kajsa kvist - Academia.edu (original) (raw)

Papers by kajsa kvist

Research paper thumbnail of Association Between Glucagon-Like Peptide 1 Receptor Agonist and Sodium–Glucose Cotransporter 2 Inhibitor Use and COVID-19 Outcomes

Diabetes Care, Jul 1, 2021

Research paper thumbnail of DEVOTE 3: temporal relationships between severe hypoglycaemia, cardiovascular outcomes and mortality

Carolina Digital Repository (University of North Carolina at Chapel Hill), 2018

Research paper thumbnail of Applying the causal roadmap to longitudinal national Danish registry data: a case study of second-line diabetes medication and dementia

arXiv (Cornell University), Oct 3, 2023

Research paper thumbnail of Targeted Maximum Likelihood Based Estimation for Longitudinal Mediation Analysis

arXiv (Cornell University), Apr 10, 2023

Causal mediation analysis with random interventions has become an area of significant interest fo... more Causal mediation analysis with random interventions has become an area of significant interest for understanding time-varying effects with longitudinal and survival outcomes. To tackle causal and statistical challenges due to the complex longitudinal data structure with time-varying confounders, competing risks, and informative censoring, there exists a general desire to combine machine learning techniques and semiparametric theory. In this manuscript, we focus on targeted maximum likelihood estimation (TMLE) of longitudinal natural direct and indirect effects defined with random interventions. The proposed estimators are multiply robust, locally efficient, and directly estimate and update the conditional densities that factorize data likelihoods. We utilize the highly adaptive lasso (HAL) and projection representations to derive new estimators (HAL-EIC) of the efficient influence curves of longitudinal mediation problems and propose a fast one-step TMLE algorithm using HAL-EIC while preserving the asymptotic properties. The proposed method can be generalized for other longitudinal causal parameters that are smooth functions of data likelihoods, and thereby provides a novel and flexible statistical toolbox. 1. Introduction. There is an increasing need of methods that can analyze mechanisms of temporally varying effects, for example, in clinical trials and studies of electronic health

Research paper thumbnail of Do antidepressants precipitate youth suicide?

European Child & Adolescent Psychiatry, Feb 23, 2006

The association between treatment with Selective serotonin reuptake inhibitors (SSRIs) and suicid... more The association between treatment with Selective serotonin reuptake inhibitors (SSRIs) and suicide in children and adolescents on the individual and ecological level were examined in a nationwide Danish pharmacoepidemiological register-linkage study including all persons aged 10-17 years treated with antidepressants during the period 1995-1999 (n=2,569) and a randomly selected control population (n=50,000). A tripartite approach was used. In Part 1, changes in youth suicide and use of antidepressants were examined. In Part 2, we made an assessment of youth suicide characteristics. In Part 3, we analysed the relative risk (RR) of suicide according to antidepressant treatment corrected for psychiatric hospital contact to minimize the problem of confounding by indication. The use of SSRIs among children and adolescents increased substantially during the study period, but the suicide rate remained stable (Part 1). Among 42 suicides nationally aged 10-17 years at death, none was treated with SSRIs within 2 weeks prior to suicide (Part 2). There was an increased rate of suicide associated with SSRIs (RR=4.47), however, not quite significant (95% CI: 0.95-20.96), when adjusted for severity of illness (Part 3). Conclusively, we were not able to identify an association between treatment with SSRIs and completed suicide among children and adolescents.

Research paper thumbnail of A Causal Roadmap for Hybrid Randomized and Real-World Data Designs: Case Study of Semaglutide and Cardiovascular Outcomes

arXiv (Cornell University), May 12, 2023

and board membership of the Association of Clinical and Translational Science. MvdL reports that ... more and board membership of the Association of Clinical and Translational Science. MvdL reports that he is a co-founder of the statistical software start-up company TLrevolution, Inc. MvdL and MP report personal compensation for consultation from Novo Nordisk.

Research paper thumbnail of 89-LB: Low Level of Agreement in Evaluation of Continuous Glucose Monitoring Profiles

Diabetes, Jun 1, 2019

Nearly 20 years after the introduction of Continuous Glucose Monitoring (CGM) it is clear that CG... more Nearly 20 years after the introduction of Continuous Glucose Monitoring (CGM) it is clear that CGM reports can effectively guide clinical decision making but the large amount of information contained in the CGM profile may also complicate the evaluation of the profiles. We hypothesized that there would be a relatively low level of agreement in evaluations of CGM-profiles performed by a diverse group of HCPs and patients, with higher level of agreement between diabetes experts. We investigated the interobserver variation in the two groups when asked to subjectively prioritize 9 Ambulatory Glucose Profiles from ‘best’ to ‘worst’ through an internet-based portal (www.glucoseprofile.com). A total of 88 responses were collected. The groups were defined as non-experts (N = 73) and self-evaluated experts (>10 years of experience and > 1 CGM evaluation per week) (N = 15). The interobserver variation between the groups were calculated using Interclass Correlation Coefficients (ICC) showing moderate agreement among the non-experts group (ICC = 0.60) and a good degree of agreement among the diabetes experts (ICC=0.75) [Figure]. Despite standardized CGM metrics and reports there is still a relatively low level of agreement in interpretation and ranking of CGM-profiles even among diabetes experts. This may allow for undesirable variation in diabetes management. More tools are needed to optimize the use of CGM for clinical decision making. Disclosure E. Hachmann-Nielsen: Employee; Self; Novo Nordisk A/S. K. Kvist: Employee; Self; Novo Nordisk A/S. Stock/Shareholder; Self; Novo Nordisk A/S. R.M. Bergenstal: Research Support; Self; Abbott, Boehringer Ingelheim Pharmaceuticals, Inc., Dexcom, Inc., Glooko, Inc., Hygieia, JDRF, Johnson & Johnson Medical Devices Companies, Lilly Diabetes, Medtronic MiniMed, Inc., Merck & Co., Inc., National Institute of Diabetes and Digestive and Kidney Diseases, Novo Nordisk Inc., Roche Diabetes Care, Sanofi. Other Relationship; Self; Abbott, Helmsley Charitable Trust, Hygieia, Lilly Diabetes, Novo Nordisk Inc., Onduo LLC, Sanofi, UnitedHealth Group Inc. Funding Novo Nordisk

Research paper thumbnail of 11-LB: Nonsevere Hypoglycemia Predicts Increased Risk of Subsequent Severe Events in Patients with Type 2 Diabetes

Diabetes, Jun 1, 2019

It is well-known that higher rates of non-severe hypoglycemic episodes (NSHEs) associate with a g... more It is well-known that higher rates of non-severe hypoglycemic episodes (NSHEs) associate with a greater risk of severe hypoglycemic episodes (SHEs) in patients with type 1 diabetes. We investigated whether a similar association existed in patients with type 2 diabetes (T2D). We explored if annual rate of NSHEs was associated with time to first SHE, cardiovascular (CV) death, time to first major adverse CV event (MACE), and all cause death in patients with T2D using data from LEADER, a CV outcomes trial with 9340 patients with T2D.We used a Cox proportional hazards model adjusted for randomized treatment arm, and annual rate of NSHE as a time-dependent covariate with three levels; A: <2 NSHEs per year (reference), B: 2−11 NSHEs, or C: ≥12 NSHEs. Hazard ratios were used to estimate the association between NSHE and each of the outcomes. Higher rates of NSHE were associated with a higher rate of severe hypoglycemia, MACE, CV death, and all-cause death in patients with T2D (Figure); our results suggest that in this T2D population, a high rate of NSHEs may be associated with more harmful outcomes. Disclosure S.R. Heller: Advisory Panel; Self; Eli Lilly and Company, Novo Nordisk A/S, Sanofi-Aventis, Takeda Pharmaceutical Company Limited, Zealand Pharma A/S. Consultant; Self; Novo Nordisk A/S. Speaker’s Bureau; Self; Eli Lilly and Company, Novo Nordisk A/S. E. Hachmann-Nielsen: Employee; Self; Novo Nordisk A/S. K. Kvist: Employee; Self; Novo Nordisk A/S. Stock/Shareholder; Self; Novo Nordisk A/S. Funding Novo Nordisk

Research paper thumbnail of Type 2 Diabetes Patients Reach Target Glycemic Control Faster Using IDegLira than Either Insulin Degludec or Liraglutide Given Alone

