Limin Peng

Emory University, Atlanta, Georgia, United States

Are you Limin Peng?

Claim your profile

Publications (59)205.52 Total impact

  • [Show abstract] [Hide abstract]
    ABSTRACT: In practice, disease outcomes are often measured in a continuous scale, and classification of subjects into meaningful disease categories is of substantive interest. To address this problem, we propose a general analytic framework for determining cut-points of the continuous scale. We develop a unified approach to assessing optimal cut-points based on various criteria, including common agreement and association measures. We study the nonparametric estimation of optimal cut-points. Our investigation reveals that the proposed estimator, though it has been ad-hocly used in practice, pertains to nonstandard asymptotic theory and warrants modifications to traditional inferential procedures. The techniques developed in this work are generally adaptable to study other estimators that are maximizers of nonsmooth objective functions while not belonging to the paradigm of M-estimation. We conduct extensive simulations to evaluate the proposed method and confirm the derived theoretical results. The new method is illustrated by an application to a mental health study.
    No preview · Article · Dec 2015 · Journal of Statistical Planning and Inference

  • No preview · Article · Oct 2015 · Diabetes care
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Background: Managing hyperglycemia and diabetes is challenging in geriatric patients admitted to long-term care (LTC) facilities. Methods: This randomized control trial enrolled patients with type 2 diabetes (T2D) with blood glucose (BG) >180 mg/dL or glycated hemoglobin >7.5% to receive low-dose basal insulin (glargine, starting dose 0.1 U/kg/day) or oral antidiabetic drug (OAD) therapy as per primary care provider discretion for 26 weeks. Both groups received supplemental rapid-acting insulin before meals for BG >200 mg/dL. Primary end point was difference in glycemic control as measured by fasting and mean daily glucose concentration between groups. Results: A total of 150 patients (age: 79±8 years, body mass index: 30.1±6.5 kg/m2, duration of diabetes mellitus: 8.2±5.1 years, randomization BG: 194±97 mg/ dL) were randomized to basal insulin (n=75) and OAD therapy (n=75). There were no differences in the mean fasting BG (131±27 mg/dL vs 123±23 mg/dL, p=0.06) between insulin and OAD groups, but patients treated with insulin had greater mean daily BG (163±39 mg/dL vs 138±27 mg/dL, p<0.001) compared to those treated with OADs. There were no differences in the rate of hypoglycemia (<70 mg/dL) between insulin (27%) and OAD (31%) groups, p=0.58. In addition, there were no differences in the number of hospital complications, emergency room visits, and mortality between treatment groups. Conclusions: The results of this randomized study indicate that elderly patients with T2D in LTC facilities exhibited similar glycemic control, hypoglycemic events and complications when treated with either basal insulin or with oral antidiabetic drugs. Trial registration number: ClinicalTrials.gov Identifier: NCT01131052.
    Full-text · Article · Aug 2015
  • [Show abstract] [Hide abstract]
    ABSTRACT: To evaluate the impact of different subcutaneous basal insulin regimens on glycemic variability (GV) and hospital complications in non-ICU patients with type 2 diabetes (T2D). This study is a post-hoc analysis of 279 general medicine and surgery patients treated with either 'Basal Bolus' insulin regimen, using glargine once daily and glulisine before meals or with 'Basal Plus' regimen, using glargine once daily plus correction doses of glulisine before meals for glucose >140 mg/dl. GV was calculated as mean delta daily glucose, mean standard deviation (SD), and mean amplitude of glycemic excursions (MAGE). Treatment with Basal Bolus and Basal Plus regimens resulted in similar mean daily glucose, hypoglycemia, length of stay, and hospital complications (all, p=NS). There were no differences in GV between treatment groups by delta change (72.5±36 vs. 69.3±34 mg/dl), SD (38.5±18 vs. 37.1±16 mg/dl) and MAGE (67.5±34 vs. 66.1±39 mg/dl), all p=NS. Surgery patients treated with Basal Bolus had higher GV compared to those treated with Basal Plus (delta daily glucose and SD: p=0.02, MAGE: p=0.009), but no difference in GV was found between treatment groups in general medicine patients (p=NS). Patients with hypoglycemia events had higher GV compared to subjects without hypoglycemia (p <0.05), but no association was found between GV and hospital complications (p=NS). Treating hospitalized, non-ICU diabetic patients with Basal Plus insulin regimen resulted in similar glucose control and glycemic variability compared to the standard Basal Bolus insulin regimen. Higher GV was not associated with hospital complications.
