SOLICITATION NOTICE
A -- Epidemiology of Diabetes Interventions & Complications
- Notice Date
- 12/8/2004
- Notice Type
- Solicitation Notice
- NAICS
- 518210
— Data Processing, Hosting, and Related Services
- Contracting Office
- Department of Health and Human Services, National Institutes of Health, Nat'l Institute of Diabetes, Digestive, & Kidney Diseases, 2 Democracy Plaza, Suite 700W 6707 Democracy Blvd., MSC 5455, Bethesda, MD, 20892-5455
- ZIP Code
- 20892-5455
- Solicitation Number
- N01-DK-6-2204A
- Response Due
- 1/22/2005
- Archive Date
- 2/6/2005
- Description
- The National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) is planning to negotiate an extension contract to the Epidemiology of Diabetes Intervention and Complications (EDIC), an ongoing, longitudinal follow up to the Diabetes Control and Complications Trial (DCCT). The goal of the DCCT, a multi-centered clinical trial that ran from 1982 until 1993 and involved 1441 patients with Type 1 Diabetes, was to investigate the early microvascular and neurologic complications of Type 1 Diabetes and whether intensive insulin treatment would prevent and/or ameliorate those complications. The main results of the DCCT have been reported (NEJM, September 30, 1993). In EDIC the surviving participants from the DCCT cohort are followed longitudinally to study the long-term effects of prior separation of glycemic levels, risk factors for macrovascular outcomes, and the correlation of progression of micro- and macrovascular outcomes. Ongoing EDIC assessments include: biochemical analyses of blood specimens to ascertain renal and metabolic status, electrocardiography, ultrasonography and coronary artery calcification studies by computed tomography to determine macrovascular disease; and stereo fundus photography to ascertain retinal status. The DCCT/EDIC Study Group has published a combined 90 reports and has produced an additional 150 presentations and abstracts based on the research findings (see Attachment 1). Planned future assessments include cardiac NMR imaging, blood and urine proteomics, and repeat neuropathy screening. Continued long-term follow-up of the currently active 1354 EDIC participants (see Attachment 2) will provide important scientific information on late macrovascular complications that have only recently begun to be observed in this cohort. The George Washington University Biostatistics Center has served as the Data Coordinating Center (DCC) for DCCT/EDIC since 1982. The following describes the expertise of The Biostatistics Center and its collaborators and their contributions to DCCT and to EDIC. EXPERTISE IN BIOSTATISTICS AND DIABETES EPIDEMIOLOGY The Principal Investigator of the Data Coordinating Center at the GWU Biostatistics Center since the inception of the DCCT in 1982 is John Lachin, Professor of Biostatistics and Epidemiology, and of Statistics. Dr. Lachin has authored three books on biostatistical methods, over 55 articles on statistical and clinical trials methodology, an additional 30 articles of medical research results; and is a corporate author of over 100 additional articles published by research groups in which he had participated. Since the inception of the DCCT, Ms Patricia Cleary has served as Deputy Director, and now Director, of the Coordinating Center. Dr. Lachin and/or Ms. Cleary have been an author of virtually all of the published reports from DCCT/EDIC and have each chaired writing committees for many of these reports. In addition, both Dr. Lachin and Ms. Cleary have given presentations of DCCT/EDIC results to the annual meetings of the American Diabetes Association, the International Diabetes Federation, and the International Diabetes Epidemiology Group, among others. Dr. Lachin and Ms. Cleary are recognized internationally as experts in the epidemiology of Type 1 Diabetes Mellitus. Dr. Lachin, often in collaboration with leading theoretical statisticians, has also developed statistical methods that have been employed in the design of DCCT and EDIC and the analysis of DCCT/EDIC results. These include statistical methods for the repeated interim analysis of the emerging results from the DCCT and EDIC, methods for the analysis of interval censored event-time data with time-dependent covariates, among others. In addition, the Biostatistics Center (BSC) of The George Washington University has extensive experience with diabetes trials and longitudinal epidemiologic studies. The DCCT/EDIC Study has taken a leading position in studying the role of genetics in the development of diabetes. Assessments of familial aggregation were performed during DCCT and will be repeated in EDIC in order to test whether DNA variation in a number of candidate