2. Classification and Diagnosis ofDiabetesDiabetes Care 2015;38(Suppl. 1):S8S16 | DOI: 10.2337/dc15-S005

CLASSIFICATION

Diabetes can be classied into the following general categories:

1. Type 1 diabetes (due to b-cell destruction, usually leading to absolute insulindeciency)

2. Type 2 diabetes (due to a progressive insulin secretory defect on the backgroundof insulin resistance)

3. Gestational diabetes mellitus (GDM) (diabetes diagnosed in the second or thirdtrimester of pregnancy that is not clearly overt diabetes)

4. Specic types of diabetes due to other causes, e.g., monogenic diabetes syndromes(such as neonatal diabetes and maturity-onset diabetes of the young [MODY]), dis-eases of the exocrine pancreas (such as cystic brosis), and drug- or chemical-induceddiabetes (such as in the treatment of HIV/AIDS or after organ transplantation)

This section reviews most common forms of diabetes but is not comprehensive.For additional information, see the American Diabetes Association (ADA) positionstatement Diagnosis and Classication of Diabetes Mellitus (1).

Assigning a type of diabetes to an individual often depends on the circumstancespresent at the time of diagnosis, with individuals not necessarily tting clearly into asingle category. For example, some patients cannot be clearly classied as havingtype 1 or type 2 diabetes. Clinical presentation and disease progression may varyconsiderably in both types of diabetes.

The traditional paradigms of type 2 diabetes occurring only in adults and type 1diabetes only in children are no longer accurate, as both diseases occur in both cohorts.Occasionally, patients with type 2 diabetes may present with diabetic ketoacidosis(DKA). Children with type 1 diabetes typically present with the hallmark symptomsof polyuria/polydipsia and occasionally with DKA. The onset of type 1 diabetes may bevariable in adults and may not present with the classic symptoms seen in children.However, difculties in diagnosis may occur in children, adolescents, and adults, withthe true diagnosis becoming more obvious over time.

DIAGNOSTIC TESTS FOR DIABETES

Diabetes may be diagnosed based on A1C criteria or plasma glucose criteria, either thefasting plasma glucose (FPG) or the 2-h plasma glucose (2-h PG) value after a 75-g oralglucose tolerance test (OGTT) (1,2) (Table 2.1).

The same tests are used to both screen for and diagnose diabetes. Diabetes maybe identied anywhere along the spectrum of clinical scenarios: in seemingly low-risk individuals who happen to have glucose testing, in symptomatic patients, and inhigher-risk individuals whom the provider tests because of a suspicion of diabetes.The same tests will also detect individuals with prediabetes.

A1CThe A1C test should be performed using a method that is certied by the NGSP andstandardized or traceable to the Diabetes Control and Complications Trial (DCCT)reference assay. Although point-of-care (POC) A1C assays may be NGSP certied,prociency testing is not mandated for performing the test, so use of POC assays fordiagnostic purposes may be problematic and is not recommended.

The A1C has several advantages to the FPG and OGTT, including greater conve-nience (fasting not required), greater preanalytical stability, and less day-to-dayperturbations during stress and illness. These advantages must be balanced by

Suggested citation: American Diabetes Association.Classication and diagnosis of diabetes. Sec. 2.In Standards of Medical Care in Diabetesd2015.Diabetes Care 2015;38(Suppl. 1):S8S16

2015 by the American Diabetes Association.Readers may use this article as long as the workis properly cited, the use is educational and notfor prot, and the work is not altered.

American Diabetes Association

S8 Diabetes Care Volume 38, Supplement 1, January 2015

POSITION

STATE

MEN

T

greater cost, the limited availability ofA1C testing in certain regions of thedeveloping world, and the incompletecorrelation between A1C and averageglucose in certain individuals.

It is important to take age, race/ethnicity, and anemia/hemoglobinopathiesinto consideration when using the A1C todiagnose diabetes.

Age

The epidemiological studies that formedthe framework for recommending A1Cto diagnose diabetes only included adultpopulations. Therefore, it remains un-clear if A1C and the same A1C cut pointshould be used to diagnose diabetes inchildren and adolescents (35).

Race/Ethnicity

A1C levels may vary with patients race/ethnicity (6,7). For example, AfricanAmericans may have higher A1C levelsthan non-Hispanic whites despite simi-lar fasting and postglucose load glucoselevels. A recent epidemiological studyfound that, when matched for FPG,African Americans (with and without di-abetes) had higher A1C levels than non-Hispanicwhites, but also had higher levelsof fructosamine and glycated albuminand lower levels of 1,5-anhydroglucitol,suggesting that their glycemic burden(particularly postprandially) may behigher (8).

Hemoglobinopathies/Anemias

Interpreting A1C levels in the presence ofcertain hemoglobinopathies and anemiamay be problematic. For patients with anabnormal hemoglobin but normal red cellturnover, such as thosewith the sickle celltrait, an A1C assay without interferencefrom abnormal hemoglobins should beused. An updated list of interferences isavailable at www.ngsp.org/interf.asp. Inconditions associated with increased redcell turnover, such as pregnancy (secondand third trimesters), recent blood lossor transfusion, erythropoietin therapy,or hemolysis, only blood glucose criteriashould be used to diagnose diabetes.

Fasting and 2-Hour Plasma GlucoseIn addition to the A1C test, the FPG and2-h PGmay also be used to diagnose diabe-tes (Table 2.1). The concordance betweenthe FPG and 2-h PG tests is imperfect, asis the concordance between A1C and ei-ther glucose-based test. National Healthand Nutrition Examination Survey(NHANES) data indicate that an A1C cutpoint of $6.5% identies one-thirdfewer cases of undiagnosed diabetesthan a fasting glucose cut point of$126 mg/dL (7.0 mmol/L) (9). Numer-ous studies have conrmed that, com-pared with these A1C and FPG cutpoints, the 2-h PG value diagnosesmore people with diabetes. Of note,the lower sensitivity of A1C at the desig-nated cut point may be offset by thetests ease of use and facilitation ofmore widespread testing.

Unless there is a clear clinical diagno-sis (e.g., a patient in a hyperglycemiccrisis or with classic symptoms of hyper-glycemia and a random plasma glucose$200 mg/dL), it is recommended thatthe same test be repeated immediatelyusing a new blood sample for conrma-tion because there will be a greater like-lihood of concurrence. For example, ifthe A1C is 7.0% and a repeat result is6.8%, the diagnosis of diabetes is con-rmed. If two different tests (such asA1Cand FPG) areboth above the diagnos-tic threshold, this also conrms the diag-nosis. On the other hand, if a patient hasdiscordant results from two differenttests, then the test result that is abovethe diagnostic cut point should be re-peated. The diagnosis is made on the ba-sis of the conrmed test. For example, if apatient meets the diabetes criterion ofthe A1C (two results $6.5%), but not

FPG (,126 mg/dL [7.0 mmol/L]), thatperson should nevertheless be consid-ered to have diabetes.

