetes” (16) and in an update to theAmerican Association of Clinical Endo-crinologists consensus on use of CGM(68). TheAGP report, in slightlymodifiedformats, has been adopted by most ofthe CGM device manufacturers in theirdownload software. An example of theAGP report, updated to incorporate tar-gets, is presented in Fig. 2. In the AGP

report, glucose ranges are defined as“Very High” (Level 2), “High” (Level 1),“Low” (Level 1), and “Very Low” (Level 2).An “mmol/L” version is provided inSupplementary Fig. 1.

There is a general consensus that auseful CGM report is one that can beunderstood by clinicians and people withdiabetes. While there may be some terms(e.g., glucose variability) that are lessfamiliar to many people with diabetes, asingle-page report that the medical teamcan reviewandfile in theelectronicmedicalrecord and that can be used as a shareddecision-making tool with people with di-abetes was considered to be of value(69–72). More detailed reports (e.g., ad-justabledata ranges,detaileddaily reports)should remain available for individualizedreview by or with people with diabetes.

Clinical Application of Time in RangesDespite its demonstrated value, clinicalutilization of CGM data has remainedsuboptimal. Although time constraintsand reimbursement issues are clearlyobstacles, clinician inexperience in datainterpretation and lack of standardiza-tion software for visualization of CGMdata have also played a role (73). Theproposed standardized report enablesclinicians to readily identify importantmetrics such as the percentage of timespent within, below, and above eachindividual’s target range, allowing forgreater personalization of therapy throughshared decision making.

Using the standardized report, theclinician can also address glucose vari-ability (e.g., the coefficient of variation

Table 5—Estimate of A1C for a given TIR level based on type 1 diabetes and type 2 diabetes studies

Beck et al. (26) (n = 545 participants with type 1 diabetes)Vigersky and McMahon (27) (n = 1,137

participants with type 1 or type 2 diabetes)

TIR 70–180 mg/dL(3.9–10.0 mmol/L)

A1C, %(mmol/mol)

95% CI for predictedA1C values, %

TIR 70–180 mg/dL(3.9–10.0 mmol/L)

A1C, %(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 (RCTswith subjects with type 1 diabetes vs. RCTs with subjects with type 1 or type 2 diabetes with CGM and SMBG).

1598 International Consensus Report Diabetes Care Volume 42, August 2019

Figure 2—Ambulatory Glucose Profile.

care.diabetesjournals.org Battelino and Associates 1599

[%CV] metric) (74) or use the glucosemanagement indicator (GMI) metric (75)to discuss the possible discrepanciesnoted in glucose exposure derived fromCGMdataversus the individual’s laboratory-measured A1C (41,76). With appropriateeducational materials, time, and experi-ence, clinicians will develop a systematicapproach to CGM data analysis and themost effective ways to discuss the datawith patients in person or remotely.

Goal Setting

Numerous studies have demonstratedthe clinical benefits of early achievementof near-normal glycemic control in indi-viduals with type 1 diabetes and type 2diabetes (77–83). However, when advis-ing people with diabetes, goal-settingmust be collaborative and take into ac-count the individual needs/capabilities ofeach patient and start with the goals thatare most achievable. An early study byDeWalt et al. (84) found that setting small,achievable goals not only enhances peo-ple’s ability to copewith their diabetes, butthat people with diabetes who set andachieved their goals often initiated addi-tional behavioral changes on their own.One approach to consider is the SMARTgoal (Specific, Measurable, Achievable,Relevant,Time-bound) intervention,whichis directly applicable to setting targets fortime in ranges. First described by Lawlorand Hornyak in 2012 (85), this approachincorporates four key components ofbehavioral change relevant to goal set-ting: 1) the goal is specific and definesexactly what is to be achieved, 2) the goalis measurable and there is tangible evi-dence when it has been achieved, 3) thegoal is achievable but stretches the pa-tient slightly so that he/she feels chal-lenged, and 4) the goal should beattainable over a short period of time.Effective goals should utilize CGMdata

to identify specific instances for thepatient to take measurable action toprevent hypoglycemia. Although analysisof the AGP reports provides an oppor-tunity for meaningful discussion, individ-uals should be counseled to look atpatterns throughout the day to seewhen low glucose events are occurringand make adjustments in their therapyto reduce these events.When applying the CGM metrics in

clinical practice, it may be more mean-ingful and motivating to communicateto people with diabetes the importance

of working to reduce the time spent,70 mg/dL (,3.9 mmol/L) to lessthan 1 h per day and time spent,54 mg/dL (,3.0 mmol/L) to lessthan 15 min per day, rather than us-ing ,4% and ,1%, respectively, as thegoal.However,asdiscussedearlier, targetsmust be personalized to meet the needsand capabilities of each person, focusingon small steps and small successes. Indi-viduals with diabetes should work withtheir provider and/or educator to developa SMART goal to reduce TBR.