genes is related to susceptibility to diabetes and its complications. To this end, DNA is being collected from all available parents of DCCT/EDIC participants and at least one full non-diabetic sibling. In addition, measures of diabetic complications are also being assessed on diabetic siblings of participants. The DCC is currently initiating the establishment of a repository of DNA samples from the DCCT/EDIC participants and their relatives. Mechanisms for accessing these samples by qualified researchers are also being generated. PROFESSIONAL AND TECHNICAL EXPERTISE All procedures necessary for data collection, editing, and processing; and for the maintenance of permanent study files and documentation related to the DCCT/EDIC database are implemented by the DCC staff as specified in the Internal Procedures Manual and in accordance with HIPAA regulations. The DCC professional and technical staff includes the principal and co-principal investigators, database manager, statisticians, programmers, technology and web support specialists, research assistants, contractual, financial and administrative specialists, and data entry clerks. Overall, the DCC staff has expertise in statistics, programming (SAS and FoxPro), contract and sub-recipient agreement management, clinical trials management as well as web site development and maintenance. COMPUTING FACILITY The DCC utilizes the computing infrastructure of the George Washington University Biostatistics Center. At the core of the computing infrastructure is an IBM Multiprise 3000, S/390 class enterprise server. This server acts as the central data warehouse for all data received for the DCCT/EDIC study. The server provides one of the most fault tolerant, secure and stable platforms available in the industry. Access to the computing facility is restricted to Biostatistics Center staff. All access is monitored and logged. Unauthorized access is prevented by way of account deactivation and is monitored 24 hours a day. Data stored for the DCCT/EDIC study is kept separate from other studies conducted by the Biostatistics Center. Access to the DCCT/EDIC data is restricted to the DCC staff and is password protected. Physical site security is maintained by way of electronic access key entry system and closed circuit video cameras. DATA SECURITY Data is stored on the enterprise server and is accessible to DCC staff for analysis and reporting needs. All data for the DCC is backed up on a daily basis. Dozens of backups are maintained for extended period of time so that they would reflect the state of the databases and files at various points in time. Extensive systems are in place to ensure proper backup of all information so that data loss is prevented. All desktop systems are backed up using a central backup server. Disaster recovery measures have been taken to prevent data loss in the event of a catastrophic disaster. All data and files stored on the enterprise server and various network file servers are backed up to tape and sent off-site to a remote location where they are kept in an environmentally controlled safe. In addition, designated desktop computers that house particularly important information participate in this off-site disaster recovery program. PROJECT, CONTRACTUAL AND FINANCIAL MANAGEMENT The Administrative Group of the Biostatistics Center has extensive experience handling the various contractual mechanisms necessary including: government and subcontract contracts, clinic, and consulting and service agreements. The Administrative and Programming group have also developed an automated system to generate clinic reimbursement reports and payment requests linked to receipt of central laboratory data. This system is programmed to interface with the University payment system. The DCC also coordinates the existing subcontracts and service agreements with central laboratories, reading centers and coding units needed to perform subject assessments in accordance with study protocol. FORM DESIGN AND SOFTWARE DEVELOPMENT Sixty-eight EDIC forms and Twenty-four Genetic Studies forms were designed and developed by DCC staff for data collection, adverse event verification, patient status notification, certification of technicians, and mailing status. Using SAS FSEDIT, software was developed and is maintained by DCC programming staff for data entry and data editing of EDIC forms. Procedures used to create the data entry system include: creation of fifty-three map files and range edit files, software development to integrate map files and range edit files, software development to create main menus and selection menus for new and continuing data entry, software development to produce 131 data entry