Clinical Drug Investigation, Feb 19, 2016

Background and Objectives The time-course when changes in glycemic control and body weight were f... more Background and Objectives The time-course when changes in glycemic control and body weight were first manifest in patients with type 2 diabetes mellitus (T2DM) treated with a combination of insulin degludec and liraglutide (IDegLira) was assessed, comparing IDegLira to its individual components. Methods Data from weeks 0-12 from two studies were analyzed, one comparing IDegLira to each component (DUAL I), and one comparing IDegLira to insulin degludec titrated to a maximum 50 units (DUAL II). Efficacy endpoints included glycated hemoglobin (HbA 1c) and fasting plasma glucose (FPG) reduction, proportion of patients achieving HbA 1c [\7.0 % (\53.0 mmol/mol)] and FPG (B7.2 mmol/L) targets, and proportion achieving HbA 1c target without hypoglycemia and without hypoglycemia and weight gain. Results Mean HbA 1c was lower, and the proportion of patients reaching target HbA 1c greater, with IDegLira versus comparators (both studies) at weeks 8 and 12. Proportions of patients reaching target HbA 1c without hypoglycemia and without hypoglycemia and weight gain were higher for IDegLira versus insulin degludec, though not versus liraglutide. Mean FPG was lower with IDegLira, and the proportion achieving target FPG higher, versus components (both studies) from weeks 4-12. IDegLira was associated with mean weight reduction from weeks 4-12, although less than with liraglutide alone. Hypoglycemia occurred infrequently in weeks 0-12, with no difference in incidence between IDegLira and insulin degludec in either study. Conclusions IDegLira reduces plasma glucose to a greater extent than its components, measurable within the first 12 weeks of therapy, and without weight gain or an increased hypoglycemia risk versus insulin degludec.

Research paper thumbnail of Adherence to lithium in naturalistic settings: results from a nationwide pharmacoepidemiological study

Bipolar Disorders, Nov 1, 2007

Non-adherence to lithium seems to be prevalent in clinical practice (1, 2) and poor adherence to ... more Non-adherence to lithium seems to be prevalent in clinical practice (1, 2) and poor adherence to mood stabilizers has been found to predict increased risk of recurrence and psychiatric hospitalization (3, 4). Rates of non-adherence to lithium have been found in the range of 18-53% in one review (5) and 20-66% with a median rate of 41% in another (6). Women (7-9) and younger patients (7-11) are at increased risk of relapse and recurrence of affective episodes but reasons for these associations are unclear. One hypothesis is that women and younger patients have decreased adherence to treatment and thus are at increased risk of recurrence. However, only two minor naturalistic studies including 86 (12) and 133 patients (13), respectively, have investigated the influence of age and gender on adherence to lithium and both found non-significant associations, possibly due to small samples of patients and poor statistical power. In general, the number of patients included in prior studies are rather small [up to 400 patients (3)] and patients are mainly recruited from specialized mood disorder clinics or lithium clinics at academic medical centers (3, 5, 6, 12, 14, 15), which most likely leads to selection bias toward patients with the most severe illness. Only one study, by Johnson and McFarland (16), has included a larger number of patients. In this study, adherence to lithium was calculated using a large database of medical records from prepaid group practices in the metropolitan area of Portland, OR, USA, during a six-year period. Among 1,594 patients treated with lithium, the median length of the first Kessing LV, Sønderga˚rd L, Kvist K, Andersen PK. Adherence to lithium in naturalistic settings: results from a nationwide pharmacoepidemiological study.

Research paper thumbnail of 200-LB: Benefit of Dual Therapy with GLP-1 RA and SGLT2i on Cardiovascular Outcomes in Type 2 Diabetes

Diabetes, Jun 20, 2023

Background: Evidence on the cardiovascular benefit of dual glucagon like peptide-1 receptor agoni... more Background: Evidence on the cardiovascular benefit of dual glucagon like peptide-1 receptor agonist (GLP-1 RA) and sodium-glucose co-transporter-2 inhibitor (SGLT2i) therapy is lacking. Aim: To compare the benefit of dual GLP-1 RA and SGLT2i therapy to other dual type 2 diabetes therapies with respect to heart failure and major cardiovascular adverse events (MACE). Methods: From 2010-2021 patients from the Danish nationwide registries were followed from start of dual second-line type 2 diabetes treatment. Primary outcome was heart failure, and secondary outcomes were MACE and all-cause death. The estimated risk following 1 of 4 dual therapy combinations was determined using a longitudinal causal inference framework assuming all patients followed one dual therapy for 5 years. Results: A total of 87,201 persons were included (GLP-1 RA and SGLT2i: 14,831, GLP-1 RA and DPP4/SU/TZD: 20,417, SGLT2i and DPP4/SU/TZD: 22,803, dual DPP4/SU/TZD: 29,150). The 5-year risk ratio (95% CI) of dual GLP-1 RA and SGLT2i therapy compared to reference (dual DPP4/SU/TZD) for heart failure was: 0.93 (0.87;1.00), for MACE: 0.91 (0.87;0.95) and for all cause death: 0.78 (0.74;0.82), see Figure. Conclusion: Dual therapy with GLP-1 RA and SGLT2i showed a benefit on heart failure compared to other dual therapies. Dual therapy with GLP-1 RA and SGLT2i showed a benefit on MACE and all-cause death compared to the reference treatment. Disclosure B. Zareini: None. T. A. Gerds: None. K. K. Sørensen: None. K. K. B. Clemmensen: Employee; Novo Nordisk A/S, Stock/Shareholder; Novo Nordisk A/S. K. Kvist: Employee; Novo Nordisk A/S. J. David: Employee; Novo Nordisk A/S, Sanofi. C. Torp-pedersen: Research Support; Novo Nordisk, Bayer Inc.

Research paper thumbnail of 201-LB: Benefit of Dual Therapy with GLP-1 RA and SGLT2i on Renal Outcomes in Type 2 Diabetes

Diabetes, Jun 20, 2023

Background: There is increased use of dual glucagon like peptide-1 receptor agonist (GLP-1 RA) an... more Background: There is increased use of dual glucagon like peptide-1 receptor agonist (GLP-1 RA) and sodium-glucose co-transporter-2 inhibitor (SGLT2i) therapy, but real-world evidence on the renal benefit is lacking. Aim: To compare the dual use of GLP-1 RA and SGLT2i to other dual second-line type 2 diabetes therapies with respect to renal outcomes. Methods: From 2010-2021 patients from the Danish nationwide registries were followed from start of dual second-line type 2 diabetes treatment. Outcomes were chronic renal disease, end-stage renal disease and >50% eGFR decrease from baseline. The estimated risk following 1 of 4 dual therapy combinations was determined using a longitudinal causal inference framework assuming all patients followed one dual therapy for 5 years. Results: In total 87,201 persons were included (GLP-1 RA and SGLT2i: 14,831, GLP-1 RA and DPP4/SU/TZD: 20,417, SGLT2i and DPP4/SU/TZD: 22,803, dual DPP4/SU/TZD: 29,150). The 5-year risk ratio (95%CI) of dual GLP-1 RA and SGLT2i therapy compared to reference (dual DPP4/SU/TZD) for chronic renal disease: 1.03 (0.97;1.08), end-stage renal disease: 0.12 (0.09;0.15) and >50% decrease in eGFR: 0.52 (0.45;0.59). Conclusion: Dual therapy with GLP-1 RA and SGLT2i compared to other dual therapies results in greater risk reduction of end-stage renal disease and eGFR decline, while dual SGLT2i and DPP4/SU/TZD showed greater risk reduction of chronic renal disease. Disclosure B. Zareini: None. T. A. Gerds: None. K. K. Sørensen: None. K. K. Clemmensen: Employee; Novo Nordisk A/S, Stock/Shareholder; Novo Nordisk A/S. K. Kvist: Employee; Novo Nordisk A/S. J. David: Employee; Novo Nordisk A/S, Sanofi. C. Torp-pedersen: Research Support; Novo Nordisk, Bayer Inc.