    No preview · Article · Aug 2015 · Endocrine Practice
  • [Show abstract] [Hide abstract]
    ABSTRACT: In many observational longitudinal studies, the outcome of interest presents a skewed distribution, is subject to censoring due to detection limit or other reasons, and is observed at irregular times that may follow a outcome-dependent pattern. In this work, we consider quantile regression modeling of such longitudinal data, because quantile regression is generally robust in handling skewed and censored outcomes and is flexible to accommodate dynamic covariate-outcome relationships. Specifically, we study a longitudinal quantile regression model that specifies covariate effects on the marginal quantiles of the longitudinal outcome. Such a model is easy to interpret and can accommodate dynamic outcome profile changes over time. We propose estimation and inference procedures that can appropriately account for censoring and irregular outcome-dependent follow-up. Our proposals can be readily implemented based on existing software for quantile regression. We establish the asymptotic properties of the proposed estimator, including uniform consistency and weak convergence. Extensive simulations suggest good finite-sample performance of the new method. We also present an analysis of data from a long-term study of a population exposed to polybrominated biphenyls (PBB), which uncovers an inhomogeneous PBB elimination pattern that would not be detected by traditional longitudinal data analysis. © 2015, The International Biometric Society.
    No preview · Article · Aug 2015 · Biometrics
  • [Show abstract] [Hide abstract]
    ABSTRACT: Objective: The optimal level of glycemic control needed to improve outcomes in cardiac surgery patients remains controversial. Research design and methods: We randomized patients with diabetes (n = 152) and without diabetes (n = 150) with hyperglycemia to an intensive glucose target of 100-140 mg/dL (n = 151) or to a conservative target of 141-180 mg/dL (n = 151) after coronary artery bypass surgery (CABG) surgery. After the intensive care unit (ICU), patients received a single treatment regimen in the hospital and 90 days postdischarge. Primary outcome was differences in a composite of complications, including mortality, wound infection, pneumonia, bacteremia, respiratory failure, acute kidney injury, and major cardiovascular events. Results: Mean glucose in the ICU was 132 ± 14 mg/dL (interquartile range [IQR] 124-139) in the intensive and 154 ± 17 mg/dL (IQR 142-164) in the conservative group (P < 0.001). There were no significant differences in the composite of complications between intensive and conservative groups (42 vs. 52%, P = 0.08). We observed heterogeneity in treatment effect according to diabetes status, with no differences in complications among patients with diabetes treated with intensive or conservative regimens (49 vs. 48%, P = 0.87), but a significant lower rate of complications in patients without diabetes treated with intensive compared with conservative treatment regimen (34 vs. 55%, P = 0.008). Conclusions: Intensive insulin therapy to target glucose of 100 and 140 mg/dL in the ICU did not significantly reduce perioperative complications compared with target glucose of 141 and 180 mg/dL after CABG surgery. Subgroup analysis showed a lower number of complications in patients without diabetes, but not in patients with diabetes treated with the intensive regimen. Large prospective randomized studies are needed to confirm these findings.