Since all the tests have preanalytic andanalytic variability, it is possible that an ab-normal result (i.e., above the diagnosticthreshold), when repeated, will producea value below the diagnostic cut point.This scenario is least likely for A1C,morelikely for FPG, andmost likely for the 2-hPG, especially if the glucose samples arecollected at room temperature and notcentrifuged promptly. Barring labora-tory error, such patients will likelyhave test results near the margins ofthe diagnostic threshold. The healthcare professional should follow thepatient closely and repeat the test in36 months.

CATEGORIES OF INCREASED RISKFOR DIABETES (PREDIABETES)

Recommendations

c Testing to assess risk for future di-abetes in asymptomatic peopleshould be considered in adults ofany age who are overweight orobese (BMI $25 kg/m2 or $23kg/m2 in Asian Americans) andwho have one or more additionalrisk factors for diabetes. For allpatients, particularly those whoare overweight or obese, testingshould begin at age 45 years. B

c If tests are normal, repeat testingcarried out at a minimum of 3-year intervals is reasonable. C

c To test for prediabetes, the A1C,FPG, and 2-h PG after 75-g OGTTare appropriate. B

c In patients with prediabetes, iden-tify and, if appropriate, treat othercardiovascular disease (CVD) riskfactors. B

c Testing to detect prediabetesshould be considered in childrenand adolescents who are over-weight or obese and who havetwo or more additional risk factorsfor diabetes. E

DescriptionIn 1997 and 2003, the Expert Commit-tee on Diagnosis and Classication ofDiabetes Mellitus (10,11) recognized agroup of individuals whose glucose lev-els did not meet the criteria for diabetesbut were too high to be considered

Table 2.1Criteria for the diagnosisof diabetesA1C$6.5%. The test should be performed

in a laboratory using a method that isNGSP certied and standardized to theDCCT assay.*

OR

FPG $126 mg/dL (7.0 mmol/L). Fasting isdened as no caloric intake for at least8 h.*

OR

2-h PG$200 mg/dL (11.1 mmol/L) duringan OGTT. The test should be performedas described by the WHO, usinga glucose load containing theequivalent of 75 g anhydrous glucosedissolved in water.*

OR

In a patient with classic symptoms ofhyperglycemia or hyperglycemic crisis,a random plasma glucose $200 mg/dL(11.1 mmol/L).

*In the absence of unequivocalhyperglycemia, results should be conrmedby repeat testing.

care.diabetesjournals.org Position Statement S9

normal. Prediabetes is the term usedfor individuals with impaired fastingglucose (IFG) and/or impaired glucosetolerance (IGT) and indicates an in-creased risk for the future develop-ment of diabetes. IFG and IGT shouldnot be viewed as clinical entities intheir own right but rather risk factorsfor diabetes (Table 2.2) and CVD. IFGand IGT are associated with obesity(especially abdominal or visceral obe-sity), dyslipidemia with high triglycer-ides and/or low HDL cholesterol, andhypertension.

DiagnosisIn 1997 and 2003, the Expert Commit-tee on Diagnosis and Classication ofDiabetes Mellitus (10,11) dened IFGas FPG levels 100125 mg/dL (5.66.9mmol/L) and IGT as 2-h PG after 75-gOGTT levels 140199 mg/dL (7.811.0mmol/L). It should be noted that theWorld Health Organization (WHO) andnumerous diabetes organizations de-ne the IFG cutoff at 110 mg/dL (6.1mmol/L).

As with the glucose measures, sev-eral prospective studies that used A1Cto predict the progression to diabetesdemonstrated a strong, continuousassociation between A1C and sub-sequent diabetes. In a systematic re-view of 44,203 individuals from 16cohort studies with a follow-up intervalaveraging 5.6 years (range 2.812years), those with an A1C between5.56.0% had a substantially increasedrisk of diabetes (5-year incidence from

9 to 25%). An A1C range of 6.06.5%had a 5-year risk of developing diabe-tes between 2550% and a relative risk20 times higher compared with an A1Cof 5.0% (12). In a community-basedstudy of African American and non-Hispanic white adults without diabetes,baseline A1C was a stronger predictorof subsequent diabetes and cardiovas-cular events than fasting glucose (13).Other analyses suggest that an A1C of5.7% is associated with a diabetes risksimilar to that of the high-risk partici-pants in the Diabetes Prevention Pro-gram (DPP) (14).

Hence, it is reasonable to consider anA1C range of 5.76.4% as identifying in-dividuals with prediabetes. As with thosewith IFG and/or IGT, individuals with anA1C of 5.76.4% should be informed oftheir increased risk for diabetes and CVDand counseled about effective strategiesto lower their risks (see Section 5. Preven-tion or Delay of Type 2 Diabetes). Similarto glucosemeasurements, the continuumof risk is curvilinear, so as A1C rises, thediabetes risk rises disproportionately(12). Aggressive interventions and vigilantfollow-up should be pursued for thoseconsidered at very high risk (e.g., thosewith A1C.6.0%).

Table 2.3 summarizes the categoriesof prediabetes. For recommendationsregarding risk factors and screening forprediabetes, see p. S12 (Testing forType 2 Diabetes and Prediabetes inAsymptomatic Adults and Testing forType 2 Diabetes and Prediabetes in Chil-dren and Adolescents).

TYPE 1 DIABETES

Recommendationc Inform the relatives of patients with

type 1 diabetes of the opportunityto be tested for type 1 diabetes risk,but only in the setting of a clinicalresearch study. E

Immune-Mediated DiabetesThis form, previously called insulin-dependent diabetes or juvenile-onsetdiabetes, accounts for 510% of diabetesand isduetocellular-mediatedautoimmunedestruction of the pancreatic b-cells.Autoimmune markers include islet cellautoantibodies, autoantibodies to insu-lin, autoantibodies to GAD (GAD65),autoantibodies to the tyrosine phospha-tases IA-2 and IA-2b, and autoantibodiesto zinc transporter 8 (ZnT8). Type 1 di-abetes is dened by the presence of oneor more of these autoimmune markers.The disease has strong HLA associations,with linkage to the DQA and DQB genes.These HLA-DR/DQ alleles can be eitherpredisposing or protective.