Research paper thumbnail of 65-LB: Real-World Continuous Glucose Monitoring Data on Time-in-Range from a U.S. Population, 2015–2019

Diabetes, Jun 1, 2021

Aims: Improvements in technology and access have led to growing adoption of continuous glucose mo... more Aims: Improvements in technology and access have led to growing adoption of continuous glucose monitoring (CGM). Targets for glucose metrics including time in range (TIR) have been published. We aimed to investigate i) the proportion of people with ≥70% TIR and ii) the proportion with ≥70% TIR and Methods: Data were collected from 2015 to 2019 from the Cornerstone4Care (C4C) database, a freely available patient support program for people with both type 1 diabetes (T1D) and type 2 diabetes (T2D) on any treatment type. CGM traces were divided into 14-day periods according to the ambulatory glucose profile (AGP)-reporting system. Only profiles with data aligned with these standards were included. Patient data are presented using the first or the mean of all AGP profiles. Results: In total, 484 persons uploaded CGM-data to the database (4727 AGPs); 242 had T1D and 74 had T2D (self-reported), the rest were unknown (Table). People uploaded between 1 and 75 AGPs (mean=10). Average TIR (70-180 mg/dL) based on mean profiles was 63%, 68% and 64% for T1D, T2D and all, respectively. Conclusions: Less than half of the population achieved ≥70% TIR, ~30% with ≥70% TIR and Disclosure R. M. Bergenstal: Advisory Panel; Self; Abbott Diabetes, Eli Lilly and Company, Novo Nordisk, Onduo LLC., Roche, Sanofi, United Healthcare, Consultant; Self; Abbott Diabetes, Ascensia, Dexcom, Inc., Eli Lilly and Company, Hygieia, Johnson & Johnson, Medtronic, Novo Nordisk, Onduo LLC., Roche, Sanofi, United Healthcare, Other Relationship; Self; HealthPartners Institute, Research Support; Self; Abbott Diabetes, Dexcom, Inc., Eli Lilly and Company, Helmsley Charitable Trust, Hygieia, Johnson & Johnson, Medtronic, NIDDK, Novo Nordisk, Onduo LLC., Roche, Sanofi, United Healthcare. E. Hachmann-nielsen: Employee; Self; Novo Nordisk A/S. J. Tarp: Employee; Self; Novo Nordisk. K. Kvist: Employee; Self; Novo Nordisk A/S. J. B. Buse: Consultant; Self; Cirius Therapeutics, CSL Behring, Fortress Biotch, Mellitus Health, Moderna, Pendulum Therapeutics, Praetego Inc., Stability Health, Zealand Pharma A/S, Other Relationship; Self; AstraZeneca, Eli Lilly and Company, Novo Nordisk, vTv Therapeutics, Research Support; Self; NovaTarg Therapeutics, Novo Nordisk, Sanofi, Tolerion, Inc., vTv Therapeutics, Stock/Shareholder; Self; Mellitus Health, Pendulum Therapeutics, PhaseBio Pharmaceuticals, Inc., Stability Health. Funding Novo Nordisk A/S

Research paper thumbnail of IDegLira, eine Fixkombination aus Insulin degludec und Liraglutid, verbessert sowohl die prä- als auch die postprandiale Plasmaglucose bei Patienten mit Typ 2 Diabetes

Diabetologie Und Stoffwechsel, Apr 29, 2015

Research paper thumbnail of 21-LB: Derived Time-in-Range Is Associated with MACE in T2D: Data from the DEVOTE Trial

Diabetes, Jun 1, 2020

Background: There is a need to validate time-in-range (TIR; percentage of time with plasma glucos... more Background: There is a need to validate time-in-range (TIR; percentage of time with plasma glucose between 70 and 180 mg/dL (3.9-10.0 mmol/L) as a surrogate endpoint for long-term clinical outcomes. Methods: We used data from patients with 8-point glucose profiles (8pp) from the double-blind cardiovascular outcomes trial, DEVOTE (NCT01959529). In total, 7637 patients with T2D and either established CVD or at high risk for CVD were included in the trial. The primary endpoint in DEVOTE was time to first MACE. The 8pp were collected at 1 year, 2 years and end-of-trial. Median length of follow-up was 2 years. For 5644 patients, 8pps with at least 7 points existed. Among the 681 major adverse cardiovascular events (MACEs) in DEVOTE, 360 were among patients with 8pps. Individual TIR was derived as the proportion of the 8pp within range. A Cox model was used to estimate the association between derived TIR and time to first MACE. Hazard ratios (HR) were estimated for patients with TIR>70% vs. TIR≤70%, and for TIR>70% and TIR 50 −70% vs. TIR≤50%. Results: Derived TIR was >70% for 65% of the patients. Estimated rate of first MACE was lower for TIR >70% and TIR 50-70% vs. TIR≤50% (Figure) and for TIR>70% vs. TIR≤70% (HR: 0.74 [0.60;0.91]95% CI; p<0.01). The associations were maintained when analyses were adjusted for baseline characteristics. Conclusions: Derived TIR was associated with rate of first MACE for T2D patients in DEVOTE. Disclosure R.M. Bergenstal: Consultant; Self; Ascensia Diabetes Care, Johnson & Johnson. Other Relationship; Self; Abbott, Dexcom, Inc., Hygieia, Lilly Diabetes, Medtronic, Novo Nordisk A/S, Onduo, Roche Diabetes Care, Sanofi, UnitedHealth Group. E. Hachmann-Nielsen: Employee; Self; Novo Nordisk A/S. K. Kvist: Employee; Self; Novo Nordisk A/S. J.B. Buse: Consultant; Self; Cirius Therapeutics, CSL Behring, Neurimmune. Research Support; Self; American Diabetes Association, National Institutes of Health, Novo Nordisk A/S, Patient-Centered Outcomes Research Institute, Sanofi, Tolerion, Inc., vTv Therapeutics. Stock/Shareholder; Self; Mellitus Health, Pendulum Therapeutics, PhaseBio Pharmaceuticals, Inc., Stability Health. Other Relationship; Self; ADOCIA, AstraZeneca, Dance Biopharm Holdings, Inc., Eli Lilly and Company, MannKind Corporation, NovaTarg Therapeutics, Novo Nordisk A/S, Senseonics, Inc, vTv Therapeutics, Zafgen, Inc. Funding Novo Nordisk A/S

Research paper thumbnail of Benefits of combination of insulin degludec and liraglutide are independent of baseline glycated haemoglobin level and duration of type 2 diabetes

Diabetes, Obesity and Metabolism, Nov 5, 2015

Aim: To evaluate, using post hoc analyses, whether the novel combination of a basal insulin, insu... more Aim: To evaluate, using post hoc analyses, whether the novel combination of a basal insulin, insulin degludec, and a glucagon-like peptide-1 receptor agonist, liraglutide (IDegLira), was consistently effective in patients with type 2 diabetes (T2D), regardless of the stage of T2D progression. Methods: Using data from the DUAL I extension [insulin-naïve patients uncontrolled on oral antidiabetic drugs (OADs), n = 1660, 52 weeks] and DUAL II (patients uncontrolled on basal insulin plus OADs, n = 398, 26 weeks) randomized trials, the efficacy of IDegLira was investigated with regard to measures of disease progression stage including baseline glycated haemoglobin (HbA1c), disease duration and previous insulin dose. Results: Across four categories of baseline HbA1c (≤7.5-9.0%), HbA1c reductions were significantly greater with IDegLira (1.1-2.5%) compared with IDeg or liraglutide alone in DUAL I. In DUAL II, HbA1c reductions were significantly greater with IDegLira (0.9-2.5%) than with IDeg in all but the lowest HbA1c category. In DUAL I, insulin dose and hypoglycaemia rate were lower across all baseline HbA1c categories for IDegLira versus IDeg, while hypoglycaemia was higher with IDegLira than liraglutide, irrespective of baseline HbA1c. In DUAL II, insulin dose and hypoglycaemia rate were similar with IDegLira and IDeg (maximum dose limited to 50 U) independent of baseline HbA1c. The reduction in HbA1c with IDegLira was independent of disease duration and previous insulin dose but varied depending on pre-trial OAD treatment. Conclusions: IDegLira effectively lowered HbA1c across a range of measures, implying suitability for patients with either early or advanced T2D.

Research paper thumbnail of A Cross-Validated Targeted Maximum Likelihood Estimator for Data-Adaptive Experiment Selection Applied to the Augmentation of RCT Control Arms with External Data

arXiv (Cornell University), Oct 11, 2022

Augmenting the control arm of a randomized controlled trial (RCT) with external data may increase... more Augmenting the control arm of a randomized controlled trial (RCT) with external data may increase power at the risk of introducing bias. Existing data fusion estimators generally rely on stringent assumptions or may have decreased coverage or power in the presence of bias. Framing the problem as one of data-adaptive experiment selection, potential experiments include the RCT only or the RCT combined with different candidate real-world datasets. To select and analyze the experiment with the optimal bias-variance tradeoff, we develop a novel experiment-selector cross-validated targeted maximum likelihood estimator (ES-CVTMLE). The ES-CVTMLE uses two bias estimates: 1) a function of the difference in conditional mean outcome under control between the RCT and combined experiments and 2) an estimate of the average treatment effect on a negative control outcome (NCO). We define the asymptotic distribution of the ES-CVTMLE under varying magnitudes of bias and construct confidence intervals by Monte Carlo simulation. In simulations involving violations of identification assumptions, the ES-CVTMLE had better coverage than test-then-pool approaches and an NCO-based bias adjustment approach and higher power than one implementation of a Bayesian dynamic borrowing approach. We further demonstrate the ability of the ES-CVTMLE to distinguish biased from unbiased external controls through a re-analysis of the effect of liraglutide on glycemic control from the LEADER trial. The ES-CVTMLE has the potential to improve power while providing relatively robust inference for future hybrid RCT-RWD studies.