Individualized goals are particularlyimportant for pediatric and young adultpopulations. The International Societyfor Pediatric and Adolescent Diabetesrecommends that targets for individuals#25 years of age aim for the lowestachievable A1C without undue exposureto severe hypoglycemia or negative ef-fects on quality of life and burden of care(86). An A1C target of 7.0% (53 mmol/mol) canbeused in children, adolescents,and adults #25 years old who haveaccess to comprehensive care (86).However, a higher A1C goal (e.g., ,7.5%[,58 mmol/mol]) may be more appro-priate in the following situations: inabilityto articulate hypoglycemia symptoms,hypoglycemia unawareness, history ofsevere hypoglycemia, lack of access toanalog insulins and/or advanced insulindelivery technology, or inability to regu-larly check glucose (86). This wouldequate to a TIR target of;60% (Table 4).

The consensus group recognized thatachieving the targets for the various timein ranges is aspirational in some situa-tions, and many individuals will requireongoing support, both educational andtechnological, from their health careteam. Importantly, as demonstrated byBeck et al. (26), Vigersky and McMahon(27), and Feig et al. (59), even small,incremental improvements yield signifi-cant glycemic benefits. Therefore, whenadvising individuals with diabetes (par-ticularly children, adolescents, and high-risk individuals) about their glycemic goals,it is important to take a stepwise ap-proach, emphasizing that what may ap-pear to be small, incremental successes(e.g., 5% increase in TIR) are, in fact,clinically significant in improving theirglycemia (26,27,59). However, when coun-seling women planning pregnancy andpregnant women, greater emphasis shouldbe placed on getting to goal as soon aspossible (59,60).

CONCLUSIONS

Use of CGM continues to expand inclinical practice. As a component of di-abetes self-management, daily use ofCGM provides the ability to obtain im-mediate feedback on current glucoselevels as well as direction and rate ofchange in glucose levels. This informationallows people with diabetes to optimizedietary intake and exercise, make in-formed therapy decisions regardingmealtime and correction of insulin dos-ing, and, importantly, react immediatelyand appropriately to mitigate or preventacute glycemic events (87–89). Retro-spective analysis of CGM data, usingstandardized data management toolssuch as the AGP, enables cliniciansand people with diabetes to work col-laboratively in identifying problem areasand then set achievable goals (70–72).We conclude that, in clinical practice,time in ranges (within target range, be-low range, above range) are both appro-priate and useful as clinical targets andoutcome measurements that comple-ment A1C for a wide range of peoplewith diabetes and that the target valuesspecified in this article should be con-sidered an integral component of CGMdata analysis and day-to-day treatmentdecision making.

Acknowledgments. The consensus group par-ticipants wish to thank the ATTD Congress fororganizing and coordinating the meeting andRachel Naveh (The Jesse Z and Sara Lea ShaferInstitute for Endocrinology and Diabetes, Na-tional Center for Childhood Diabetes, SchneiderChildren’sMedical Centerof Israel) for assistancein organizing the meeting. They also thankCourtney Lias from the U.S. Food and DrugAdministration for her participation as an ob-server at the consensus conference.Funding and Duality of Interest. Support forthe CGM consensus conference and develop-ment of this consensus report was provided bythe ATTD Congress. Abbott Diabetes Care, AstraZeneca, Dexcom Inc., Eli Lilly and Company,Insulet Corporation, Medtronic, Novo Nordisk,Roche Diabetes Care, and Sanofi provided fund-ing to ATTD to support the consensus meeting.Consensus participants were reimbursed fortravel to the ATTD conference and one nightof lodging; no honoraria were provided. ATTDprovided funding to Christopher G. Parkin,CGParkin Communications, Inc., for his medicalwriting and editorial support. T.Ba. has receivedhonoraria for participation on advisory boardsfor Novo Nordisk, Sanofi, Eli Lilly and Company,Boehringer,Medtronic, andBayerHealthCareandas a speaker for AstraZeneca, Eli Lilly and Com-pany, Bayer, Novo Nordisk, Medtronic, Sanofi,and Roche. T.Ba. owns stocks of DreaMed