The rate of b-cell destruction is quitevariable, being rapid in some individuals(mainly infants and children) and slow inothers (mainly adults). Children andadolescents may present with ketoaci-dosis as the rst manifestation of thedisease. Others have modest fasting hy-perglycemia that can rapidly change tosevere hyperglycemia and/or ketoacido-sis with infection or other stress. Adultsmay retain sufcient b-cell function toprevent ketoacidosis for many years;such individuals eventually become de-pendent on insulin for survival and areat risk for ketoacidosis. At this latterstage of the disease, there is little orno insulin secretion, as manifested bylow or undetectable levels of plasmaC-peptide. Immune-mediated diabetes

Table 2.2Criteria for testing for diabetes or prediabetes in asymptomatic adults1. Testing should be considered in all adults who are overweight (BMI$25 kg/m2 or$23 kg/m2 in

Asian Americans) and have additional risk factors:c physical inactivityc rst-degree relative with diabetesc high-risk race/ethnicity (e.g., African American, Latino, Native American, AsianAmerican, Pacic Islander)

c women who delivered a baby weighing .9 lb or were diagnosed with GDMc hypertension ($140/90 mmHg or on therapy for hypertension)c HDL cholesterol level,35 mg/dL (0.90 mmol/L) and/or a triglyceride level.250 mg/dL(2.82 mmol/L)

c women with polycystic ovary syndromec A1C $5.7%, IGT, or IFG on previous testingc other clinical conditions associated with insulin resistance (e.g., severe obesity,acanthosis nigricans)

c history of CVD

2. For all patients, particularly those who are overweight or obese, testing shouldbegin at age 45 years.

3. If results are normal, testing should be repeated at a minimum of 3-year intervals, withconsideration of more frequent testing depending on initial results (e.g., those withprediabetes should be tested yearly) and risk status.

Table 2.3Categories of increased riskfor diabetes (prediabetes)*FPG 100mg/dL (5.6 mmol/L) to 125 mg/dL

(6.9 mmol/L) (IFG)

OR

2-h PG in the 75-g OGTT 140 mg/dL (7.8mmol/L) to 199mg/dL (11.0mmol/L) (IGT)

OR

A1C 5.76.4%

*For all three tests, risk is continuous,extending below the lower limit of the rangeand becoming disproportionately greater athigher ends of the range.

S10 Position Statement Diabetes Care Volume 38, Supplement 1, January 2015

commonly occurs in childhood and ado-lescence, but it can occur at any age,even in the 8th and 9th decades of life.

Autoimmune destruction ofb-cells hasmultiple genetic predispositions and isalso related to environmental factorsthat are still poorly dened. Although pa-tients are not typically obese when theypresent with type 1 diabetes, obesityshould not preclude the diagnosis. Thesepatients are also prone to other autoim-mune disorders such as Graves disease,Hashimotos thyroiditis, Addisons dis-ease, vitiligo, celiac disease, autoimmunehepatitis, myasthenia gravis, and perni-cious anemia.

Idiopathic DiabetesSome forms of type 1 diabetes have noknownetiologies. These patients haveper-manent insulinopenia and are prone toketoacidosis, but have no evidence of au-toimmunity. Although only a minority ofpatients with type 1 diabetes fall into thiscategory, of those who do, most are ofAfrican or Asian ancestry. Individualswith this form of diabetes suffer from ep-isodic ketoacidosis and exhibit varyingdegrees of insulin deciency between epi-sodes. This form of diabetes is stronglyinherited, lacks immunological evidencefor b-cell autoimmunity, and is not HLAassociated. An absolute requirement forinsulin replacement therapy in affectedpatients may come and go.

Testing for Type 1 Diabetes

The incidence and prevalence of type 1diabetes is increasing (15). Type 1 dia-betic patients often present with acutesymptoms of diabetes and markedly ele-vated blood glucose levels, and some arediagnosed with life-threatening keto-acidosis. Several studies suggest thatmea-suring islet autoantibodies in relatives ofthose with type 1 diabetes may identifyindividuals who are at risk for developingtype 1 diabetes. Such testing, coupledwith education about diabetes symptomsand close follow-up in an observationalclinical study, may enable earlier identi-cation of type 1 diabetes onset. There isevidence to suggest that early diagnosismay limit acute complications (16) andextend long-term endogenous insulinproduction (17).

A recent study reported the risk of pro-gression to type 1 diabetes from the timeof seroconversion to autoantibody posi-tivity in three pediatric cohorts from Fin-land, Germany, and the U.S. Of the 585

children who developed more than twoautoantibodies, nearly 70% developedtype 1 diabetes within 10 years and 84%within 15 years (16,18). Thesendings arehighly signicant because, while the Ger-man group was recruited from offspringof parents with type 1 diabetes, the Finn-ish and American groups were recruitedfrom the general population. Remark-ably, the ndings in all three groupswere the same, suggesting that thesame sequence of events led to clinicaldisease in both sporadic and geneticcases of type 1 diabetes.

While there is currently a lack ofaccepted screening programs, oneshould consider referring relatives ofthose with type 1 diabetes for antibodytesting for risk assessment in thesetting of a clinical research study(http://www2.diabetestrialnet.org).Widespread clinical testing of asymptom-atic low-risk individuals is not currentlyrecommended due to lack of approvedtherapeutic interventions. Higher-risk in-dividuals may be tested, but only in thecontext of a clinical research setting. In-dividuals who test positive will be coun-seled about the risk of developingdiabetes, diabetes symptoms, and DKAprevention. Numerous clinical studiesare being conducted to test variousmeth-ods of preventing type 1 diabetes inthose with evidence of autoimmunity(www.clinicaltrials.gov).

TYPE 2 DIABETES

Recommendations

c Testing to detect type 2 diabetesin asymptomatic people shouldbe considered in adults of anyage who are overweight orobese (BMI $25 kg/m2 or $23kg/m2 in Asian Americans) andwho have one or more addi-tional risk factors for diabetes.For all patients, particularlythose who are overweight orobese, testing should begin atage 45 years. B

c If tests are normal, repeat testingcarried out at a minimum of 3-yearintervals is reasonable. C

c To test for diabetes, the A1C, FPG,and 2-h PG after 75-g OGTT areappropriate. B

c In patients with diabetes, identifyand, if appropriate, treat otherCVD risk factors. B

c Testing to detect type 2 diabetesshould be considered in childrenand adolescents who are over-weight or obese and who havetwo ormore additional risk factorsfor diabetes. E

DescriptionThis form, previously referred to as non-insulin-dependent diabetes or adult-onset diabetes, accounts for ;9095%of all diabetes. Type 2 diabetes encom-passes individuals who have insulin resis-tance and usually relative (rather thanabsolute) insulin deciency. At least ini-tially, and often throughout their lifetime,these individuals may not need insulintreatment to survive.

There are various causes of type 2 di-abetes. Although the specic etiologiesare not known, autoimmune destruc-tion of b-cells does not occur, and pa-tients do not have any of the otherknown causes of diabetes. Most, butnot all, patients with type 2 diabetesare obese. Obesity itself causes somedegree of insulin resistance. Patientswho are not obese by traditional weightcriteria may have an increased percent-age of body fat distributed predomi-nantly in the abdominal region.

Ketoacidosis seldom occurs sponta-neously in type 2 diabetes; when seen,it usually arises in association withthe stress of another illness such as in-fection. Type 2 diabetes frequently goesundiagnosed for many years because hy-perglycemia develops gradually and atearlier stages is often not severe enoughfor the patient to notice the classic di-abetes symptoms. Nevertheless, suchpatients are at an increased risk ofdeveloping macrovascular and micro-vascular complications.