    No preview · Article · Jul 2015 · Diabetes care
  • Qi Zheng · Limin Peng · Xuming He
    [Show abstract] [Hide abstract]
    ABSTRACT: Quantile regression has become a valuable tool to analyze heterogeneous covaraite-response associations that are often encountered in practice. The development of quantile regression methodology for high-dimensional covariates primarily focuses on examination of model sparsity at a single or multiple quantile levels, which are typically pre-specified ad hoc by the users. The resulting models may be sensitive to the specific choices of the quantile levels, leading to difficulties in interpretation and erosion of confidence in the results. In this article, we propose a new penalization framework for quantile regression in the high-dimensional setting. We employ adaptive L1 penalties, and more importantly, propose a uniform selector of the tuning parameter for a set of quantile levels to avoid some of the potential problems with model selection at individual quantile levels. Our proposed approach achieves consistent shrinkage of regression quantile estimates across a continuous range of quantiles levels, enhancing the flexibility and robustness of the existing penalized quantile regression methods. Our theoretical results include the oracle rate of uniform convergence and weak convergence of the parameter estimators. We also use numerical studies to confirm our theoretical findings and illustrate the practical utility of our proposal
    No preview · Article · Jul 2015 · The Annals of Statistics
  • [Show abstract] [Hide abstract]
    ABSTRACT: In survival analysis, quantile regression has become a useful approach to account for covariate effects on the distribution of an event time of interest. In this paper, we discuss how quantile regression can be extended to model counting processes, and thus lead to a broader regression framework for survival data. We specifically investigate the proposed modeling of counting processes for recurrent events data. We show that the new recurrent events model retains the desirable features of quantile regression such as easy interpretation and good model flexibility, while accommodating various observation schemes encountered in observational studies. We develop a general theoretical and inferential framework for the new counting process model, which unifies with an existing method for censored quantile regression. As another useful contribution of this work, we propose a sample-based covariance estimation procedure, which provides a useful complement to the prevailing bootstrapping approach. We demonstrate the utility of our proposals via simulation studies and an application to a dataset from the US Cystic Fibrosis Foundation Patient Registry (CFFPR).
    No preview · Article · Mar 2015 · Journal of the American Statistical Association
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Clinical guidelines recommend point-of-care glucose testing and the use of supplemental doses of rapid-acting insulin before meals and at bedtime for correction of hyperglycemia. The efficacy and safety of this recommendation, however, have not been tested in the hospital setting. In this open-label, randomized controlled trial, 206 general medicine and surgery patients with type 2 diabetes treated with a basal-bolus regimen were randomized to receive either supplemental insulin (n = 106) at bedtime for blood glucose (BG) >7.8 mmol/L or no supplemental insulin (n = 100) except for BG >19.4 mmol/L. Point-of-care testing was performed before meals, at bedtime, and at 3:00 a.m. The primary outcome was the difference in fasting BG. In addition to the intention-to-treat analysis, an as-treated analysis was performed where the primary outcome was analyzed for only the bedtime BG levels between 7.8 and 19.4 mmol/L. There were no differences in mean fasting BG for the intention-to-treat (8.8 ± 2.4 vs. 8.6 ± 2.2 mmol/L, P = 0.76) and as-treated (8.9 ± 2.4 vs. 8.8 ± 2.4 mmol/L, P = 0.92) analyses. Only 66% of patients in the supplement and 8% in the no supplement groups received bedtime supplemental insulin. Hypoglycemia (BG <3.9 mmol/L) did not differ between groups for either the intention-to-treat (30% vs. 26%, P = 0.50) or the as-treated (4% vs. 8%, P = 0.37) analysis. The use of insulin supplements for correction of bedtime hyperglycemia was not associated with an improvement in glycemic control. We conclude that routine use of bedtime insulin supplementation is not indicated for management of inpatients with type 2 diabetes. © 2015 by the American Diabetes Association. Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered.
