3 results
2248: Screening for diabetes in high-risk women: Building the data infrastructure to study postpartum diabetes screening among low-income women with gestational diabetes
- Cynthia Joan Herrick, Ben Cooper, Matthew Keller, Margaret Olsen, Graham Colditz
-
- Journal:
- Journal of Clinical and Translational Science / Volume 1 / Issue S1 / September 2017
- Published online by Cambridge University Press:
- 10 May 2018, pp. 71-72
-
- Article
-
- You have access Access
- Open access
- Export citation
-
OBJECTIVES/SPECIFIC AIMS: Women with GDM have a 7-fold higher risk of developing T2DM, and rates of GDM are higher among racial and ethnic minorities and women of lower socio-economic status. There are no data on postpartum diabetes screening after the first postpartum year or among women receiving care in FQHCs. We aim to address this gap in the literature by (1) defining the rates of follow-up screening for T2DM at 6–12 weeks and 1–3 years postpartum and (2) characterizing patient, provider, and healthcare system attributes that are associated with lack of follow-up screening for T2DM in a population of low-income women with GDM. METHODS/STUDY POPULATION: This is a retrospective cohort study of women with GDM during pregnancy receiving care in Missouri FQHCs from 2010 to 2015. Electronic health records (EHR) data from 26 FQHCs is housed in a central repository through the Missouri Primary Care Association (MPCA). This includes patient demographic, lab, and medication information as well as encounter level patient and provider data for the prenatal and postpartum period. EHR data does not include accurate delivery information, however. Pregnancies during the study time frame were identified using CPT and ICD9/10 codes. Deidentified data on individuals with a pregnancy was utilized to identify a subpopulation of “GDM candidates,” using a broad definition of glucose abnormalities as follows: ICD-9/ICD-10 codes for diabetes, medications and testing supplies used for diabetes, infant birth weight ≥4000 g or 8 lb or 13 oz, or abnormal glucose labs [defined as fasting glucose≥95, gestational glucose screen≥130, 1 h test≥130 (or ≥180 if 2 h test and 3 h test recorded on same day), 2 h test≥155, 3 h test≥140, A1C≥6, any glucose≥130, or any positive urine glucose]. This subpopulation was then linked to Medicaid administrative claims data [housed at the University of Missouri Office of Social and Economic Development Analysis (OSEDA)], providing detailed information on delivery, to further characterize patients with GDM in the time frame and provide all dates necessary to classify pregnancy and postpartum periods. RESULTS/ANTICIPATED RESULTS: From the de-identified pregnancy data set including 45,810 individuals, we identified 8008 “GDM candidates.” EHR data were linked to Medicaid claims data for these individuals from 2010 to 2015. Utilizing the enhanced data set, we are defining a pregnancy for each individual by the delivery date in the Medicaid record and an algorithm using lab and ultrasound record dates to define gestational age at delivery. This will result in a pregnancy level data set linked with individual encrypted identifiers with each record representing 1 pregnancy and postpartum period. GDM in pregnancy will be defined as having a baby with birth weight≥4000 g or 8 lb or 13 oz, ICD-9 or ICD-10 code for GDM during pregnancy or at delivery, or an oral glucose tolerance test (oGTT) 12–16 weeks before delivery with 2 or more abnormal results by Carpenter and Coustan criteria. We anticipate that our final GDM data set will include 2000–3000 individuals. We will then calculate the percentage of individuals receiving recommended screening tests at 6–12 weeks (fasting glucose or 2 h oGTT) and 1–3 years postpartum (fasting glucose, 2 h oGTT, HbA1C). We will use multivariable regression techniques to identify risk factors for lack of screening. We will be able to incorporate predictors not previously evaluated including distance from home to health center, access to public transport, specialty and training of the patient’s providers, pregnancy weight gain, postpartum appointment time of day, and number of various types of office visits. DISCUSSION/SIGNIFICANCE OF IMPACT: The creation of a linked data set of pregnancies complicated by GDM in women receiving care in FQHCs in Missouri is the first step toward better characterizing follow-up diabetes screening rates in this population and understanding patient, provider, and healthcare system variables that affect postpartum screening. The ultimate goal is to translate evidence-based patient-centered sustainable interventions into practice for low-income women with a history of GDM and improve population outcomes with the ability to track progress prospectively over time.
Acknowledgements: The authors thank Susan Wilson (MPCA), Jill Lucht, and Bhawani Mishra (OSEDA).
Scaling of Statistical and Physical Electromigration Characteristics in Cu Interconnects
- Martin Gall, Meike Hauschildt, Patrick Justison, Koneru Ramakrishna, Richard Hernandez, Matthew Herrick, Lynne Michaelson, Hisao Kawasaki
-
- Journal:
- MRS Online Proceedings Library Archive / Volume 914 / 2006
- Published online by Cambridge University Press:
- 01 February 2011, 0914-F06-01
- Print publication:
- 2006
-
- Article
- Export citation
-
Even after the successful introduction of Cu-based metallization, the electromigration (EM) failure risk has remained one of the important reliability concerns for most advanced process technologies. Ever increasing operating current densities and the introduction of low-k materials in the backend process scheme are some of the issues that threaten reliable, long-term operation at elevated temperatures. The main factors requiring attention and careful control are the activation energy related to the dominating diffusion mechanism, the resulting median lifetimes, and the lognormal standard deviation of experimentally acquired failure time distributions. Whereas the origin of the EM activation energy and the behavior of median lifetimes with continuing device scaling are relatively well understood, detailed models explaining the origin and scaling behavior of the lognormal standard deviation are scarce. The statistical behavior of EM-induced void sizes and resulting lifetime distributions appear to be explainable by geometrical variations of the void shapes and the consideration of kinetic aspects of the EM process. Using these models, expected lifetime distributions for future technology nodes can be simulated from current, experimentally obtained void size and lifetime distributions. These simulations have to include geometrical factors of the EM test structures and actual, on-chip interconnects, as well as kinetic aspects of the mass transport process, such as differences in interface diffusivity between the lines. By extrapolating the expected lifetime distributions for future technology nodes from current EM data, it is possible to predict when insertion of new process schemes, such as Cu-alloys and/or metallic coating of the Cu/passivation interface is required.
The Use of the Four-Point Bending Technique for Determining the Strength of Low K Dielectric/Barrier Interface
- Ting Tsui, Cindy Goldberg, Greg Braeckelman, Stan Filipiak, Bradley M. Ekstrom, J.J. Lee, Eric Jackson, Matthew Herrick, John Iacoponi, Jeremy Martin, David Sieloff
-
- Journal:
- MRS Online Proceedings Library Archive / Volume 612 / 2000
- Published online by Cambridge University Press:
- 17 March 2011, D1.2.1
- Print publication:
- 2000
-
- Article
- Export citation
-
One of the important reliability challenges in integrating copper/Low-K dielectric technology has been adhesion between the Low-K dielectric and barrier metal. This investigation explored the applicability of the four-point bend technique for determining the adhesion strength of a fluorine doped low dielectric constant oxide in contact with tantalum barrier layer. Time of flight secondary ion mass spectroscopy (ToFSIMS) was used for surface chemical analyses of the delaminated surfaces to identify the fractured interface and its chemical compositions. The effect of annealing on mechanical strength was coupled with chemical analysis to discern the adhesion properties. Experimental results suggested that fluorine rich interfacial layer formation was associated with degraded adhesion characteristics between Low-K dielectric and tantalum barrier metal.