1600 International Consensus Report Diabetes Care Volume 42, August 2019

Diabetes, andhis institutionhas receivedresearchgrant support and travel expenses from AbbottDiabetes Care, Medtronic, Novo Nordisk, Glu-Sense, Sanofi, Sandoz, and Diamyd. T.D. hasreceived speaker honoraria, research support,and consulting fees from Abbott Diabetes Care,Bayer, Bristol-Myers Squibb, AstraZeneca, Boeh-ringer Ingelheim, Dexcom, Eli Lilly and Company,Medtronic, Novo Nordisk, Sanofi, and RocheDiabetes Care; and he is a shareholder ofDreaMed Diabetes. S.A.A. has received honorariafor participation on advisory boards for Rocheand Medtronic and has given a lecture funded bySanofi. R.B. is an employee the Jaeb Center forHealth Research, which has received grant sup-port from Dexcom, Animas, Bigfoot, and Tan-dem; nonfinancial study support from Dexcom,Abbott Diabetes Care, and Roche Diabetes Care;and consulting fees from Eli Lilly and Companyand Insulet. He has no personal financial arrange-ments with any company. R.M.B. has receivedresearch funding and served as a consultant andon advisory boards for Abbott Diabetes Care,BectonDickinson,Dexcom,Eli Lilly andCompany,Glooko, Helmsley Charitable Trust, Hygieia,Johnson&Johnson,Medtronic,Merck,NovoNordisk,Roche, Sanofi, and Senseonics. His employer,nonprofit HealthPartners Institute, contractsfor his services and no personal income goes toR.M.B. E.B. received honoraria for participationon advisory boards and speakers’ bureaus fromAbbott Diabetes Care, AstraZeneca, Medtronic,Novartis, Roche, and Sanofi. B.A.B. is on medicaladvisory boards forMedtronic and Convatec andhas received research funding from the NationalInstitutesofHealth, JDRF, theLeonaM.andHarryB. Helmsley Charitable Trust, Medtronic Diabe-tes, ConvaTec, Dexcom, Tandem, and Insulet.K.L.C. is an employee of Close Concerns and ThediaTribe Foundation, which receive funding fromCGM manufacturers, including Medtronic, Dex-com, and Abbott Diabetes Care. C.C. reports10 patents and patent applications related tocontinuous glucose sensors and artificial pan-creas. E.D. has received consulting fees andhonoraria for participation on advisory boardsfrom Roche, Insulet, and Eli Lilly and Companyand research support from Dexcom, Insulet, Ani-mas,Xeris, andRoche. J.H.D. has received speakerhonoraria and research support from and hasconsulted for Abbott Diabetes Care, Dexcom,Medtronic, Merck Sharp & Dohme, Novo Nor-disk, Sanofi, Roche, Senseonics, and Zealand.F.J.D. has received consulting fees from Mod-eAGC and research support from Dexcom, In-sulet, Animas, and Xeris. S.G. has receivedconsulting fees and honoraria for participationon advisory boards for Medtronic, Roche Di-abetes Care, Merck, Lexicon, Novo Nordisk,Sanofi, MannKind, Senseonics, Zealand, and EliLilly and Company and research grants from EliLilly and Company, Novo Nordisk, Merck, Lexi-con, Medtronic, Dario, the National Cancer In-stitute, T1D Exchange, the National Institute ofDiabetes and Digestive and Kidney Diseases,JDRF, Animas, Dexcom, and Sanofi. G.G. hasreceived research support from Novo Nordiskand Medtronic and honoraria for participationon speakers’ bureaus from Novo Nordisk, Eli Lillyand Company, Boehringer Ingelheim, and Sanofi.S.H. has served as a consultant or speaker forEli Lilly and Company, Novo Nordisk, Takeda,

Boehringer Ingelheim, MannKind, Sanofi, Zea-land Pharma, and UNEEG. L.H. is a consultantfor companies developing novel diagnostic andtherapeutic options for diabetes. He is a share-holder of Profil Institut fur StoffwechselforschungGmbH and ProSciento. I.B.H. receives researchfunding from Medtronic Diabetes and has re-ceived consulting fees from Abbott DiabetesCare, Bigfoot, Roche, and Becton Dickinson.R.H. reports having received speaker honorariafrom Eli Lilly and Company, Novo Nordisk, andAstraZeneca, serving on an advisory panel for EliLilly and Company and Novo Nordisk, and re-ceiving license fees from B. Braun and Medtronic.O.K. received honoraria fromAmring, Eli Lilly andCompany, Novo Nordisk, and Sanofi and ownsshares from DreaMed Diabetes. B.K. declaresresearch support handled by the University ofVirginia from Dexcom, Roche, Sanofi, and Tan-dem; patent royalties handled by the Universityof Virginia from Johnson & Johnson, Sanofi, andDexcom; and has served as a consultant for Sanofiand Tandem and on a speakers’ bureau forDexcom. B.L. is an employee of Close Concernsand The diaTribe Foundation, which receivefunding from CGM manufacturers, includingMedtronic, Dexcom, and Abbott Diabetes Care.C.M. serves or has served on the advisory panelor speakers’ bureau for Novo Nordisk, Sanofi,Merck Sharp & Dohme, Eli Lilly and Company,Novartis, AstraZeneca, Boehringer Ingelheim,Hanmi Pharmaceuticals, Roche, Medtronic,ActoBio Therapeutics, Pfizer, Dianax, and UCB.Financial compensation for these activities hasbeen received by KU Leuven. H.R.M. receivedhonoraria from participation on advisory boardsfor Medtronic and research support from Dex-com, Medtronic, Abbott Diabetes Care, andJohnson & Johnson. R.N. received honorariafor participation on the speakers’ bureau ofNovo Nordisk, Pfizer, Eli Lilly and Company,and Sanofi. R.N. owns DreaMed stock and reportstwo patent applications. K.N. owns shares inNovo Nordisk and has received consulting feesfromMedtronic,AbbottDiabetesCare, andNovoNordisk; speaker honoraria from Medtronic,Roche Diabetes Care, Rubin Medical, Sanofi,Novo Nordisk, Zealand Pharma, and Bayer;and research support from Novo Nordisk, Zea-land Pharma, Medtronic, and Roche DiabetesCare. C.G.P. has received consulting fees fromDexcom, Diasome, Onduo, Proteus, Roche Di-abetes Care, and Senseonics. E.R. has receivedconsulting fees from A. Menarini Diagnostics,Abbott Diabetes Care, Air Liquide SI, BectonDickinson, Cellnovo, Dexcom, Eli Lilly and Com-pany, Insulet, Johnson & Johnson, Medtronic,Novo Nordisk, Roche, and Sanofi and researchsupport from Abbott Diabetes Care, Dexcom,Insulet, Tandem, and Roche. D.R. has receivedconsulting fees from Eli Lilly and Company andBetter Therapeutics. K.S. is an employee of dQ&AMarket Research, Inc., whose clients includeseveral device and pharmaceutical companiesin the diabetes field. S.A.W. has received con-sulting fees from Eli Lillyand Company, Sanofi,and Zealand and speaker honoraria from Med-tronic, Insulet, and Tandem.M.P. is a member ofthe advisory board of AstraZeneca, Sanofi, Med-tronic, Eli Lilly and Company, Novo Nordisk, andInsulet and is a consultant to RSP Systems A/S,Qulab Medical, and Pfizer. The institute headed