Whereas patients with type 2 diabetesmay have insulin levels that appear nor-mal or elevated, the higher blood glucoselevels in these patientswould be expectedto result in even higher insulin values hadtheir b-cell function been normal. Thus,insulin secretion is defective in these pa-tients and insufcient to compensate forinsulin resistance. Insulin resistance mayimprove with weight reduction and/orpharmacological treatment of hyper-glycemia but is seldom restored to normal.

The risk of developing type 2 diabetesincreases with age, obesity, and lack ofphysical activity. It occurs more frequently

care.diabetesjournals.org Position Statement S11

in women with prior GDM, in those withhypertension or dyslipidemia, and in cer-tain racial/ethnic subgroups (AfricanAmerican, American Indian, Hispanic/Latino, and Asian American). It is oftenassociated with a strong genetic predis-position, more so than type 1 diabetes.However, the genetics of type 2 diabetesis poorly understood.

Testing for Type 2 Diabetes andPrediabetes in Asymptomatic AdultsPrediabetes and diabetes meet criteria forconditions in which early detection is ap-propriate. Both conditions are commonand impose signicant clinical and publichealth burdens. There is often a long pre-symptomatic phase before the diagnosisof type 2 diabetes. Simple tests to detectpreclinical disease are readily available.Thedurationof glycemic burden is a strongpredictor of adverse outcomes. There areeffective interventions that prevent pro-gression from prediabetes to diabetes(see Section 5. Prevention or Delay ofType 2 Diabetes) and reduce the riskof diabetes complications (see Section8. Cardiovascular Disease and RiskManagement and Section 9. Microvas-cular Complications and Foot Care).

Approximately one-quarter of peoplewith diabetes in the U.S. are undiag-nosed. Although screening of asymptom-atic individuals to identify those withprediabetes or diabetes might seem rea-sonable, rigorous clinical trials to provethe effectiveness of such screening havenot been conducted and are unlikely tooccur. A large European randomized con-trolled trial compared the impact ofscreening for diabetes and intensivemultifactorial intervention with that ofscreening and routine care (19). Generalpractice patients between the ages of4069 years were screened for diabetesand randomized by practice to intensivetreatment of multiple risk factors or rou-tine diabetes care. After 5.3 years offollow-up, CVD risk factorsweremodestlybut signicantly improved with intensivetreatment compared with routine care,but the incidence of rst CVD events ormortality was not signicantly differentbetween the groups (19). The excellentcare provided to patients in the routinecare group and the lack of an unscreenedcontrol arm limit our ability to provethat screening and early intensive treat-ment impact outcomes. Mathematicalmodeling studies suggest that screening,

beginning at age 30 or 45 years andindependent of risk factors, may becost-effective (,$11,000 per quality-adjusted life-year gained) (20).

Additional considerations regardingtesting for type 2 diabetes and predia-betes in asymptomatic patients includethe following:

Age

Testing recommendations for diabetes inasymptomatic adults are listed in Table2.2. Age is amajor risk factor for diabetes.Testing should begin at age 45 years for allpatients, particularly those who are over-weight or obese.

BMI and Ethnicity

Testing should be considered in adultsof any age with BMI$25 kg/m2 and oneor more additional risk factors for dia-betes. However, recent data (21) andevidence from the ADA position state-ment BMI Cut Points to Identify At-RiskAsian Americans for Type 2 DiabetesScreening (22) suggest that the BMIcut point should be lower for the AsianAmerican population. For diabetesscreening purposes, the BMI cut pointsfall consistently between 2324 kg/m2

(sensitivity of 80%) for nearly all AsianAmerican subgroups (with levels slightlylower for Japanese Americans). Thismakes a rounded cut point of 23 kg/m2

practical. In determining a single BMIcut point, it is important to balance sen-sitivity and specicity so as to provide avaluable screening tool without numer-ous false positives. An argument can bemade to push the BMI cut point to lowerthan 23 kg/m2 in favor of increasedsensitivity; however, this would leadto an unacceptably low specicity(13.1%). Data from the WHO also sug-gest that a BMI $23 kg/m2 should beused to dene increased risk in AsianAmericans (23).

Evidence also suggests that otherpopulations may benet from lowerBMI cut points. For example, in a largemultiethnic cohort study, for an equiv-alent incidence rate of diabetes, a BMIof 30 kg/m2 in non-Hispanic whites wasequivalent to a BMI of 26 kg/m2 in Afri-can Americans (24).

Medications

Certain medications, such as glucocorti-coids, thiazide diuretics, and atypical anti-psychotics (25), are known to increase therisk of diabetes and should be consideredwhen ascertaining a diagnosis.

Diagnostic Tests

The A1C, FPG, and 2-h PG after 75-gOGTTare appropriate for testing. It should benoted that the tests do not necessarilydetect diabetes in the same individuals.The efcacy of interventions for primaryprevention of type 2 diabetes (2632) hasprimarily been demonstrated among in-dividualswith IGT, not for individualswithisolated IFG or for those with prediabetesdened by A1C criteria.

Testing Interval

The appropriate interval between tests isnot known (33). The rationale for the3-year interval is that with this interval,the number of false-positive tests that re-quire conrmatory testing will be reducedand individuals with false-negative testswill be retested before substantial timeelapses and complications develop (33).

Community Screening

Ideally, testing should be carried outwithin a health care setting because ofthe need for follow-up and treatment.Community testing outside a health caresetting is not recommended because peo-ple with positive tests may not seek, orhave access to, appropriate follow-up test-ing and care. Community testing may alsobe poorly targeted; i.e., it may fail to reachthe groups most at risk and inappropri-ately test those at very low risk or eventhose who have already been diagnosed.

Testing for Type 2 Diabetes andPrediabetes in Children andAdolescentsIn the last decade, the incidence and prev-alence of type2diabetes in adolescents hasincreased dramatically, especially in ethnicpopulations (15). Recent studies questionthe validity of A1C in the pediatric popula-tion, especially among certain ethnicities,and suggest OGTT or FPG as more suitablediagnostic tests (34). However, many ofthese studies do not recognize that diabe-tes diagnostic criteria are based on long-term health outcomes, and validations arenot currently available in the pediatric pop-ulation (35). The ADA acknowledges thelimited data supporting A1C for diagnosingdiabetes in children and adolescents. How-ever, aside from rare instances, such as cys-tic brosis and hemoglobinopathies, theADA continues to recommend A1C in thiscohort (36,37). Themodied recommenda-tions of the ADA consensus report Type 2Diabetes in Children and Adolescents aresummarized in Table 2.4.