    Full-text · Article · Feb 2015 · Diabetes Care
  • Ruosha Li · Limin Peng
    [Show abstract] [Hide abstract]
    ABSTRACT: Semi-competing risks data frequently arise in biomedical studies when time to a disease landmark event is subject to dependent censoring by death, the observation of which however is not precluded by the occurrence of the landmark event. In observational studies, the analysis of such data can be further complicated by left truncation. In this work, we study a varying coefficient subdistribution regression model for left-truncated semi-competing risks data. Our method appropriately accounts for the specifical truncation and censoring features of the data, and moreover has the flexibility to accommodate potentially varying covariate effects. The proposed method can be easily implemented and the resulting estimators are shown to have nice asymptotic properties. We also present inference, such as Kolmogorov–Smirnov type and Cramér–Von-Mises type hypothesis testing procedures for the covariate effects. Simulation studies and an application to the Denmark diabetes registry demonstrate good finite-sample performance and practical utility of the proposed method.
    No preview · Article · Oct 2014 · Journal of Multivariate Analysis
  • Limin Peng · Jinfeng Xu · Nancy Kutner
    [Show abstract] [Hide abstract]
    ABSTRACT: Varying covariate effects often manifest meaningful heterogeneity in covariate-response associations. In this paper, we adopt a quantile regression model that assumes linearity at a continuous range of quantile levels as a tool to explore such data dynamics. The consideration of potential non-constancy of covariate effects necessitates a new perspective for variable selection, which, under the assumed quantile regression model, is to retain variables that have effects on all quantiles of interest as well as those that influence only part of quantiles considered. Current work on l 1-penalized quantile regression either does not concern varying covariate effects or may not produce consistent variable selection in the presence of covariates with partial effects, a practical scenario of interest. In this work, we propose a shrinkage approach by adopting a novel uniform adaptive LASSO penalty. The new approach enjoys easy implementation without requiring smoothing. Moreover, it can consistently identify the true model (uniformly across quantiles) and achieve the oracle estimation efficiency. We further extend the proposed shrinkage method to the case where responses are subject to random right censoring. Numerical studies confirm the theoretical results and support the utility of our proposals.
    No preview · Article · Sep 2014 · Statistics and Computing
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Objective: Effective treatment algorithms are needed to guide diabetes care at hospital discharge in general medicine and surgery patients with type 2 diabetes. Research design and methods: This was a prospective, multicenter open-label study aimed to determine the safety and efficacy of a hospital discharge algorithm based on admission HbA1c. Patients with HbA1c <7% (53.0 mmol/mol) were discharged on their preadmission diabetes therapy, HbA1c between 7 and 9% (53.0-74.9 mmol/mol) were discharged on a preadmission regimen plus glargine at 50% of hospital daily dose, and HbA1c >9% were discharged on oral antidiabetes agents (OADs) plus glargine or basal bolus regimen at 80% of inpatient dose. The primary outcome was HbA1c concentration at 12 weeks after hospital discharge. Results: A total of 224 patients were discharged on OAD (36%), combination of OAD and glargine (27%), basal bolus (24%), glargine alone (9%), and diet (4%). The admission HbA1c was 8.7 ± 2.5% (71.6 mmol/mol) and decreased to 7.3 ± 1.5% (56 mmol/mol) at 12 weeks of follow-up (P < 0.001). The change of HbA1c from baseline at 12 weeks after discharge was -0.1 ± 0.6, -0.8 ± 1.0, and -3.2 ± 2.4 in patients with HbA1c <7%, 7-9%, and >9%, respectively (P < 0.001). Hypoglycemia (<70 mg/dL) was reported in 22% of patients discharged on OAD only, 30% on OAD plus glargine, 44% on basal bolus, and 25% on glargine alone and was similar in patients with admission HbA1c ≤7% (26%) compared with those with HbA1c >7% (31%, P = 0.54). Conclusions: Measurement of HbA1c on admission is beneficial in tailoring treatment regimens at discharge in general medicine and surgery patients with type 2 diabetes.