by M.P. received research support from Med-tronic, Novo Nordisk, Eli Lilly and Company,Dexcom, Sanofi, Insulet, OPKO Health, DreaMedDiabetes, Bristol-Myers Squibb, andMerck. M.P.is a stockholder/shareholder of DreaMed Dia-betes, NG Solutions, and Nutriteen Professionalsand reports two patent applications. No otherpotential conflicts of interest relevant to thisarticle were reported.

References1. Lind M, Polonsky W, Hirsch IB, et al. Contin-uous glucose monitoring vs conventional therapyfor glycemic control in adults with type 1 diabetestreated with multiple daily insulin injections: theGOLD randomized clinical trial. JAMA 2017;317:379–3872. Aleppo G, Ruedy KJ, Riddlesworth TD, et al.;REPLACE-BG Study Group. REPLACE-BG: A ran-domized trial comparing continuous glucosemonitoring with and without routine bloodglucose monitoring in adults with well-controlledtype 1 diabetes. Diabetes Care 2017;40:538–5453. Beck RW, Riddlesworth T, Ruedy K, et al.;DIAMOND Study Group. Effect of continuousglucose monitoring on glycemic control in adultswith type 1 diabetes using insulin injections: TheDIAMOND randomized clinical trial. JAMA 2017;317:371–3784. Beck RW, Riddlesworth TD, Ruedy K, et al.;DIAMOND Study Group. Continuous glucosemonitoring versus usual care in patients withtype 2 diabetes receiving multiple daily insulininjections: a randomized trial. Ann Intern Med2017;167:365–3745. Polonsky WH, Hessler D, Ruedy KJ, Beck RW;DIAMOND Study Group. The impact of contin-uous glucose monitoring on markers of qualityof life in adults with type 1 diabetes: furtherfindings from the DIAMOND randomized clinicaltrial. Diabetes Care 2017;40:736–7416. Soupal J, Petruzelkova L, Flekac M, et al.Comparison of different treatment modalitiesfor type 1 diabetes, including sensor-augmentedinsulin regimens, in 52 weeks of follow-up:a COMISAIR study. Diabetes Technol Ther2016;18:532–5387. van Beers CA, DeVries JH, Kleijer SJ, et al.Continuous glucosemonitoring for patients withtype 1 diabetes and impaired awareness ofhypoglycaemia (IN CONTROL): a randomised,open-label, crossover trial. Lancet Diabetes En-docrinol 2016;4:893–9028. Bolinder J, Antuna R, Geelhoed-Duijvestijn P,Kroger J, Weitgasser R. Novel glucose-sensingtechnology and hypoglycaemia in type 1 diabe-tes: a multicentre, non-masked, randomisedcontrolled trial. Lancet 2016;388:2254–22639. Haak T, Hanaire H, Ajjan R, Hermanns N,Riveline JP, Rayman G. Flash glucose-sensingtechnology as a replacement for blood glucosemonitoring for the management of insulin-treatedtype 2 diabetes: a multicenter, open-label ran-domized controlled trial. Diabetes Ther 2017;8:55–7310. Choudhary P, Olsen BS, Conget I, Welsh JB,Vorrink L, Shin JJ. Hypoglycemia prevention anduser acceptance of an insulin pump system withpredictive low glucose management. DiabetesTechnol Ther 2016;18:288–29111. Bergenstal RM, Garg S, Weinzimer SA, et al.Safety of a hybrid closed-loop insulin delivery