S12 Position Statement Diabetes Care Volume 38, Supplement 1, January 2015

GESTATIONAL DIABETES MELLITUS

Recommendations

c Test for undiagnosed type 2 diabe-tes at the rst prenatal visit inthose with risk factors, using stan-dard diagnostic criteria. B

c Test for GDMat 2428weeks of ges-tation in pregnant women not pre-viously known to have diabetes. A

c Screen women with GDM for per-sistent diabetes at 612 weekspostpartum, using the OGTT andclinically appropriate nonpreg-nancy diagnostic criteria. E

c Women with a history of GDMshould have lifelong screening forthe development of diabetes orprediabetes at least every 3 years. B

c Women with a history of GDMfound to have prediabetes shouldreceive lifestyle interventions ormetformin to prevent diabetes. A

DenitionFor many years, GDM was dened asany degree of glucose intolerance thatwas rst recognized during pregnancy(10), regardless of whether the condi-tion may have predated the pregnancyor persisted after the pregnancy. Thisdenition facilitated a uniform strategyfor detection and classication of GDM,but it was limited by imprecision.

The ongoing epidemic of obesity anddiabetes has led to more type 2 diabetesinwomenof childbearing age, resulting inan increase in the number of pregnantwomenwith undiagnosed type2 diabetes(38). Because of the number of pregnantwomen with undiagnosed type 2 diabe-tes, it is reasonable to test women withrisk factors for type 2 diabetes (Table 2.2)at their initial prenatal visit, using stan-dard diagnostic criteria (Table 2.1).Women with diabetes in the rst trimes-ter would be classied as having type 2diabetes. GDM is diabetes diagnosed inthe second or third trimester of preg-nancy that is not clearly overt diabetes.

DiagnosisGDM carries risks for the mother andneonate. Not all adverse outcomes areof equal clinical importance. The Hyper-glycemia and Adverse Pregnancy Out-come (HAPO) study (39), a large-scale(;25,000 pregnant women) multina-tional cohort study, demonstrated thatrisk of adverse maternal, fetal, and neo-natal outcomes continuously increasedas a function of maternal glycemia at2428 weeks, even within ranges previ-ously considered normal for pregnancy.For most complications, there was nothreshold for risk. These results haveled to careful reconsideration of the di-agnostic criteria for GDM. GDM diagno-sis (Table 2.5) can be accomplished witheither of two strategies:

1. One-step 75-g OGTT or2. Two-step approach with a 50-g

(nonfasting) screen followed by a100-g OGTT for those who screenpositive

Different diagnostic criteria will identifydifferent degrees of maternal hypergly-cemia and maternal/fetal risk, leadingsome experts to debate, and disagreeon, optimal strategies for the diagnosisof GDM.

One-Step Strategy

In the 2011 Standards of Care (40), theADA for the rst time recommendedthat all pregnant women not known tohave prior diabetes undergo a 75-gOGTT at 2428 weeks of gestation,based on a recommendation of the In-ternational Association of the Diabetesand Pregnancy Study Groups (IADPSG)(41). The IADPSG dened diagnostic cut

points for GDM as the average glucosevalues (fasting, 1-h, and 2-h PG) in theHAPO study at which odds for adverseoutcomes reached 1.75 times the esti-mated odds of these outcomes at themean glucose levels of the study popu-lation. This one-step strategy was an-ticipated to signicantly increase theincidence of GDM (from 56% to ;1520%), primarily becauseonly oneabnormalvalue, not two, became sufcient to makethe diagnosis. The ADA recognized that theanticipated increase in the incidence ofGDM would have signicant impact onthe costs, medical infrastructure capacity,and potential for increased medicaliza-tion of pregnancies previously catego-rized as normal, but recommended thesediagnostic criteria changes in the contextof worrisomeworldwide increases in obe-sity and diabetes rates with the intent ofoptimizing gestational outcomes forwomen and their offspring.

The expected benets to these preg-nancies and offspring are inferred fromintervention trials that focused onwomen with lower levels of hyperglyce-mia than identied using older GDM di-agnostic criteria and that found modestbenets including reduced rates of large-for-gestational-age births and preeclamp-sia (42,43). It is important to note that8090% of women being treated formild GDM in two randomized controlledtrials (whose glucose values overlappedwith the thresholds recommended bythe IADPSG) could be managed with life-style therapy alone. Data are lacking onhow the treatment of lower levels of hy-perglycemia affects a mothers risk forthe development of type 2 diabetes inthe future and her offsprings risk forobesity, diabetes, and other metabolicdysfunction. Additional well-designedclinical studies are needed to determinethe optimal intensity of monitoring andtreatment of women with GDM diag-nosed by the one-step strategy.

Two-Step Strategy

In 2013, the National Institutes of Health(NIH) convened a consensus develop-ment conference on diagnosing GDM.The15-memberpanel had representativesfrom obstetrics/gynecology, maternal-fetal medicine, pediatrics, diabetes re-search, biostatistics, and other relatedelds to consider diagnostic criteria (44).The panel recommended the two-stepapproach of screening with a 1-h 50-g

Table 2.4Testing for type 2 diabetes orprediabetes in asymptomatic children*Criteria

c Overweight (BMI .85th percentilefor age and sex, weight for height.85th percentile, or weight .120%of ideal for height)

Plus any two of the following risk factors:c Family history of type 2 diabetes inrst- or second-degree relative

c Race/ethnicity (Native American,African American, Latino, AsianAmerican, Pacic Islander)

c Signs of insulin resistance orconditions associated with insulinresistance (acanthosis nigricans,hypertension, dyslipidemia,polycystic ovary syndrome, or small-for-gestational-age birth weight)

c Maternal history of diabetes or GDMduring the childs gestation

Age of initiation: age 10 years or at onsetof puberty, if puberty occurs at ayounger age

Frequency: every 3 years

*Persons aged #18 years.

care.diabetesjournals.org Position Statement S13

glucose load test (GLT) followed by a 3-h100-g OGTT for those who screenpositive, a strategy commonly used inthe U.S.

Key factors reported in the NIH panelsdecision-making process were the lack ofclinical trial interventions demonstratingthe benets of the one-step strategyand the potential negative consequencesof identifying a large new group ofwomen with GDM, including medicaliza-tion of pregnancy with increased inter-ventions and costs. Moreover, screeningwith a 50-g GLT does not require fast-ing and is therefore easier to accomplishfor many women. Treatment of higherthreshold maternal hyperglycemia, asidentied by the two-step approach, re-duces rates of neonatal macrosomia,large-for-gestational-age births, andshoulder dystocia, without increasingsmall-for-gestational-age births (45).The American College of Obstetriciansand Gynecologists (ACOG) updated itsguidelines in 2013 and supported thetwo-step approach (46).

Future Considerations

The conicting recommendations fromexpert groups underscore the fact thatthere are data to support each strategy.The decision of which strategy to

implementmust therefore bemade basedon the relative values placed on factorsthat have yet to be measured (e.g., cost-benet estimation, willingness to changepractice based on correlation studiesrather than clinical intervention trialresults, relative role of cost consid-erations, and available infrastructure lo-cally, nationally, and internationally).