Research paper thumbnail of 8-LB: Outcomes of Type 2 Diabetes (T2D) Clustering Replicated in the DEVOTE Trial

Diabetes, Jun 1, 2019

T2D is a heterogeneous disease. Individuals in the Swedish All New Diabetics in Scania (ANDIS) co... more T2D is a heterogeneous disease. Individuals in the Swedish All New Diabetics in Scania (ANDIS) cohort with newly diagnosed T2D were grouped by 6 demographic and clinical variables to show 4 distinct T2D subtypes with differential risk for nephropathy and retinopathy. We tested the predictive validity of this clustering system for patients with advanced T2D in DEVOTE (a large, global, randomized, double-blind, cardiovascular outcomes trial; median observation time: 1.99 years) for major adverse cardiovascular event (MACE)-free survival, severe hypoglycemia (SH)-free survival, and overall survival rates. Subjects (N=7637, mean age=65.0 years, mean T2D duration=16.4 years, mean glycated hemoglobin [A1C]=8.43%) were assigned to a cluster for which they had the smallest Euclidean distance to the cluster center based on available baseline variables: A1C, BMI, age, age at diagnosis. Insulin resistance and sensitivity measures were not available. The 4 DEVOTE clusters showed baseline characteristics consistent with the original ANDIS clusters, with significant differences in MACE-incidence and SH-incidence (Table). The results were confirmed using data from the LEADER trial (data not shown). The study suggests that clusters derived from early T2D can be replicated in long-standing T2D. Future work should characterize differences in treatment response across clusters to improve outcomes across the heterogeneous T2D population. Disclosure A.R. Kahkoska: None. E. Hachmann-Nielsen: Employee; Self; Novo Nordisk A/S. K. Klein: None. K.G. Kongsbak: Employee; Self; Novo Nordisk A/S. Stock/Shareholder; Self; Novo Nordisk A/S. K. Kvist: Employee; Self; Novo Nordisk A/S. Stock/Shareholder; Self; Novo Nordisk A/S. J.B. Buse: Consultant; Self; Neurimmune AG. Research Support; Self; AstraZeneca, National Center for Advancing Translational Sciences, National Institute of Diabetes and Digestive and Kidney Diseases, Novo Nordisk A/S, Sanofi, vTv Therapeutics. Stock/Shareholder; Self; Mellitus Health, PhaseBio Pharmaceuticals, Inc., Stability Health. Other Relationship; Self; ADOCIA, AstraZeneca, Dance Biopharm Holdings Inc., Eli Lilly and Company, MannKind Corporation, NovaTarg, Novo Nordisk A/S, Senseonics, vTv Therapeutics, Zafgen, Inc. Funding Novo Nordisk

Research paper thumbnail of 2160-PUB: Development of an Evidence-Based Tool to Facilitate Individualized Treatment in the Clinic for Patients with Type 2 Diabetes

Diabetes, Jun 1, 2020

Type 2 diabetes (T2D) is a heterogeneous disease with many therapies available. Current guideline... more Type 2 diabetes (T2D) is a heterogeneous disease with many therapies available. Current guidelines recommend individualized therapy, and this could be facilitated by using estimates of outcomes based on individual characteristics instead of population-level outcomes. There are challenges to defining and implementing individualized therapy in a data-driven and patient-oriented way. We propose criteria to aid implementation of individualized medicine, including: 1) robust insights based on high-quality clinical data, to guide treatment selection; 2) integration of patient-oriented outcomes with engagement of providers and patients; 3) provision of a user-friendly, efficient tool for use in the clinic. We are developing an interactive tool that draws on data from a wide range of randomized controlled trials, e.g., all treatment arms from the full phase 3a program for once-weekly semaglutide. The tool selects the appropriate trial from a database and shows the predicted change in outcomes on initiation/continuation with different therapies over 26-30 weeks. Estimates are based on characteristics such as age, BMI, and T2D duration (Figure). In conclusion, a tool such as this could help clinicians and patients use high-quality, large-scale data from contemporary trials to select individualized treatment regimens. Disclosure A.R. Kahkoska: Other Relationship; Self; Novo Nordisk A/S. J.B. Buse: Consultant; Self; Cirius Therapeutics, CSL Behring, Neurimmune. Research Support; Self; American Diabetes Association, National Institutes of Health, Novo Nordisk A/S, Patient-Centered Outcomes Research Institute, Sanofi, Tolerion, Inc., vTv Therapeutics. Stock/Shareholder; Self; Mellitus Health, Pendulum Therapeutics, PhaseBio Pharmaceuticals, Inc., Stability Health. Other Relationship; Self; ADOCIA, AstraZeneca, Dance Biopharm Holdings, Inc., Eli Lilly and Company, MannKind Corporation, NovaTarg Therapeutics, Novo Nordisk A/S, Senseonics, Inc, vTv Therapeutics, Zafgen, Inc. S. Harring: Employee; Self; Novo Nordisk A/S. Stock/Shareholder; Self; Novo Nordisk A/S. I. Holst: Employee; Self; Novo Nordisk A/S. F.K. Knop: Advisory Panel; Self; AstraZeneca, Merck Sharp & Dohme Corp., Mundipharma International, Novo Nordisk A/S, Sanofi. Consultant; Self; Carmot Therapeutics, Inc., Eli Lilly and Company, Novo Nordisk A/S. Research Support; Self; AstraZeneca, Gubra, Novo Nordisk A/S, Sanofi, Zealand Pharma A/S. Speaker’s Bureau; Self; AstraZeneca, Lupin Pharmaceuticals, Inc., Merck Sharp & Dohme Corp., Norgine B.V., Novo Nordisk A/S. K. Kvist: Employee; Self; Novo Nordisk A/S. R.E. Pratley: Other Relationship; Self; AstraZeneca, Eli Lilly and Company, GlaxoSmithKline plc., Glytec, Janssen Scientific Affairs, LLC., Lexicon Pharmaceuticals, Inc., Ligand Pharmaceuticals Incorporated, Merck & Co., Inc., Novo Nordisk Inc., Sanofi. Funding Novo Nordisk A/S

Research paper thumbnail of 15-LB: Diabetes Duration Did Not Moderate the Reduction in Severe Hypoglycemia Seen with Degludec vs. Glargine U100 in DEVOTE

Diabetes, Jun 1, 2019

Hypoglycemia is associated with significant morbidity in type 2 diabetes (T2D). Progression of T2... more Hypoglycemia is associated with significant morbidity in type 2 diabetes (T2D). Progression of T2D leads to declining counterregulatory responses and heightened hypoglycemia risk. In this secondary analysis of data from DEVOTE, a CVOT with 7637 patients with T2D at high risk of cardiovascular events, we investigated diabetes duration as a moderator of severe hypoglycemia (SH) with use of glargine U100 (IGlar U100) versus insulin degludec. Patients in DEVOTE were categorized by diabetes duration in years: <10, 10-15, 15-20 and ≥20. Prevalence of SH was examined by treatment arm (Table) using a negative binomial model with log-observation time as an offset term, and diabetes duration categories as moderators. Both proportion of patients with SH events and rate of SH increased with progression of disease in both arms. Degludec preserved its relative benefit over IGlar U100 for SH across diabetes duration groups, even in less progressed disease groups, i.e., an observed 52% relative risk reduction for degludec vs. IGlar U100 for patients within 10 years of disease onset (95% CI: 0.29; 0.81). There was no significant difference in treatment ratios across categories (p-value for interaction=0.50). Similar findings were confirmed in the SWITCH 2 trial (data not shown). Thus, diabetes duration did not moderate the relative reduction in rate of SH for degludec vs. IGlar U100 in DEVOTE. Disclosure A. Alexopoulos: None. K. Kvist: Employee; Self; Novo Nordisk A/S. Stock/Shareholder; Self; Novo Nordisk A/S. E. Hachmann-Nielsen: Employee; Self; Novo Nordisk A/S. J.B. Buse: Consultant; Self; Neurimmune AG. Research Support; Self; AstraZeneca, National Center for Advancing Translational Sciences, National Institute of Diabetes and Digestive and Kidney Diseases, Novo Nordisk A/S, Sanofi, vTv Therapeutics. Stock/Shareholder; Self; Mellitus Health, PhaseBio Pharmaceuticals, Inc., Stability Health. Other Relationship; Self; ADOCIA, AstraZeneca, Dance Biopharm Holdings Inc., Eli Lilly and Company, MannKind Corporation, NovaTarg, Novo Nordisk A/S, Senseonics, vTv Therapeutics, Zafgen, Inc. Funding Novo Nordisk A/S