care.diabetesjournals.org Battelino and Associates 1601

system in patients with type 1 diabetes. JAMA2016;316:1407–140812. Heinemann L, Freckmann G, Erdmann D,et al. Real-time continuous glucose monitoringuse in adults with type 1 diabetes and impairedhypoglycaemia awareness or severe hypoglycae-mia treated with multiple daily insulin injections(HypoDE): a multicentre, randomised controlledtrial. Lancet 2018;391:1367–137713. Reddy M, Jugnee N, El Laboudi A,Spanudakis E, Anantharaja S, Oliver N. A ran-domised controlled pilot study of continuousglucosemonitoring and flash glucosemonitoringin people with type 1 diabetes and impairedawareness of hypoglycaemia. Diabet Med 2018;35:483–49014. Battelino T, Nimri R, Dovc K, Phillip M,Bratina N. Prevention of hypoglycemia withpredictive low glucose insulin suspension inchildren with type 1 diabetes: a randomizedcontrolled trial. Diabetes Care 2017;40:764–77015. Dovc K, Cargnelutti K, SturmA, Selb J, BratinaN, Battelino T. Continuous glucose monitoringuse and glucose variability in pre-school childrenwith type 1 diabetes. Diabetes Res Clin Pract2019;147:76–8016. American Diabetes Association. 7. Diabe-tes technology: Standards of Medical Care inDiabetesd2019. Diabetes Care 2019;42(Suppl.1):S71–S8017. Fonseca VA, Grunberger G, Anhalt H, et al.;Consensus ConferenceWriting Committee. Con-tinuous glucose monitoring: a consensus con-ference of the American Association of ClinicalEndocrinologists and American College of Endo-crinology. Endocr Pract 2016;22:1008–102118. Danne T, Nimri R, Battelino T, et al. In-ternational consensus on use of continuousglucose monitoring. Diabetes Care 2017;40:1631–164019. Petrie JR, Peters AL, Bergenstal RM,Holl RW,Fleming GA, Heinemann L. Improving the clinicalvalue and utility of CGM Systems: issues andrecommendations: a joint statement of theEuropean Association for the Study of Diabetesand the American Diabetes Association Diabe-tes Technology Working Group. Diabetes Care2017;40:1614–162120. Agiostratidou G, Anhalt H, Ball D, et al.Standardizing clinically meaningful outcomemeasures beyond HbA1c for type 1 diabetes:a consensus report of the American Associationof Clinical Endocrinologists, the American Asso-ciation of Diabetes Educators, the AmericanDiabetes Association, the Endocrine Society,JDRF International, The Leona M. and Harry B.Helmsley Charitable Trust, the Pediatric Endo-crine Society, and the T1D Exchange. DiabetesCare 2017;40:1622–163021. Sherr JL, Tauschmann M, Battelino T, et al.ISPAD Clinical Practice Consensus Guidelines2018: diabetes technologies. Pediatr Diabetes2018;19(Suppl. 27):302–32522. FosterNC, Beck RW,Miller KM, et al. State oftype 1 diabetes management and outcomes fromthe T1D Exchange in 2016–2018. Diabetes TechnolTher 2019;21:66–7223. Foster NC,Miller K, Dimeglio L, et al.Markedincreases in CGM use has not prevented in-creases in HbA1c levels in participants in theT1D Exchange (T1DX) clinic network (Abstract).Diabetes 2018;67(Suppl. 1):A451