As the IADPSG criteria have beenadopted internationally, further evi-dence has emerged to support im-proved pregnancy outcomes with costsavings (47) and may be the preferredapproach. In addition, pregnanciescomplicated by GDM per IADPSG crite-ria, but not recognized as such, havecomparable outcomes to pregnancies di-agnosed as GDM by the more stringenttwo-step criteria (48). There remainsstrong consensus that establishing a uni-form approach to diagnosing GDM willbenet patients, caregivers, and policy-makers. Longer-term outcome studies arecurrently underway.

MONOGENIC DIABETESSYNDROMES

Monogenic defects that cause b-cell dys-function, such as neonatal diabetes andMODY, represent a small fraction of pa-tients with diabetes (,5%). These forms

of diabetes are frequently characterizedby onset of hyperglycemia at an early age(generally before age 25 years).

Neonatal DiabetesDiabetes diagnosed in therst 6months oflife has been shown not to be typical au-toimmune type 1 diabetes. This so-calledneonatal diabetes can either be transientor permanent. The most common geneticdefect causing transient disease is a defecton ZAC/HYAMI imprinting, whereaspermanent neonatal diabetes is mostcommonly a defect in the gene encodingthe Kir6.2 subunit of the b-cell KATP chan-nel. Diagnosing the latter has implications,since such children can be well managedwith sulfonylureas.

Maturity-Onset Diabetes of the YoungMODY is characterized by impaired insulinsecretion with minimal or no defects in in-sulin action. It is inherited in an autosomaldominant pattern. Abnormalities at six ge-netic loci on different chromosomes havebeen identied to date. Themost commonform is associated with mutations on chro-mosome12 in ahepatic transcription factorreferred to as hepatocyte nuclear factor(HNF)-1a. A second form is associatedwith mutations in the glucokinase geneon chromosome 7p and results in a defec-tive glucokinase molecule. Glucokinaseconverts glucose to glucose-6-phosphate,the metabolism of which, in turn, stimu-lates insulin secretion by the b-cell. Theless common forms of MODY result frommutations in other transcription factors, in-cluding HNF-4a, HNF-1b, insulin promoterfactor (IPF)-1, and NeuroD1.

DiagnosisReadily available commercial genetictesting now enables a true genetic diag-nosis. It is important to correctly diag-nose one of the monogenic forms ofdiabetes because these children maybe incorrectly diagnosed with type 1 ortype 2 diabetes, leading to suboptimaltreatment regimens and delays in diag-nosing other family members (49).

The diagnosis of monogenic diabetesshould be considered in children withthe following ndings:

Diabetes diagnosed within the rst 6months of life

Strong family history of diabetes butwithout typical features of type 2 di-abetes (nonobese, low-risk ethnicgroup)

Table 2.5Screening for and diagnosis of GDMOne-step strategy

Perform a 75-g OGTT, with plasma glucose measurement when patient is fasting and at 1 and2 h, at 2428 weeks of gestation in women not previously diagnosed with overt diabetes.

The OGTT should be performed in the morning after an overnight fast of at least 8 h.

The diagnosis of GDM is made when any of the following plasma glucose values are met orexceeded:

c Fasting: 92 mg/dL (5.1 mmol/L)

c 1 h: 180 mg/dL (10.0 mmol/L)

c 2 h: 153 mg/dL (8.5 mmol/L)

Two-step strategy

Step 1: Perform a 50-g GLT (nonfasting), with plasma glucose measurement at 1 h, at 2428weeks of gestation in women not previously diagnosed with overt diabetes.

If the plasma glucose level measured 1 h after the load is$140 mg/dL* (7.8 mmol/L), proceedto a 100-g OGTT.

Step 2: The 100-g OGTT should be performed when the patient is fasting.

The diagnosis of GDM is made if at least two of the following four plasma glucose levels(measured fasting and 1 h, 2 h, 3 h after the OGTT) are met or exceeded:

Carpenter/Coustan (56) or NDDG (57)

c Fasting 95 mg/dL (5.3 mmol/L) 105 mg/dL (5.8 mmol/L)

c 1 h 180 mg/dL (10.0 mmol/L) 190 mg/dL (10.6 mmol/L)

c 2 h 155 mg/dL (8.6 mmol/L) 165 mg/dL (9.2 mmol/L)

c 3 h 140 mg/dL (7.8 mmol/L) 145 mg/dL (8.0 mmol/L)

NDDG, National Diabetes Data Group.*The ACOG recommends a lower threshold of 135 mg/dL (7.5 mmol/L) in high-risk ethnicpopulations with higher prevalence of GDM; some experts also recommend 130 mg/dL(7.2 mmol/L).

S14 Position Statement Diabetes Care Volume 38, Supplement 1, January 2015

Mild fasting hyperglycemia (100150mg/dL [5.58.5mmol/L]), especially ifyoung and nonobese

Diabetes with negative autoantibod-ies and without signs of obesity or in-sulin resistance

CYSTIC FIBROSISRELATEDDIABETES

Recommendations

c Annual screening for cystic brosisrelated diabetes (CFRD) with OGTTshould begin by age 10 years in allpatients with cystic brosis whodo not have CFRD. B A1C as ascreening test for CFRD is not rec-ommended. B

c Patients with CFRD should betreated with insulin to attain in-dividualized glycemic goals. A

c In patients with cystic brosis andIGT without conrmed diabetes,prandial insulin therapy should beconsidered to maintain weight. B

c Annual monitoring for complica-tions of diabetes is recommended,beginning 5 years after the diagno-sis of CFRD. E

CFRD is themost common comorbidityin people with cystic brosis, occurring inabout 20% of adolescents and 4050% ofadults. Diabetes in this population is as-sociated with worse nutritional status,more severe inammatory lung disease,and greater mortality from respiratoryfailure. Insulin insufciency related topartial brotic destruction of the isletmass is the primary defect in CFRD. Ge-netically determined function of the re-maining b-cells and insulin resistanceassociated with infection and inamma-tion may also play a role. While screeningfor diabetes before the age of 10 yearscan identify risk for progression to CFRDin thosewith abnormal glucose tolerance,there appears to be no benet with re-spect toweight, height, BMI, or lung func-tion compared with those with normalglucose tolerance ,10 years of age. Theuse of continuous glucose monitoringmay be more sensitive than OGTT to de-tect risk for progression to CFRD, but thislikely needs more evidence.

Encouraging data suggest that im-proved screening (50,51) and aggressiveinsulin therapy have narrowed the gap inmortality between cystic brosis patients

with and without diabetes and have elim-inated the sex difference in mortality (52).Recent trials comparing insulin with oralrepaglinide showed no signicant differ-ence between the groups. However, an-other study compared three differentgroups: premeal insulin aspart, repagli-nide, or oral placebo in cystic brosis pa-tients with abnormal glucose tolerance.Patients all had weight loss; however, inthe insulin-treated group, this pattern wasreversed, and they gained 0.39 (6 0.21)BMI units (P 5 0.02). Patients in therepaglinide-treated group had initial weightgain, but this was not sustained by 6months. The placebo group continued tolose weight (53). Insulin remains the mostwidely used therapy for CFRD (54).