    Full-text · Article · Aug 2014 · Diabetes Care
  • Shuang Ji · Limin Peng · Ruosha Li · Michael J Lynn
    [Show abstract] [Hide abstract]
    ABSTRACT: Dependent censoring occurs in many biomedical studies and poses considerable methodological challenges for survival analysis. In this work, we develop a new approach for analyzing dependently censored data by adopting quantile regression models. We formulate covariate effects on the quantiles of the marginal distribution of the event time of interest. Such a modeling strategy can accommodate a more dynamic relationship between covariates and survival time compared to traditional regression models in survival analysis, which usually assume constant covariate effects. We propose estimation and inference procedures, along with an efficient and stable algorithm. We establish the uniform consistency and weak convergence of the resulting estimators. Extensive simulation studies demonstrate good finite-sample performance of the proposed inferential procedures. We illustrate the practical utility of our method via an application to a multicenter clinical trial that compared warfarin and aspirin in treating symptomatic intracranial arterial stenosis.
    No preview · Article · Jul 2014 · Statistica Sinica
  • Ruosha Li · Limin Peng
    [Show abstract] [Hide abstract]
    ABSTRACT: We study quantile regression when the response is an event time subject to potentially dependent censoring. We consider the semicompeting risks setting, where the time to censoring remains observable after the occurrence of the event of interest. Although such a scenario frequently arises in biomedical studies, most of current quantile regression methods for censored data are not applicable because they generally require the censoring time and the event time to be independent. By imposing quite mild assumptions on the association structure between the time-to-event response and the censoring time variable, we propose quantile regression procedures, which allow us to garner a comprehensive view of the covariate effects on the event time outcome as well as to examine the informativeness of censoring. An efficient and stable algorithm is provided for implementing the new method. We establish the asymptotic properties of the resulting estimators including uniform consistency and weak convergence. The theoretical development may serve as a useful template for addressing estimating settings that involve stochastic integrals. Extensive simulation studies suggest that the method proposed performs well with moderate sample sizes. We illustrate the practical utility of our proposals through an application to a bone marrow transplant trial.
    No preview · Article · Mar 2014 · Journal Of The Royal Statistical Society
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Few studies have reported on the quality of diabetes care and glycemic control adjusted for medication use in long term care (LTC) facilities. This observational study analyzed diabetes prevalence and management and the impact of glycemic control on clinical outcome in elderly subjects admitted to 3 community LTC facilities. Among 1409 LTC residents (age 79.7 ± 12 years), the prevalence of diabetes was 34.2%. Subjects with diabetes were either on no pharmacological agents (10%) or were treated with sliding scale regular insulin (SSI, 25%), oral antidiabetic drugs (OAD, 5%), insulin (34%), or with combination of OAD and insulin (26%). Patients with diabetes had a mean daily BG of 156 ± 39 mg/dL and a mean admission HbA1c of 6.7% ± 1.1%. Compared with nondiabetes, residents with diabetes had higher number of complications (54% vs 45%, P < .001), infections (26% vs 21%, P = .036), emergency room (ER) and hospital transfers (37% vs 30%, P = .003), but similar mortality (15% vs 14%, P = .56). A total of 43% of residents with diabetes had a BG less than 70 mg/dL, and those with hypoglycemia had longer median length of stay (LOS, 52 vs 29 days, P < .001), more ER or hospital transfers (56% vs 69%, P = .005), and mortality (20% vs 10%, P = .002) compared with residents without hypoglycemia. Diabetes is common in LTC residents and is associated with higher resource utilization and complications. Hypoglycemia is common and is associated with increased need of emergency room visits and hospitalization and higher mortality. Our findings emphasize the need for randomized trials evaluating the impact of different approaches to glycemic management on clinical outcome in LTC residents with diabetes.