24. Bergenstal RM, Ahmann AJ, Bailey T, et al.Recommendations for standardizing glucose re-porting and analysis to optimize clinical decisionmaking in diabetes: the Ambulatory GlucoseProfile (AGP). Diabetes Technol Ther 2013;15:198–21125. American Diabetes Association. Introduc-tion: Standards of Medical Care in Diabetesd2017. Diabetes Care 2017;41(Suppl. 1):S1–S226. Beck RW, Bergenstal RM, Cheng P, et al. Therelationships between time in range, hypergly-cemia metrics, and HbA1c. J Diabetes Sci Technol.13 January 2019 [Epub ahead of print]. DOI:10.1177/193229681882249627. VigerskyRA,McMahonC.The relationshipofhemoglobin A1C to time-in-range in patientswith diabetes. Diabetes Technol Ther 2019;21:81–8528. Brod M, Christensen T, Thomsen TL,Bushnell DM. The impact of non-severe hy-poglycemic events on work productivity anddiabetes management. Value Health 2011;14:665–67129. Brod M, Rana A, Barnett AH. Impact of self-treated hypoglycaemia in type 2 diabetes: amul-tinational survey in patients and physicians. CurrMed Res Opin 2012;28:1947–195830. Seaquist ER, Anderson J, Childs B, et al.Hypoglycemia and diabetes: a report of a work-group of the American Diabetes Association andthe Endocrine Society. Diabetes Care 2013;36:1384–139531. International Hypoglycaemia Study Group.Glucose concentrations of less than 3.0 mmol/L(54 mg/dL) should be reported in clinical trials:a joint position statement of the American Di-abetesAssociation and the EuropeanAssociationfor the Study of Diabetes. Diabetes Care 2017;40:155–15732. Novodvorsky P, Bernjak A, Chow E, et al.Diurnal differences in risk of cardiac arrhythmiasduring spontaneous hypoglycemia in youngpeople with type 1 diabetes. Diabetes Care2017;40:655–66233. Battelino T, Conget I, Olsen B, et al.; SWITCHStudy Group. The use and efficacy of continu-ous glucose monitoring in type 1 diabetestreated with insulin pump therapy: a rando-mised controlled trial. Diabetologia 2012;55:3155–316234. Bergenstal RM, TamborlaneWV, Ahmann A,et al.; STAR 3 Study Group. Sensor-augmentedpump therapy for A1C reduction (STAR 3) study:results from the 6-month continuation phase.Diabetes Care 2011;34:2403–240535. Cox DJ, Kovatchev BP, Julian DM, et al.Frequency of severe hypoglycemia in insulin-dependent diabetes mellitus can be predictedfrom self-monitoring blood glucose data. J ClinEndocrinol Metab 1994;79:1659–166236. Qu Y, Jacober SJ, Zhang Q,Wolka LL, DeVriesJH. Rate of hypoglycemia in insulin-treated pa-tientswith type 2 diabetes can be predicted fromglycemic variability data. Diabetes Technol Ther2012;14:1008–101237. National Institute of Diabetes and Digestiveand Kidney Diseases Health Information Center.Sickle cell trait & other hemoglobinopathies &diabetes (for providers) [Internet]. Available fromhttps://www.niddk.nih.gov/health-information/diagnostic-tests/sickle-cell-trait-hemoglobinopathies-diabetes. Accessed 12 January 2018

38. Bry L, Chen PC, Sacks DB. Effects of hemo-globin variants and chemically modified deriva-tives on assays for glycohemoglobin. Clin Chem2001;47:153–16339. Ford ES, Cowie CC, Li C, Handelsman Y,Bloomgarden ZT. Iron-deficiency anemia, non-iron-deficiency anemia and HbA1c among adultsin the US. J Diabetes 2011;3:67–7340. Nielsen LR, Ekbom P, Damm P, et al. HbA1clevels are significantly lower in early and latepregnancy. Diabetes Care 2004;27:1200–120141. Beck RW, Connor CG, Mullen DM, WesleyDM, Bergenstal RM. The fallacy of average: howusing HbA1c alone to assess glycemic control canbe misleading. Diabetes Care 2017;40:994–99942. Xing D, Kollman C, Beck RW, et al.; JuvenileDiabetes Research Foundation Continuous Glu-cose Monitoring Study Group. Optimal samplingintervals to assess long-term glycemic controlusing continuous glucose monitoring. DiabetesTechnol Ther 2011;13:351–35843. Riddlesworth TD, Beck RW, Gal RL, et al.Optimal sampling duration for continuousglucose monitoring to determine long-term gly-cemic control. Diabetes Technol Ther 2018;20:314–31644. AmericanDiabetesAssociation. 14.Manage-ment of diabetes in pregnancy. Standards ofMedical Care in Diabetesd2019. Diabetes Care2019;42(Suppl. 1):S165–S17245. Rama Chandran S, Tay WL, Lye WK, et al.Beyond HbA1c: comparing glycemic variabilityand glycemic indices in predicting hypoglycemiain type 1 and type 2 diabetes. Diabetes TechnolTher 2018;20:353–36246. Famulla S, Pieber TR, Eilbracht J, et al.Glucose exposure and variability with empagli-flozin as adjunct to insulin in patients with type 1diabetes: continuous glucose monitoring datafrom a 4-week, randomized, placebo-controlledtrial (EASE-1). Diabetes Technol Ther 2017;19:49–6047. Dandona P, Mathieu C, Phillip M, et al.;DEPICT-1 Investigators. Efficacy and safety ofdapagliflozin in patients with inadequately con-trolled type 1 diabetes: the DEPICT-1 52-weekstudy. Diabetes Care 2018;41:2552–255948. Mathieu C, Dandona P, Phillip M, et al.;DEPICT-1 and DEPICT-2 Investigators. Glucosevariables in type 1 diabetes studies with dapagli-flozin: pooled analysis of continuous glucosemonitoring data from DEPICT-1 and -2. DiabetesCare 2019;42:1081–108749. Lu J, Ma X, Zhou J, et al. Association of timein range, as assessed by continuous glucosemonitoring, with diabetic retinopathy in type2 diabetes. Diabetes Care 2018;41:2370–237650. Beck RW, Bergenstal RM, Riddlesworth TD,et al. Validation of time in range as an outcomemeasure for diabetes clinical trials. DiabetesCare 2019;42:400–40551. Weinstock RS, DuBose SN, Bergenstal RM,et al.; T1D Exchange Severe Hypoglycemia inOlder Adults With Type 1 Diabetes Study Group.Risk factors associated with severe hypoglycemiain older adults with type 1 diabetes. DiabetesCare 2016;39:603–61052. Bremer JP, Jauch-Chara K, Hallschmid M,Schmid S, Schultes B. Hypoglycemia unaware-ness in older compared with middle-aged pa-tients with type 2 diabetes. Diabetes Care 2009;32:1513–1517