Research paper thumbnail of Association Between Glucagon-Like Peptide 1 Receptor Agonist and Sodium–Glucose Cotransporter 2 Inhibitor Use and COVID-19 Outcomes

Diabetes Care, Jul 1, 2021

Research paper thumbnail of DEVOTE 3: temporal relationships between severe hypoglycaemia, cardiovascular outcomes and mortality

Carolina Digital Repository (University of North Carolina at Chapel Hill), 2018

Research paper thumbnail of Applying the causal roadmap to longitudinal national Danish registry data: a case study of second-line diabetes medication and dementia

arXiv (Cornell University), Oct 3, 2023

Research paper thumbnail of Targeted Maximum Likelihood Based Estimation for Longitudinal Mediation Analysis

arXiv (Cornell University), Apr 10, 2023

Causal mediation analysis with random interventions has become an area of significant interest fo... more Causal mediation analysis with random interventions has become an area of significant interest for understanding time-varying effects with longitudinal and survival outcomes. To tackle causal and statistical challenges due to the complex longitudinal data structure with time-varying confounders, competing risks, and informative censoring, there exists a general desire to combine machine learning techniques and semiparametric theory. In this manuscript, we focus on targeted maximum likelihood estimation (TMLE) of longitudinal natural direct and indirect effects defined with random interventions. The proposed estimators are multiply robust, locally efficient, and directly estimate and update the conditional densities that factorize data likelihoods. We utilize the highly adaptive lasso (HAL) and projection representations to derive new estimators (HAL-EIC) of the efficient influence curves of longitudinal mediation problems and propose a fast one-step TMLE algorithm using HAL-EIC while preserving the asymptotic properties. The proposed method can be generalized for other longitudinal causal parameters that are smooth functions of data likelihoods, and thereby provides a novel and flexible statistical toolbox. 1. Introduction. There is an increasing need of methods that can analyze mechanisms of temporally varying effects, for example, in clinical trials and studies of electronic health

Research paper thumbnail of Do antidepressants precipitate youth suicide?

European Child & Adolescent Psychiatry, Feb 23, 2006

The association between treatment with Selective serotonin reuptake inhibitors (SSRIs) and suicid... more The association between treatment with Selective serotonin reuptake inhibitors (SSRIs) and suicide in children and adolescents on the individual and ecological level were examined in a nationwide Danish pharmacoepidemiological register-linkage study including all persons aged 10-17 years treated with antidepressants during the period 1995-1999 (n=2,569) and a randomly selected control population (n=50,000). A tripartite approach was used. In Part 1, changes in youth suicide and use of antidepressants were examined. In Part 2, we made an assessment of youth suicide characteristics. In Part 3, we analysed the relative risk (RR) of suicide according to antidepressant treatment corrected for psychiatric hospital contact to minimize the problem of confounding by indication. The use of SSRIs among children and adolescents increased substantially during the study period, but the suicide rate remained stable (Part 1). Among 42 suicides nationally aged 10-17 years at death, none was treated with SSRIs within 2 weeks prior to suicide (Part 2). There was an increased rate of suicide associated with SSRIs (RR=4.47), however, not quite significant (95% CI: 0.95-20.96), when adjusted for severity of illness (Part 3). Conclusively, we were not able to identify an association between treatment with SSRIs and completed suicide among children and adolescents.

Research paper thumbnail of A Causal Roadmap for Hybrid Randomized and Real-World Data Designs: Case Study of Semaglutide and Cardiovascular Outcomes

arXiv (Cornell University), May 12, 2023

and board membership of the Association of Clinical and Translational Science. MvdL reports that ... more and board membership of the Association of Clinical and Translational Science. MvdL reports that he is a co-founder of the statistical software start-up company TLrevolution, Inc. MvdL and MP report personal compensation for consultation from Novo Nordisk.

Research paper thumbnail of 89-LB: Low Level of Agreement in Evaluation of Continuous Glucose Monitoring Profiles

Diabetes, Jun 1, 2019

Nearly 20 years after the introduction of Continuous Glucose Monitoring (CGM) it is clear that CG... more Nearly 20 years after the introduction of Continuous Glucose Monitoring (CGM) it is clear that CGM reports can effectively guide clinical decision making but the large amount of information contained in the CGM profile may also complicate the evaluation of the profiles. We hypothesized that there would be a relatively low level of agreement in evaluations of CGM-profiles performed by a diverse group of HCPs and patients, with higher level of agreement between diabetes experts. We investigated the interobserver variation in the two groups when asked to subjectively prioritize 9 Ambulatory Glucose Profiles from ‘best’ to ‘worst’ through an internet-based portal (www.glucoseprofile.com). A total of 88 responses were collected. The groups were defined as non-experts (N = 73) and self-evaluated experts (>10 years of experience and > 1 CGM evaluation per week) (N = 15). The interobserver variation between the groups were calculated using Interclass Correlation Coefficients (ICC) showing moderate agreement among the non-experts group (ICC = 0.60) and a good degree of agreement among the diabetes experts (ICC=0.75) [Figure]. Despite standardized CGM metrics and reports there is still a relatively low level of agreement in interpretation and ranking of CGM-profiles even among diabetes experts. This may allow for undesirable variation in diabetes management. More tools are needed to optimize the use of CGM for clinical decision making. Disclosure E. Hachmann-Nielsen: Employee; Self; Novo Nordisk A/S. K. Kvist: Employee; Self; Novo Nordisk A/S. Stock/Shareholder; Self; Novo Nordisk A/S. R.M. Bergenstal: Research Support; Self; Abbott, Boehringer Ingelheim Pharmaceuticals, Inc., Dexcom, Inc., Glooko, Inc., Hygieia, JDRF, Johnson & Johnson Medical Devices Companies, Lilly Diabetes, Medtronic MiniMed, Inc., Merck & Co., Inc., National Institute of Diabetes and Digestive and Kidney Diseases, Novo Nordisk Inc., Roche Diabetes Care, Sanofi. Other Relationship; Self; Abbott, Helmsley Charitable Trust, Hygieia, Lilly Diabetes, Novo Nordisk Inc., Onduo LLC, Sanofi, UnitedHealth Group Inc. Funding Novo Nordisk

Research paper thumbnail of 11-LB: Nonsevere Hypoglycemia Predicts Increased Risk of Subsequent Severe Events in Patients with Type 2 Diabetes

Diabetes, Jun 1, 2019

It is well-known that higher rates of non-severe hypoglycemic episodes (NSHEs) associate with a g... more It is well-known that higher rates of non-severe hypoglycemic episodes (NSHEs) associate with a greater risk of severe hypoglycemic episodes (SHEs) in patients with type 1 diabetes. We investigated whether a similar association existed in patients with type 2 diabetes (T2D). We explored if annual rate of NSHEs was associated with time to first SHE, cardiovascular (CV) death, time to first major adverse CV event (MACE), and all cause death in patients with T2D using data from LEADER, a CV outcomes trial with 9340 patients with T2D.We used a Cox proportional hazards model adjusted for randomized treatment arm, and annual rate of NSHE as a time-dependent covariate with three levels; A: <2 NSHEs per year (reference), B: 2−11 NSHEs, or C: ≥12 NSHEs. Hazard ratios were used to estimate the association between NSHE and each of the outcomes. Higher rates of NSHE were associated with a higher rate of severe hypoglycemia, MACE, CV death, and all-cause death in patients with T2D (Figure); our results suggest that in this T2D population, a high rate of NSHEs may be associated with more harmful outcomes. Disclosure S.R. Heller: Advisory Panel; Self; Eli Lilly and Company, Novo Nordisk A/S, Sanofi-Aventis, Takeda Pharmaceutical Company Limited, Zealand Pharma A/S. Consultant; Self; Novo Nordisk A/S. Speaker’s Bureau; Self; Eli Lilly and Company, Novo Nordisk A/S. E. Hachmann-Nielsen: Employee; Self; Novo Nordisk A/S. K. Kvist: Employee; Self; Novo Nordisk A/S. Stock/Shareholder; Self; Novo Nordisk A/S. Funding Novo Nordisk

Research paper thumbnail of Type 2 Diabetes Patients Reach Target Glycemic Control Faster Using IDegLira than Either Insulin Degludec or Liraglutide Given Alone