Recommendations for the clinicalman-agement of CFRD can be found in theADAposition statement Clinical Care Guide-lines for Cystic FibrosisRelated Diabetes:A Position Statement of the AmericanDiabetes Association and a Clinical Prac-tice Guideline of the Cystic FibrosisFoundation, Endorsed by the PediatricEndocrine Society (55).

References1. American Diabetes Association. Diagnosisand classication of diabetes mellitus. DiabetesCare 2014;37(Suppl. 1):S81S902. The International Expert Committee. Inter-national Expert Committee report on the roleof the A1C assay in the diagnosis of diabetes.Diabetes Care 2009;32:132713343. Nowicka P, Santoro N, Liu H, et al. Utility ofhemoglobin A1c for diagnosing prediabetes anddiabetes in obese children and adolescents. Di-abetes Care 2011;34:130613114. Garca de Guadiana Romualdo L, GonzalezMorales M, Albaladejo Oton MD, et al. [Thevalue of hemoglobin A1c for diagnosis of diabe-tes mellitus and other changes in carbohydratemetabolism in women with recent gestationaldiabetes mellitus] Endocrinol Nutr 2012;59:362366 [in Spanish]5. Cowie CC, Rust KF, Byrd-Holt DD, et al. Prev-alence of diabetes and high risk for diabetesusing A1C criteria in the U.S. population in1988-2006. Diabetes Care 2010;33:5625686. Ziemer DC, Kolm P, Weintraub WS, et al.Glucose-independent, black-white differences inhemoglobin A1c levels: a cross-sectional analysisof 2 studies. Ann Intern Med 2010;152:7707777. Kumar PR, Bhansali A, RavikiranM, et al. Util-ity of glycated hemoglobin in diagnosing type 2diabetes mellitus: a community-based study. JClin Endocrinol Metab 2010;95:283228358. Selvin E, Steffes MW, Ballantyne CM,Hoogeveen RC, Coresh J, Brancati FL. Racial dif-ferences in glycemic markers: a cross-sectionalanalysis of community-based data. Ann InternMed 2011;154:3033099. Picon MJ, Murri M, Mu~noz A, Fernandez-Garca JC, Gomez-Huelgas R, Tinahones FJ.

Hemoglobin A1c versus oral glucose tolerancetest in postpartum diabetes screening. DiabetesCare 2012;35:1648165310. The Expert Committee on the Diagnosis andClassication of DiabetesMellitus. Report of theExpert Committee on the Diagnosis and Classi-cation of Diabetes Mellitus. Diabetes Care1997;20:1183119711. Genuth S, Alberti KG, Bennett P, et al.; Ex-pert Committee on the Diagnosis and Classica-tion of Diabetes Mellitus. Follow-up report onthe diagnosis of diabetes mellitus. DiabetesCare 2003;26:3160316712. Zhang X, Gregg EW,Williamson DF, et al. A1Clevel and future risk of diabetes: a systematic re-view. Diabetes Care 2010;33:1665167313. Selvin E, Steffes MW, Zhu H, et al. Glycatedhemoglobin, diabetes, and cardiovascular risk innondiabetic adults.NEngl JMed2010;362:80081114. Ackermann RT, Cheng YJ, Williamson DF,Gregg EW. Identifying adults at high risk for di-abetes and cardiovascular disease using hemo-globin A1c National Health and NutritionExamination Survey 2005-2006. Am J PrevMed 2011;40:111715. Dabelea D, Mayer-Davis EJ, Saydah S, et al.;SEARCH for Diabetes in Youth Study. Prevalenceof type 1 and type 2 diabetes among childrenand adolescents from 2001 to 2009. JAMA 2014;311:1778178616. Ziegler AG, Rewers M, Simell O, et al. Sero-conversion to multiple islet autoantibodies andrisk of progression to diabetes in children. JAMA2013;309:2473247917. Sorensen JS, Johannesen J, Pociot F, et al.Residual b-cell function 3-6 years after onset oftype 1 diabetes reduces risk of severe hypogly-cemia in children and adolescents. DiabetesCare 2013;36:3454345918. Sosenko JM, Skyler JS, Palmer JP, et al.; Type1 Diabetes TrialNet Study Group; Diabetes Pre-vention Trial-Type 1 Study Group. The predictionof type 1 diabetes bymultiple autoantibody levelsand their incorporation into an autoantibody riskscore in relatives of type 1 diabetic patients. Di-abetes Care 2013;36:2615262019. Grifn SJ, Borch-Johnsen K, DaviesMJ, et al.Effect of early intensive multifactorial therapyon 5-year cardiovascular outcomes in individu-als with type 2 diabetes detected by screening(ADDITION-Europe): a cluster-randomised trial.Lancet 2011;378:15616720. Kahn R, Alperin P, Eddy D, et al. Age at ini-tiation and frequency of screening to detecttype 2 diabetes: a cost-effectiveness analysis.Lancet 2010;375:1365137421. Araneta MR, Gandinetti A, Chang HK. Opti-mum BMI cut points to screen Asian Americansfor type 2 diabetes: the UCSD Filipino HealthStudy and the North Kohala Study. Diabetes2014;63(Suppl. 1):A2022. Hsu WC, Araneta MR, Kanaya AM, ChiangJL, FujimotoW. BMI cut points to identify at-riskAsian Americans for type 2 diabetes screening.Diabetes Care 2015;38:15015823. WHO Expert Consultation. Appropriatebody-mass index for Asian populations and itsimplications for policy and intervention strate-gies. Lancet 2004;363:15716324. Chiu M, Austin PC, Manuel DG, Shah BR, TuJV. Deriving ethnic-specic BMI cutoff points for