1602 International Consensus Report Diabetes Care Volume 42, August 2019

53. Punthakee Z, Miller ME, Launer LJ, et al.;ACCORD Group of Investigators; ACCORD-MINDInvestigators. Poor cognitive function and risk ofsevere hypoglycemia in type 2 diabetes: post hocepidemiologic analysis of the ACCORD trial. Di-abetes Care 2012;35:787–79354. GiordaCB,OzzelloA, Gentile S, et al.; HYPOS-1 Study Group of AMD. Incidence and risk fac-tors for severe and symptomatic hypoglycemiain type 1 diabetes. Results of the HYPOS-1 study.Acta Diabetol 2015;52:845–85355. Cariou B, Fontaine P, Eschwege E, et al.Frequency and predictors of confirmed hypogly-caemia in type 1 and insulin-treated type 2diabetes mellitus patients in a real-life setting:results from the DIALOG study. Diabetes Metab2015;41:116–12556. Abdelhafiz AH, Rodrıguez-Ma~nas L, MorleyJE, Sinclair AJ. Hypoglycemia in older people – aless well recognized risk factor for frailty. AgingDis 2015;6:156–16757. Beck RW, Bergenstal RM, Riddlesworth TD,Kollman C. The association of biochemical hy-poglycemia with the subsequent risk of a severehypoglycemic event: analysis of the DCCT dataset. Diabetes Technol Ther 2019;21:1–558. Murphy HR, Rayman G, Duffield K, et al.Changes in the glycemic profiles of women withtype 1 and type 2 diabetes during pregnancy.Diabetes Care 2007;30:2785–279159. Feig DS, Donovan LE, Corcoy R, et al.;CONCEPTT Collaborative Group. Continuous glu-cose monitoring in pregnant women with type 1diabetes (CONCEPTT): a multicentre interna-tional randomised controlled trial. Lancet 2017;390:2347–235960. Kristensen K, Ogge LE, Sengpiel V, et al.Continuous glucose monitoring in pregnantwomen with type 1 diabetes: an observationalcohort study of 186 pregnancies. Diabetologia.23 March 2019 [Epub ahead of print]. DOI:10.1007/s00125-019-4850-061. Stewart ZA, Wilinska ME, Hartnell S, et al.Closed-loop insulin delivery during pregnancy inwomenwith type 1 diabetes. N Engl J Med 2016;375:644–65462. Stewart ZA, Wilinska ME, Hartnell S, et al.Day-and-night closed-loop insulin delivery in abroad population of pregnant women with type 1diabetes: a randomized controlled crossovertrial. Diabetes Care 2018;41:1391–139963. Law GR, Alnaji A, Alrefaii L, et al. Suboptimalnocturnal glucose control is associated with largefor gestational age in treated gestational diabe-tes mellitus. Diabetes Care 2019;42:810–81564. Murphy HR, Rayman G, Lewis K, et al. Ef-fectiveness of continuous glucose monitoring inpregnant women with diabetes: randomisedclinical trial. BMJ 2008;337:a168065. Secher AL, Ringholm L, Andersen HU, Damm P,Mathiesen ER. The effect of real-time continuousglucose monitoring in pregnant women with