Clinical Drug Investigation, Feb 19, 2016

Background and Objectives The time-course when changes in glycemic control and body weight were f... more Background and Objectives The time-course when changes in glycemic control and body weight were first manifest in patients with type 2 diabetes mellitus (T2DM) treated with a combination of insulin degludec and liraglutide (IDegLira) was assessed, comparing IDegLira to its individual components. Methods Data from weeks 0-12 from two studies were analyzed, one comparing IDegLira to each component (DUAL I), and one comparing IDegLira to insulin degludec titrated to a maximum 50 units (DUAL II). Efficacy endpoints included glycated hemoglobin (HbA 1c) and fasting plasma glucose (FPG) reduction, proportion of patients achieving HbA 1c [\7.0 % (\53.0 mmol/mol)] and FPG (B7.2 mmol/L) targets, and proportion achieving HbA 1c target without hypoglycemia and without hypoglycemia and weight gain. Results Mean HbA 1c was lower, and the proportion of patients reaching target HbA 1c greater, with IDegLira versus comparators (both studies) at weeks 8 and 12. Proportions of patients reaching target HbA 1c without hypoglycemia and without hypoglycemia and weight gain were higher for IDegLira versus insulin degludec, though not versus liraglutide. Mean FPG was lower with IDegLira, and the proportion achieving target FPG higher, versus components (both studies) from weeks 4-12. IDegLira was associated with mean weight reduction from weeks 4-12, although less than with liraglutide alone. Hypoglycemia occurred infrequently in weeks 0-12, with no difference in incidence between IDegLira and insulin degludec in either study. Conclusions IDegLira reduces plasma glucose to a greater extent than its components, measurable within the first 12 weeks of therapy, and without weight gain or an increased hypoglycemia risk versus insulin degludec.

Research paper thumbnail of Adherence to lithium in naturalistic settings: results from a nationwide pharmacoepidemiological study

Bipolar Disorders, Nov 1, 2007

Non-adherence to lithium seems to be prevalent in clinical practice (1, 2) and poor adherence to ... more Non-adherence to lithium seems to be prevalent in clinical practice (1, 2) and poor adherence to mood stabilizers has been found to predict increased risk of recurrence and psychiatric hospitalization (3, 4). Rates of non-adherence to lithium have been found in the range of 18-53% in one review (5) and 20-66% with a median rate of 41% in another (6). Women (7-9) and younger patients (7-11) are at increased risk of relapse and recurrence of affective episodes but reasons for these associations are unclear. One hypothesis is that women and younger patients have decreased adherence to treatment and thus are at increased risk of recurrence. However, only two minor naturalistic studies including 86 (12) and 133 patients (13), respectively, have investigated the influence of age and gender on adherence to lithium and both found non-significant associations, possibly due to small samples of patients and poor statistical power. In general, the number of patients included in prior studies are rather small [up to 400 patients (3)] and patients are mainly recruited from specialized mood disorder clinics or lithium clinics at academic medical centers (3, 5, 6, 12, 14, 15), which most likely leads to selection bias toward patients with the most severe illness. Only one study, by Johnson and McFarland (16), has included a larger number of patients. In this study, adherence to lithium was calculated using a large database of medical records from prepaid group practices in the metropolitan area of Portland, OR, USA, during a six-year period. Among 1,594 patients treated with lithium, the median length of the first Kessing LV, Sønderga˚rd L, Kvist K, Andersen PK. Adherence to lithium in naturalistic settings: results from a nationwide pharmacoepidemiological study.

Research paper thumbnail of 200-LB: Benefit of Dual Therapy with GLP-1 RA and SGLT2i on Cardiovascular Outcomes in Type 2 Diabetes

Diabetes, Jun 20, 2023

Background: Evidence on the cardiovascular benefit of dual glucagon like peptide-1 receptor agoni... more Background: Evidence on the cardiovascular benefit of dual glucagon like peptide-1 receptor agonist (GLP-1 RA) and sodium-glucose co-transporter-2 inhibitor (SGLT2i) therapy is lacking. Aim: To compare the benefit of dual GLP-1 RA and SGLT2i therapy to other dual type 2 diabetes therapies with respect to heart failure and major cardiovascular adverse events (MACE). Methods: From 2010-2021 patients from the Danish nationwide registries were followed from start of dual second-line type 2 diabetes treatment. Primary outcome was heart failure, and secondary outcomes were MACE and all-cause death. The estimated risk following 1 of 4 dual therapy combinations was determined using a longitudinal causal inference framework assuming all patients followed one dual therapy for 5 years. Results: A total of 87,201 persons were included (GLP-1 RA and SGLT2i: 14,831, GLP-1 RA and DPP4/SU/TZD: 20,417, SGLT2i and DPP4/SU/TZD: 22,803, dual DPP4/SU/TZD: 29,150). The 5-year risk ratio (95% CI) of dual GLP-1 RA and SGLT2i therapy compared to reference (dual DPP4/SU/TZD) for heart failure was: 0.93 (0.87;1.00), for MACE: 0.91 (0.87;0.95) and for all cause death: 0.78 (0.74;0.82), see Figure. Conclusion: Dual therapy with GLP-1 RA and SGLT2i showed a benefit on heart failure compared to other dual therapies. Dual therapy with GLP-1 RA and SGLT2i showed a benefit on MACE and all-cause death compared to the reference treatment. Disclosure B. Zareini: None. T. A. Gerds: None. K. K. Sørensen: None. K. K. B. Clemmensen: Employee; Novo Nordisk A/S, Stock/Shareholder; Novo Nordisk A/S. K. Kvist: Employee; Novo Nordisk A/S. J. David: Employee; Novo Nordisk A/S, Sanofi. C. Torp-pedersen: Research Support; Novo Nordisk, Bayer Inc.

Research paper thumbnail of 201-LB: Benefit of Dual Therapy with GLP-1 RA and SGLT2i on Renal Outcomes in Type 2 Diabetes

Diabetes, Jun 20, 2023

Background: There is increased use of dual glucagon like peptide-1 receptor agonist (GLP-1 RA) an... more Background: There is increased use of dual glucagon like peptide-1 receptor agonist (GLP-1 RA) and sodium-glucose co-transporter-2 inhibitor (SGLT2i) therapy, but real-world evidence on the renal benefit is lacking. Aim: To compare the dual use of GLP-1 RA and SGLT2i to other dual second-line type 2 diabetes therapies with respect to renal outcomes. Methods: From 2010-2021 patients from the Danish nationwide registries were followed from start of dual second-line type 2 diabetes treatment. Outcomes were chronic renal disease, end-stage renal disease and >50% eGFR decrease from baseline. The estimated risk following 1 of 4 dual therapy combinations was determined using a longitudinal causal inference framework assuming all patients followed one dual therapy for 5 years. Results: In total 87,201 persons were included (GLP-1 RA and SGLT2i: 14,831, GLP-1 RA and DPP4/SU/TZD: 20,417, SGLT2i and DPP4/SU/TZD: 22,803, dual DPP4/SU/TZD: 29,150). The 5-year risk ratio (95%CI) of dual GLP-1 RA and SGLT2i therapy compared to reference (dual DPP4/SU/TZD) for chronic renal disease: 1.03 (0.97;1.08), end-stage renal disease: 0.12 (0.09;0.15) and >50% decrease in eGFR: 0.52 (0.45;0.59). Conclusion: Dual therapy with GLP-1 RA and SGLT2i compared to other dual therapies results in greater risk reduction of end-stage renal disease and eGFR decline, while dual SGLT2i and DPP4/SU/TZD showed greater risk reduction of chronic renal disease. Disclosure B. Zareini: None. T. A. Gerds: None. K. K. Sørensen: None. K. K. Clemmensen: Employee; Novo Nordisk A/S, Stock/Shareholder; Novo Nordisk A/S. K. Kvist: Employee; Novo Nordisk A/S. J. David: Employee; Novo Nordisk A/S, Sanofi. C. Torp-pedersen: Research Support; Novo Nordisk, Bayer Inc.