care.diabetesjournals.org Position Statement S15

assessing diabetes risk. Diabetes Care 2011;34:1741174825. Erickson SC, Le L, Zakharyan A, et al. New-onset treatment-dependent diabetes mellitusand hyperlipidemia associated with atypical an-tipsychotic use in older adults without schizo-phrenia or bipolar disorder. J Am Geriatr Soc2012;60:47447926. Knowler WC, Barrett-Connor E, Fowler SE,et al.; Diabetes Prevention Program ResearchGroup. Reduction in the incidence of type 2 di-abetes with lifestyle intervention or metformin.N Engl J Med 2002;346:39340327. Tuomilehto J, Lindstrom J, Eriksson JG,et al.; Finnish Diabetes Prevention Study Group.Prevention of type 2 diabetes mellitus bychanges in lifestyle among subjects with im-paired glucose tolerance. N Engl J Med 2001;344:1343135028. Pan X-R, Li G-W, Hu Y-H, et al. Effects of dietand exercise in preventing NIDDM in people withimpaired glucose tolerance: the Da Qing IGT andDiabetes Study. Diabetes Care 1997;20:53754429. Buchanan TA, Xiang AH, Peters RK, et al.Preservation of pancreatic b-cell function andprevention of type 2 diabetes by pharmacolog-ical treatment of insulin resistance in high-riskHispanic women. Diabetes 2002;51:2796280330. Chiasson J-L, Josse RG, Gomis R, HanefeldM, Karasik A, Laakso M; STOP-NIDDM Trial Re-search Group. Acarbose for prevention of type 2diabetesmellitus: the STOP-NIDDM randomisedtrial. Lancet 2002;359:2072207731. Gerstein HC, Yusuf S, Bosch J, et al.; DREAM(Diabetes REduction Assessment with ramipriland rosiglitazone Medication) Trial Investigators.Effect of rosiglitazone on the frequency of diabe-tes in patients with impaired glucose tolerance orimpaired fasting glucose: a randomised con-trolled trial. Lancet 2006;368:1096110532. Ramachandran A, Snehalatha C, Mary S,Mukesh B, Bhaskar AD, Vijay V; Indian DiabetesPrevention Programme (IDPP). The Indian Dia-betes Prevention Programme shows that life-style modication and metformin prevent type2 diabetes in Asian Indian subjects with im-paired glucose tolerance (IDPP-1). Diabetologia2006;49:28929733. Johnson SL, Tabaei BP, Herman WH. Theefcacy and cost of alternative strategies forsystematic screening for type 2 diabetes in theU.S. population 45-74 years of age. DiabetesCare 2005;28:30731134. Buse JB, Kaufman FR, Linder B, Hirst K, ElGhormli L, Willi S; HEALTHY Study Group. Diabe-tes screening with hemoglobin A(1c) versus fast-ing plasma glucose in amultiethnicmiddle-schoolcohort. Diabetes Care 2013;36:429435

35. Kapadia C, Zeitler P; Drugs and TherapeuticsCommittee of the Pediatric Endocrine Society.Hemoglobin A1cmeasurement for the diagnosisof type 2 diabetes in children. Int J Pediatr En-docrinol 2012;2012:3136. Kester LM, Hey H, Hannon TS. Using hemoglo-bin A1c for prediabetes and diabetes diagnosis inadolescents: can adult recommendations be upheldfor pediatric use? JAdolescHealth2012;50:32132337. Wu E-L, Kazzi NG, Lee JM. Cost-effectivenessof screening strategies for identifying pediatricdiabetes mellitus and dysglycemia. JAMA Pe-diatr 2013;167:323938. Lawrence JM, Contreras R, Chen W, SacksDA. Trends in the prevalence of preexistingdiabetes and gestational diabetes mellitusamong a racially/ethnically diverse populationof pregnant women, 1999-2005. Diabetes Care2008;31:89990439. Metzger BE, Lowe LP, Dyer AR, et al.; HAPOStudy Cooperative Research Group. Hyperglyce-mia and adverse pregnancy outcomes. N Engl JMed 2008;358:1991200240. American Diabetes Association. Standardsof medical care in diabetes2011. DiabetesCare 2011;34(Suppl. 1):S11S6141. Metzger BE, Gabbe SG, Persson B, et al.;International Association of Diabetes and Preg-nancy Study Groups Consensus Panel. Interna-tional Association of Diabetes and PregnancyStudy Groups recommendations on the diagno-sis and classication of hyperglycemia in preg-nancy. Diabetes Care 2010;33:67668242. Landon MB, Spong CY, Thom E, et al.; Eu-nice Kennedy Shriver National Institute ofChild Health and Human Development Ma-ternal-Fetal Medicine Units Network. A multi-center, randomized trial of treatment for mildgestational diabetes. N Engl J Med 2009;361:1339134843. Crowther CA, Hiller JE, Moss JR, McPheeAJ, Jeffries WS, Robinson JS; Australian Carbo-hydrate Intolerance Study in PregnantWomen (ACHOIS) Trial Group. Effect of treat-ment of gestational diabetes mellitus on preg-nancy outcomes. N Engl J Med 2005;352:2477248644. Vandorsten JP, Dodson WC, Espeland MA,et al. NIH consensus development conference:diagnosing gestational diabetes mellitus. NIHConsens State Sci Statements 2013;29:13145. Horvath K, Koch K, Jeitler K, et al. Effects oftreatment in women with gestational diabetesmellitus: systematic review and meta-analysis.BMJ 2010;340:c139546. Committee on Practice BulletinsObstetrics.Practice Bulletin No. 137: gestational diabetesmellitus. Obstet Gynecol 2013;122:406416

47. Duran A, Saenz S, Torrejon MJ, et al. In-troduction of IADPSG criteria for the screen-ing and diagnosis of gestational diabetesmellitus results in improved pregnancy out-comes at a lower cost in a large cohort ofpregnant women: the St. Carlos GestationalDiabetes Study. Diabetes Care 2014;37:2442245048. Ethridge JK Jr, Catalano PM, Waters TP.Perinatal outcomes associated with the diagno-sis of gestational diabetes made by the Inter-national Association of the Diabetes andPregnancy Study Groups criteria. Obstet Gyne-col 2014;124:57157849. Hattersley A, Bruining J, Shield J, NjolstadP, Donaghue KC. The diagnosis and manage-ment of monogenic diabetes in children andadolescents. Pediatr Diabetes 2009;10(Suppl.12):334250. Kern AS, Prestridge AL. Improving screeningfor cystic brosis-related diabetes at a pediatriccystic brosis program. Pediatrics 2013;132:e512e51851. Waugh N, Royle P, Craigie I, et al. Screen-ing for cystic brosis-related diabetes: a sys-tematic review. Health Technol Assess 2012;16:iiiiv, 117952. Moran A, Dunitz J, Nathan B, Saeed A,Holme B, Thomas W. Cystic brosis-related di-abetes: current trends in prevalence, incidence,and mortality. Diabetes Care 2009;32:1626163153. Moran A, Pekow P, Grover P, et al.; CysticFibrosis Related Diabetes Therapy Study Group.Insulin therapy to improve BMI in cystic brosis-related diabetes without fasting hyperglycemia:results of the cystic brosis related diabetestherapy trial. Diabetes Care 2009;32:1783178854. Onady GM, Stol A. Insulin and oralagents for managing cystic brosis-related dia-betes. Cochrane Database Syst Rev 2013;7:CD00473055. Moran A, Brunzell C, Cohen RC, et al.; CFRDGuidelines Committee. Clinical care guidelinesfor cystic brosis-related diabetes: a positionstatement of the American Diabetes Associa-tion and a clinical practice guideline of the CysticFibrosis Foundation, endorsed by the PediatricEndocrine Society. Diabetes Care 2010;33:2697270856. Carpenter MW, Coustan DR. Criteria forscreening tests for gestational diabetes. Am JObstet Gynecol 1982;144:76877357. National Diabetes Data Group. Classica-tion and diagnosis of diabetes mellitus andother categories of glucose intolerance. Diabe-tes 1979;28:10391057

S16 Position Statement Diabetes Care Volume 38, Supplement 1, January 2015