diabetes: a randomized controlled trial. DiabetesCare 2013;36:1877–188366. Paramasivam SS, Chinna K, Singh AKK, et al.Continuous glucose monitoring results in lowerHbA1c in Malaysian women with insulin-treatedgestational diabetes: a randomized controlledtrial. Diabet Med 2018;35:1118–112967. Mazze RS, Lucido D, Langer O, Hartmann K,Rodbard D. Ambulatory glucose profile: repre-sentation of verified self-monitored blood glu-cose data. Diabetes Care 1987;10:111–11768. Fonseca V, Grunberger G. Letter to theeditor: standard glucose reporting: follow-upto the February 2016 AACE CGM ConsensusConference. Endocr Pract 2017;23:629–63269. Mullen DM, Bergenstal R, Criego A, ArnoldKC, Goland R, Richter S. Time savings using astandardized glucose reporting system and am-bulatory glucose profile. J Diabetes Sci Technol2018;12:614–62170. Carlson AL, Mullen DM, Bergenstal RM.Clinical use of continuous glucose monitoringin adults with type 2 diabetes. Diabetes TechnolTher 2017;19(Suppl. 2):S4–S1171. Hirsch IB, Verderese CA. Professional flashcontinuous glucose monitoring with ambulatoryglucose profile reporting to supplement A1C:rationale and practical implementation. EndocrPract 2017;23:1333–134472. Kruger DF, Edelman SV, Hinnen DA, ParkinCG. Reference guide for integrating continuousglucose monitoring into clinical practice. Diabe-tes Educ 2019;45(Suppl. 1):3S–20S73. Rodbard D. Continuous glucose monitoring:a review of successes, challenges, and opportu-nities. Diabetes Technol Ther 2016;18(Suppl. 2):S3–S1374. Rodbard D. Glucose variability: a review ofclinical applications and research developments.DiabetesTechnolTher2018;20(Suppl.2):S25–S21575. Bergenstal RM, Beck RW, Close KL, et al.Glucose management indicator (GMI): a new termfor estimating A1C from continuous glucosemonitoring. Diabetes Care 2018;41:2275–228076. Cohen RM, Franco RS, Smith EP, Higgins JM.When HbA1c and blood glucose do not match:howmuch is determined by race, by genetics, bydifferences in mean red blood cell age? J ClinEndocrinol Metab 2019;104:707–71077. Svensson E, Baggesen LM, Johnsen SP, et al.Early glycemic control and magnitude of HbA1creduction predict cardiovascular events andmortality: population-based cohort study of24,752 metformin initiators. Diabetes Care2017;40:800–80778. Safford MM, Shewchuk R, Qu H, et al.Reasons for not intensifying medications: dif-ferentiating “clinical inertia” from appropriatecare. J Gen Intern Med 2007;22:1648–165579. Nathan DM, Genuth S, Lachin J, et al.; Di-abetes Control and Complications Trial ResearchGroup. The effect of intensive treatment of

diabetes on the development and progressionof long-term complications in insulin-dependentdiabetes mellitus. N Engl J Med 1993;329:977–98680. UK Prospective Diabetes Study (UKPDS)Group. Intensive blood-glucose control withsulphonylureas or insulin compared with con-ventional treatment and risk of complications inpatients with type 2 diabetes (UKPDS 33). Lancet1998;352:837–85381. Holman RR, Paul SK, Bethel MA, MatthewsDR, Neil HAW. 10-year follow-up of intensiveglucose control in type 2 diabetes. N Engl J Med2008;359:1577–158982. Ismail-Beigi F, Craven T, Banerji MA, et al.;ACCORD trial group. Effect of intensive treat-ment of hyperglycaemia on microvascular out-comes in type 2 diabetes: an analysis of theACCORD randomised trial. Lancet 2010;376:419–43083. Hayward RA, Reaven PD, Wiitala WL, et al.;VADT Investigators. Follow-up of glycemic con-trol and cardiovascular outcomes in type 2 di-abetes. N Engl J Med 2015;372:2197–220684. DeWalt DA, Davis TC, Wallace AS, et al. Goalsetting in diabetes self-management: taking thebaby steps to success. Patient Educ Couns 2009;77:218–22385. Lawlor KB, Hornyak MJ. SMART goals: howthe application of SMART goals can contribute toachievement of student learning outcomes. De-velopments in Business Simulation and Experi-ential Learning: Proceedings of the Annual ABSELConference 2012;39:259–26786. DiMeglio LA, Acerini C, Codner E, et al. ISPADClinical Practice Consensus Guidelines 2018:glycemic control targets and glucose monitoringfor children, adolescents, and young adults withdiabetes. Pediatr Diabetes 2018;19:105–11487. Aleppo G, Laffel LM, Ahmann AJ, et al. Apractical approach to using trend arrows on theDexcom G5 CGM system for the management ofadults with diabetes. J Endocr Soc 2017;1:1445–146088. Laffel LM, Aleppo G, Buckingham BA, et al. Apractical approach to using trend arrows on theDexcomG5 CGM system tomanage children andadolescents with diabetes. J Endocr Soc 2017;1:1461–147689. Kudva YC, Ahmann AJ, Bergenstal RM, et al.Approach to using trend arrows in the FreeStyleLibre Flash Glucose Monitoring Systems in adults.J Endocr Soc 2018;2:1320–133790. Monnier L, Colette C, Wojtusciszyn A, et al.Toward defining the threshold between low andhigh glucose variability in diabetes. DiabetesCare 2017;40:832–83891. Rodbard D. Hypo- and hyperglycemia inrelation to the mean, standard deviation, co-efficient of variation, and nature of the glucosedistribution. Diabetes Technol Ther 2012;14:868–876

care.diabetesjournals.org Battelino and Associates 1603