Research paper thumbnail of 65-LB: Real-World Continuous Glucose Monitoring Data on Time-in-Range from a U.S. Population, 2015–2019

Diabetes, Jun 1, 2021

Aims: Improvements in technology and access have led to growing adoption of continuous glucose mo... more Aims: Improvements in technology and access have led to growing adoption of continuous glucose monitoring (CGM). Targets for glucose metrics including time in range (TIR) have been published. We aimed to investigate i) the proportion of people with ≥70% TIR and ii) the proportion with ≥70% TIR and Methods: Data were collected from 2015 to 2019 from the Cornerstone4Care (C4C) database, a freely available patient support program for people with both type 1 diabetes (T1D) and type 2 diabetes (T2D) on any treatment type. CGM traces were divided into 14-day periods according to the ambulatory glucose profile (AGP)-reporting system. Only profiles with data aligned with these standards were included. Patient data are presented using the first or the mean of all AGP profiles. Results: In total, 484 persons uploaded CGM-data to the database (4727 AGPs); 242 had T1D and 74 had T2D (self-reported), the rest were unknown (Table). People uploaded between 1 and 75 AGPs (mean=10). Average TIR (70-180 mg/dL) based on mean profiles was 63%, 68% and 64% for T1D, T2D and all, respectively. Conclusions: Less than half of the population achieved ≥70% TIR, ~30% with ≥70% TIR and Disclosure R. M. Bergenstal: Advisory Panel; Self; Abbott Diabetes, Eli Lilly and Company, Novo Nordisk, Onduo LLC., Roche, Sanofi, United Healthcare, Consultant; Self; Abbott Diabetes, Ascensia, Dexcom, Inc., Eli Lilly and Company, Hygieia, Johnson & Johnson, Medtronic, Novo Nordisk, Onduo LLC., Roche, Sanofi, United Healthcare, Other Relationship; Self; HealthPartners Institute, Research Support; Self; Abbott Diabetes, Dexcom, Inc., Eli Lilly and Company, Helmsley Charitable Trust, Hygieia, Johnson & Johnson, Medtronic, NIDDK, Novo Nordisk, Onduo LLC., Roche, Sanofi, United Healthcare. E. Hachmann-nielsen: Employee; Self; Novo Nordisk A/S. J. Tarp: Employee; Self; Novo Nordisk. K. Kvist: Employee; Self; Novo Nordisk A/S. J. B. Buse: Consultant; Self; Cirius Therapeutics, CSL Behring, Fortress Biotch, Mellitus Health, Moderna, Pendulum Therapeutics, Praetego Inc., Stability Health, Zealand Pharma A/S, Other Relationship; Self; AstraZeneca, Eli Lilly and Company, Novo Nordisk, vTv Therapeutics, Research Support; Self; NovaTarg Therapeutics, Novo Nordisk, Sanofi, Tolerion, Inc., vTv Therapeutics, Stock/Shareholder; Self; Mellitus Health, Pendulum Therapeutics, PhaseBio Pharmaceuticals, Inc., Stability Health. Funding Novo Nordisk A/S

Research paper thumbnail of IDegLira, eine Fixkombination aus Insulin degludec und Liraglutid, verbessert sowohl die prä- als auch die postprandiale Plasmaglucose bei Patienten mit Typ 2 Diabetes

Diabetologie Und Stoffwechsel, Apr 29, 2015

Research paper thumbnail of 21-LB: Derived Time-in-Range Is Associated with MACE in T2D: Data from the DEVOTE Trial

Diabetes, Jun 1, 2020

Background: There is a need to validate time-in-range (TIR; percentage of time with plasma glucos... more Background: There is a need to validate time-in-range (TIR; percentage of time with plasma glucose between 70 and 180 mg/dL (3.9-10.0 mmol/L) as a surrogate endpoint for long-term clinical outcomes. Methods: We used data from patients with 8-point glucose profiles (8pp) from the double-blind cardiovascular outcomes trial, DEVOTE (NCT01959529). In total, 7637 patients with T2D and either established CVD or at high risk for CVD were included in the trial. The primary endpoint in DEVOTE was time to first MACE. The 8pp were collected at 1 year, 2 years and end-of-trial. Median length of follow-up was 2 years. For 5644 patients, 8pps with at least 7 points existed. Among the 681 major adverse cardiovascular events (MACEs) in DEVOTE, 360 were among patients with 8pps. Individual TIR was derived as the proportion of the 8pp within range. A Cox model was used to estimate the association between derived TIR and time to first MACE. Hazard ratios (HR) were estimated for patients with TIR>70% vs. TIR≤70%, and for TIR>70% and TIR 50 −70% vs. TIR≤50%. Results: Derived TIR was >70% for 65% of the patients. Estimated rate of first MACE was lower for TIR >70% and TIR 50-70% vs. TIR≤50% (Figure) and for TIR>70% vs. TIR≤70% (HR: 0.74 [0.60;0.91]95% CI; p<0.01). The associations were maintained when analyses were adjusted for baseline characteristics. Conclusions: Derived TIR was associated with rate of first MACE for T2D patients in DEVOTE. Disclosure R.M. Bergenstal: Consultant; Self; Ascensia Diabetes Care, Johnson & Johnson. Other Relationship; Self; Abbott, Dexcom, Inc., Hygieia, Lilly Diabetes, Medtronic, Novo Nordisk A/S, Onduo, Roche Diabetes Care, Sanofi, UnitedHealth Group. E. Hachmann-Nielsen: Employee; Self; Novo Nordisk A/S. K. Kvist: Employee; Self; Novo Nordisk A/S. J.B. Buse: Consultant; Self; Cirius Therapeutics, CSL Behring, Neurimmune. Research Support; Self; American Diabetes Association, National Institutes of Health, Novo Nordisk A/S, Patient-Centered Outcomes Research Institute, Sanofi, Tolerion, Inc., vTv Therapeutics. Stock/Shareholder; Self; Mellitus Health, Pendulum Therapeutics, PhaseBio Pharmaceuticals, Inc., Stability Health. Other Relationship; Self; ADOCIA, AstraZeneca, Dance Biopharm Holdings, Inc., Eli Lilly and Company, MannKind Corporation, NovaTarg Therapeutics, Novo Nordisk A/S, Senseonics, Inc, vTv Therapeutics, Zafgen, Inc. Funding Novo Nordisk A/S

Research paper thumbnail of Benefits of combination of insulin degludec and liraglutide are independent of baseline glycated haemoglobin level and duration of type 2 diabetes

Diabetes, Obesity and Metabolism, Nov 5, 2015

Aim: To evaluate, using post hoc analyses, whether the novel combination of a basal insulin, insu... more Aim: To evaluate, using post hoc analyses, whether the novel combination of a basal insulin, insulin degludec, and a glucagon-like peptide-1 receptor agonist, liraglutide (IDegLira), was consistently effective in patients with type 2 diabetes (T2D), regardless of the stage of T2D progression. Methods: Using data from the DUAL I extension [insulin-naïve patients uncontrolled on oral antidiabetic drugs (OADs), n = 1660, 52 weeks] and DUAL II (patients uncontrolled on basal insulin plus OADs, n = 398, 26 weeks) randomized trials, the efficacy of IDegLira was investigated with regard to measures of disease progression stage including baseline glycated haemoglobin (HbA1c), disease duration and previous insulin dose. Results: Across four categories of baseline HbA1c (≤7.5-9.0%), HbA1c reductions were significantly greater with IDegLira (1.1-2.5%) compared with IDeg or liraglutide alone in DUAL I. In DUAL II, HbA1c reductions were significantly greater with IDegLira (0.9-2.5%) than with IDeg in all but the lowest HbA1c category. In DUAL I, insulin dose and hypoglycaemia rate were lower across all baseline HbA1c categories for IDegLira versus IDeg, while hypoglycaemia was higher with IDegLira than liraglutide, irrespective of baseline HbA1c. In DUAL II, insulin dose and hypoglycaemia rate were similar with IDegLira and IDeg (maximum dose limited to 50 U) independent of baseline HbA1c. The reduction in HbA1c with IDegLira was independent of disease duration and previous insulin dose but varied depending on pre-trial OAD treatment. Conclusions: IDegLira effectively lowered HbA1c across a range of measures, implying suitability for patients with either early or advanced T2D.

Research paper thumbnail of A Cross-Validated Targeted Maximum Likelihood Estimator for Data-Adaptive Experiment Selection Applied to the Augmentation of RCT Control Arms with External Data

arXiv (Cornell University), Oct 11, 2022

Augmenting the control arm of a randomized controlled trial (RCT) with external data may increase... more Augmenting the control arm of a randomized controlled trial (RCT) with external data may increase power at the risk of introducing bias. Existing data fusion estimators generally rely on stringent assumptions or may have decreased coverage or power in the presence of bias. Framing the problem as one of data-adaptive experiment selection, potential experiments include the RCT only or the RCT combined with different candidate real-world datasets. To select and analyze the experiment with the optimal bias-variance tradeoff, we develop a novel experiment-selector cross-validated targeted maximum likelihood estimator (ES-CVTMLE). The ES-CVTMLE uses two bias estimates: 1) a function of the difference in conditional mean outcome under control between the RCT and combined experiments and 2) an estimate of the average treatment effect on a negative control outcome (NCO). We define the asymptotic distribution of the ES-CVTMLE under varying magnitudes of bias and construct confidence intervals by Monte Carlo simulation. In simulations involving violations of identification assumptions, the ES-CVTMLE had better coverage than test-then-pool approaches and an NCO-based bias adjustment approach and higher power than one implementation of a Bayesian dynamic borrowing approach. We further demonstrate the ability of the ES-CVTMLE to distinguish biased from unbiased external controls through a re-analysis of the effect of liraglutide on glycemic control from the LEADER trial. The ES-CVTMLE has the potential to improve power while providing relatively robust inference for future hybrid RCT-RWD studies.