    Full-text · Article · Sep 2013 · Journal of the American Medical Directors Association
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Objective: Hyperglycemia is associated with increased mortality in critically ill patients treated with total parenteral nutrition (TPN). The role of glucose variability (GV) in predicting outcomes in these patients is not known.Methods: This retrospective study included medical and surgical patients receiving TPN in a community teaching hospital. GV was calculated by standard deviation (SD) of blood glucose (BG) values and by mean BG daily delta change (daily max - daily minimum).Results: 276 medical and surgical patients (mean age: 51±18 yr), 19% with history of diabetes (DM), and 74% with ICU admission were treated with TPN. During TPN mean daily BG was 142.9±33 mg/dl, frequency of hypoglycemia <70 mg/dl and <40 mg/dl was 41% and 3%, respectively, and hospital mortality was 27.2%. The mean GV by SD: 38±21 mg/dl and by mean delta change: 58±34 mg/dl. GV was significantly higher in deceased patients (SD: 48±25 vs. 34±18 mg/dL and Δ change: 75±39 vs. 51±29 mg/dl, both p<0.01) than non-deceased patients. Multivariate analysis adjusted for age, DM status, gender, APACHE score, mean daily glucose and hypoglycemia revealed that GV was an independent predictor of hospital mortality (p<0.05). The association between GV and mortality was limited to patients without a history of DM and was not present in patients with DM.Conclusion: High GV is associated with increased hospital mortality independent of the presence and severity of hyperglycemia or hypoglycemia during TPN therapy. Prospective randomized trials are needed to determine if reduction in GV with intensive glycemic control improves clinical outcomes in patients treated with TPN.
    Full-text · Article · Sep 2013 · Endocrine Practice
  • [Show abstract] [Hide abstract]
    ABSTRACT: To determine differences in inpatient glycemic control and response to two different glargine-based insulin regimens in general medicine and surgery patients with type 2 diabetes (T2D). This is a post-hoc analysis of a prospective, multicenter, randomized trial of 298 non-ICU medicine and surgery patients with T2D treated with Basal Bolus regimen with glargine once daily and glulisine before meals and with Basal Plus regimen with glargine once daily and supplemental doses of glulisine before meals for blood glucose (BG)>140mg/dl. Major study outcomes included differences in mean daily BG, frequency of treatment failures (defined as >2 consecutive BG>240mg/dl or a mean daily BG>240mg/dl), and hypoglycemia between the medicine and surgery cohorts. Patients treated with Basal Bolus or with Basal Plus experienced similar improvement in mean daily BG after 1st day of therapy (p=0.16), number of treatment failures (p=0.11) and hypoglycemic events (p=0.50). Compared to surgery patients (n=130), medicine patients (n=168) had higher admission BG (p=0.01) and HbA1c levels (p<0.01); however, they had similar response to either treatment regimen without differences in mean daily BG after 1st day of therapy (p=0.18), number of treatment failures (p=0.58), daily insulin requirements (p=0.36), or in the frequency of hypoglycemia (p=0.79). The Basal Plus regimen with glargine once daily and correction doses with glulisine before meals resulted in similar glycemic control to basal bolus regimen. We observed no differences in response to either basal insulin regimen between medicine and surgery patients with type 2 diabetes.
    No preview · Article · Aug 2013 · Journal of diabetes and its complications
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: OBJECTIVE This study investigated the safety and efficacy of sitagliptin (Januvia) for the inpatient management of type 2 diabetes (T2D) in general medicine and surgery patients.RESEARCH DESIGN AND METHODS In this pilot, multicenter, open-label, randomized study, patients (n = 90) with a known history of T2D treated with diet, oral antidiabetic agents, or low total daily dose of insulin (≤0.4 units/kg/day) were randomized to receive sitagliptin alone or in combination with glargine insulin (glargine) or to a basal bolus insulin regimen (glargine and lispro) plus supplemental (correction) doses of lispro. Major study outcomes included differences in daily blood glucose (BG), frequency of treatment failures (defined as three or more consecutive BG >240 mg/dL or a mean daily BG >240 mg/dL), and hypoglycemia between groups.RESULTSGlycemic control improved similarly in all treatment groups. There were no differences in the mean daily BG after the 1st day of treatment (P = 0.23), number of readings within a BG target of 70 and 140 mg/dL (P = 0.53), number of BG readings >200 mg/dL (P = 0.23), and number of treatment failures (P > 0.99). The total daily insulin dose and number of insulin injections were significantly less in the sitagliptin groups compared with the basal bolus group (both P < 0.001). There were no differences in length of hospital stay (P = 0.78) or in the number of hypoglycemic events between groups (P = 0.86)CONCLUSIONS Results of this pilot indicate that treatment with sitagliptin alone or in combination with basal insulin is safe and effective for the management of hyperglycemia in general medicine and surgery patients with T2D.