Research paper thumbnail of 8-LB: Outcomes of Type 2 Diabetes (T2D) Clustering Replicated in the DEVOTE Trial

Diabetes, Jun 1, 2019

T2D is a heterogeneous disease. Individuals in the Swedish All New Diabetics in Scania (ANDIS) co... more T2D is a heterogeneous disease. Individuals in the Swedish All New Diabetics in Scania (ANDIS) cohort with newly diagnosed T2D were grouped by 6 demographic and clinical variables to show 4 distinct T2D subtypes with differential risk for nephropathy and retinopathy. We tested the predictive validity of this clustering system for patients with advanced T2D in DEVOTE (a large, global, randomized, double-blind, cardiovascular outcomes trial; median observation time: 1.99 years) for major adverse cardiovascular event (MACE)-free survival, severe hypoglycemia (SH)-free survival, and overall survival rates. Subjects (N=7637, mean age=65.0 years, mean T2D duration=16.4 years, mean glycated hemoglobin [A1C]=8.43%) were assigned to a cluster for which they had the smallest Euclidean distance to the cluster center based on available baseline variables: A1C, BMI, age, age at diagnosis. Insulin resistance and sensitivity measures were not available. The 4 DEVOTE clusters showed baseline characteristics consistent with the original ANDIS clusters, with significant differences in MACE-incidence and SH-incidence (Table). The results were confirmed using data from the LEADER trial (data not shown). The study suggests that clusters derived from early T2D can be replicated in long-standing T2D. Future work should characterize differences in treatment response across clusters to improve outcomes across the heterogeneous T2D population. Disclosure A.R. Kahkoska: None. E. Hachmann-Nielsen: Employee; Self; Novo Nordisk A/S. K. Klein: None. K.G. Kongsbak: Employee; Self; Novo Nordisk A/S. Stock/Shareholder; Self; Novo Nordisk A/S. K. Kvist: Employee; Self; Novo Nordisk A/S. Stock/Shareholder; Self; Novo Nordisk A/S. J.B. Buse: Consultant; Self; Neurimmune AG. Research Support; Self; AstraZeneca, National Center for Advancing Translational Sciences, National Institute of Diabetes and Digestive and Kidney Diseases, Novo Nordisk A/S, Sanofi, vTv Therapeutics. Stock/Shareholder; Self; Mellitus Health, PhaseBio Pharmaceuticals, Inc., Stability Health. Other Relationship; Self; ADOCIA, AstraZeneca, Dance Biopharm Holdings Inc., Eli Lilly and Company, MannKind Corporation, NovaTarg, Novo Nordisk A/S, Senseonics, vTv Therapeutics, Zafgen, Inc. Funding Novo Nordisk

Research paper thumbnail of 2160-PUB: Development of an Evidence-Based Tool to Facilitate Individualized Treatment in the Clinic for Patients with Type 2 Diabetes

Diabetes, Jun 1, 2020

Type 2 diabetes (T2D) is a heterogeneous disease with many therapies available. Current guideline... more Type 2 diabetes (T2D) is a heterogeneous disease with many therapies available. Current guidelines recommend individualized therapy, and this could be facilitated by using estimates of outcomes based on individual characteristics instead of population-level outcomes. There are challenges to defining and implementing individualized therapy in a data-driven and patient-oriented way. We propose criteria to aid implementation of individualized medicine, including: 1) robust insights based on high-quality clinical data, to guide treatment selection; 2) integration of patient-oriented outcomes with engagement of providers and patients; 3) provision of a user-friendly, efficient tool for use in the clinic. We are developing an interactive tool that draws on data from a wide range of randomized controlled trials, e.g., all treatment arms from the full phase 3a program for once-weekly semaglutide. The tool selects the appropriate trial from a database and shows the predicted change in outcomes on initiation/continuation with different therapies over 26-30 weeks. Estimates are based on characteristics such as age, BMI, and T2D duration (Figure). In conclusion, a tool such as this could help clinicians and patients use high-quality, large-scale data from contemporary trials to select individualized treatment regimens. Disclosure A.R. Kahkoska: Other Relationship; Self; Novo Nordisk A/S. J.B. Buse: Consultant; Self; Cirius Therapeutics, CSL Behring, Neurimmune. Research Support; Self; American Diabetes Association, National Institutes of Health, Novo Nordisk A/S, Patient-Centered Outcomes Research Institute, Sanofi, Tolerion, Inc., vTv Therapeutics. Stock/Shareholder; Self; Mellitus Health, Pendulum Therapeutics, PhaseBio Pharmaceuticals, Inc., Stability Health. Other Relationship; Self; ADOCIA, AstraZeneca, Dance Biopharm Holdings, Inc., Eli Lilly and Company, MannKind Corporation, NovaTarg Therapeutics, Novo Nordisk A/S, Senseonics, Inc, vTv Therapeutics, Zafgen, Inc. S. Harring: Employee; Self; Novo Nordisk A/S. Stock/Shareholder; Self; Novo Nordisk A/S. I. Holst: Employee; Self; Novo Nordisk A/S. F.K. Knop: Advisory Panel; Self; AstraZeneca, Merck Sharp & Dohme Corp., Mundipharma International, Novo Nordisk A/S, Sanofi. Consultant; Self; Carmot Therapeutics, Inc., Eli Lilly and Company, Novo Nordisk A/S. Research Support; Self; AstraZeneca, Gubra, Novo Nordisk A/S, Sanofi, Zealand Pharma A/S. Speaker’s Bureau; Self; AstraZeneca, Lupin Pharmaceuticals, Inc., Merck Sharp & Dohme Corp., Norgine B.V., Novo Nordisk A/S. K. Kvist: Employee; Self; Novo Nordisk A/S. R.E. Pratley: Other Relationship; Self; AstraZeneca, Eli Lilly and Company, GlaxoSmithKline plc., Glytec, Janssen Scientific Affairs, LLC., Lexicon Pharmaceuticals, Inc., Ligand Pharmaceuticals Incorporated, Merck & Co., Inc., Novo Nordisk Inc., Sanofi. Funding Novo Nordisk A/S

Research paper thumbnail of 15-LB: Diabetes Duration Did Not Moderate the Reduction in Severe Hypoglycemia Seen with Degludec vs. Glargine U100 in DEVOTE

Diabetes, Jun 1, 2019

Hypoglycemia is associated with significant morbidity in type 2 diabetes (T2D). Progression of T2... more Hypoglycemia is associated with significant morbidity in type 2 diabetes (T2D). Progression of T2D leads to declining counterregulatory responses and heightened hypoglycemia risk. In this secondary analysis of data from DEVOTE, a CVOT with 7637 patients with T2D at high risk of cardiovascular events, we investigated diabetes duration as a moderator of severe hypoglycemia (SH) with use of glargine U100 (IGlar U100) versus insulin degludec. Patients in DEVOTE were categorized by diabetes duration in years: <10, 10-15, 15-20 and ≥20. Prevalence of SH was examined by treatment arm (Table) using a negative binomial model with log-observation time as an offset term, and diabetes duration categories as moderators. Both proportion of patients with SH events and rate of SH increased with progression of disease in both arms. Degludec preserved its relative benefit over IGlar U100 for SH across diabetes duration groups, even in less progressed disease groups, i.e., an observed 52% relative risk reduction for degludec vs. IGlar U100 for patients within 10 years of disease onset (95% CI: 0.29; 0.81). There was no significant difference in treatment ratios across categories (p-value for interaction=0.50). Similar findings were confirmed in the SWITCH 2 trial (data not shown). Thus, diabetes duration did not moderate the relative reduction in rate of SH for degludec vs. IGlar U100 in DEVOTE. Disclosure A. Alexopoulos: None. K. Kvist: Employee; Self; Novo Nordisk A/S. Stock/Shareholder; Self; Novo Nordisk A/S. E. Hachmann-Nielsen: Employee; Self; Novo Nordisk A/S. J.B. Buse: Consultant; Self; Neurimmune AG. Research Support; Self; AstraZeneca, National Center for Advancing Translational Sciences, National Institute of Diabetes and Digestive and Kidney Diseases, Novo Nordisk A/S, Sanofi, vTv Therapeutics. Stock/Shareholder; Self; Mellitus Health, PhaseBio Pharmaceuticals, Inc., Stability Health. Other Relationship; Self; ADOCIA, AstraZeneca, Dance Biopharm Holdings Inc., Eli Lilly and Company, MannKind Corporation, NovaTarg, Novo Nordisk A/S, Senseonics, vTv Therapeutics, Zafgen, Inc. Funding Novo Nordisk A/S