    Full-text · Article · Jul 2013 · Diabetes care
  • Ying Guo · Ruosha Li · Limin Peng · Amita K Manatunga
    [Show abstract] [Hide abstract]
    ABSTRACT: The need to assess agreement arises in many scenarios in biomedical sciences when measurements were taken by different methods on the same subjects. When the endpoints are survival outcomes, the study of agreement becomes more challenging given the special characteristics of time-to-event data. In this article, we propose a new framework for assessing agreement based on survival processes that can be viewed as a natural representation of time-to-event outcomes. Our new agreement measure is formulated as the chance-corrected concordance between survival processes. It provides a new perspective for studying the relationship between correlated survival outcomes and offers an appealing interpretation as the agreement between survival times on the absolute distance scale. We provide a multivariate extension of the proposed agreement measure for multiple methods. Furthermore, the new framework enables a natural extension to evaluate time-dependent agreement structure. We develop nonparametric estimation of the proposed new agreement measures. Our estimators are shown to be strongly consistent and asymptotically normal. We evaluate the performance of the proposed estimators through simulation studies and then illustrate the methods using a prostate cancer data example.
    No preview · Article · Jul 2013 · Biometrics
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: OBJECTIVE Effective and easily implemented insulin regimens are needed to facilitate hospital glycemic control in general medical and surgical patients with type 2 diabetes (T2D).RESEARCH DESIGN AND METHODS This multicenter trial randomized 375 patients with T2D treated with diet, oral antidiabetic agents, or low-dose insulin (≤0.4 units/kg/day) to receive a basal bolus regimen with glargine once daily and glulisine before meals, a basal plus regimen with glargine once daily and supplemental doses of glulisine, and sliding scale regular insulin (SSI).RESULTSImprovement in mean daily blood glucose (BG) after the first day of therapy was similar between basal bolus and basal plus groups (P = 0.16), and both regimens resulted in a lower mean daily BG than did SSI (P = 0.04). In addition, treatment with basal bolus and basal plus regimens resulted in less treatment failure (defined as >2 consecutive BG >240 mg/dL or a mean daily BG >240 mg/dL) than did treatment with SSI (0 vs. 2 vs. 19%, respectively; P < 0.001). A BG <70 mg/dL occurred in 16% of patients in the basal bolus group, 13% in the basal plus group, and 3% in the SSI group (P = 0.02). There was no difference among the groups in the frequency of severe hypoglycemia (<40 mg/dL; P = 0.76).CONCLUSIONS The use of a basal plus regimen with glargine once daily plus corrective doses with glulisine insulin before meals resulted in glycemic control similar to a standard basal bolus regimen. The basal plus approach is an effective alternative to the use of a basal bolus regimen in general medical and surgical patients with T2D.
    Full-text · Article · Feb 2013 · Diabetes care

Publication Stats

1k Citations
205.52 Total Impact Points

Institutions

  • 2007-2015
    • Emory University
      • • School of Medicine
      • • Department of Biostatistics and Bioinformatics
      Atlanta, Georgia, United States
    • University of Wisconsin–Madison
      • Department of Biostatistics and Medical Informatics
      Madison, Wisconsin, United States
  • 2006
    • UK Department of Health
      Londinium, England, United Kingdom