screens, one data entry screen for each page of a form for all forms, software development to create fifty-three driver programs, procedures to integrate data entry screens and driver programs into FSP, and procedures to incorporate text files into the SCL application using Interactive SAS. FoxPro was used to design and develop data entry screens for Genetic Studies data collection forms. Procedures for software development using FoxPro include: creation of data dictionaries with finalized data structures and set requirement levels with logical ranges, uploading the data dictionaries to the mainframe, preparation for data entry by building an executable file that when launched will give the end user the data entry screens for keying, uploading keyed FoxPro files to the mainframe and executing the production program. In addition to FoxPro software, SAS software was used for Genetic Studies to develop programs to integrate FoxPro data dictionaries and data files keyed in FoxPro, into SAS datasets on the mainframe. The Genetic Studies Database Management System consists of 48 data entry screens. The EDIC ancillary study URO-EDIC, also keyed in FoxPro, consists of 12 data entry screens. A separate data dictionary, and separate files containing valid ranges for each variable, were generated and are maintained by DCC Database Management staff for both EDIC and Genetic Studies data collection forms. The DCC has 84 master files for the EDIC and EDIC Genetics Forms, some forms with as many as 5 versions. The number of records for each file ranges between 5 and 8,530; and the number of variables in each file ranges between 12 and 466. A list of all of the EDIC and EDIC Genetics Forms is attached to this document (see Attachment 3). In addition, summary listings of the master files for the EDIC and EDIC Genetics Forms data are attached to this document (see Attachment 4). Programs and systems have also been written and established to interface with, convert, receive, batch, check, scan and/or process data receive and process information received from the central coding and reading units. A summary list of the master files for EDIC Central Reading Unit data is attached to this document (see Attachment 5). The procedures utilized at the DCC for receiving and processing this data is summarized below: The Central Ophthalmologic Reading Unit: Fundus Photographic data is received via diskette from the Central Ophthalmologic Reading Unit, for both EDIC and EDIC Genetic Studies, and uploaded to the mainframe. SAS Software is used to process the raw data and create a Master File. SAS Software is also used to produce patient feedback reports based on Fundus Photograph and Visual Acuity data. The Central Biochemistry Laboratory: EDIC CBL Data is received via diskette and uploaded to the mainframe. SAS Software is used to process the raw CBL data and create a Master File. SAS Software is used to produce patient feedback reports based on CBL data. EDIC Genetic Studies CBL Data is received via FTP at a secured web address. SAS Software is used to process the raw CBL data and create a Master File. SAS Software is also used to produce patient feedback reports based on CBL data. The Central ECG Reading Unit: Copies of ECG tracings are received via mail from the Clinical Centers. The ECG tracings are batched by a Data Control Clerk and mailed to the ECG Reading Unit for processing. Two types of ECG data are received via diskette from the ECG reading Unit: the raw Minnesota Code data and the ECG Serial Change Comparison data. Separate SAS programs are used to process the data, and separate master files created. SAS Software is used to generate patient feedback reports based on ECG data. The Central Ultrasound Reading Unit: Carotid Ultrasound data is received via email from the Central Ultrasound Reading Unit. Raw data is processed using SAS Software and a Master File was created. SAS Software was used to generate patient feedback reports. The Central Food Frequency Recall Unit: Food Frequency Questionnaires are mailed from the Clinical Centers to the Data Coordinating Center. Questionnaires are checked, batched, and mailed to the Central Food Frequency Recall Unit for scanning and processing. Food Frequency Data was previously received via diskette and are now being received via email. SAS Software is used to process the raw data and create a Master File for Food Frequency Questionnaire data. The Central Computed Tomography Reading Unit: Data was received via EUDORA and uploaded to the mainframe. SAS Software was used to process CT Scan data and create a Master File. SAS Software was used to edit CT Scan data based on specifications received from the CT Reading Unit. Patient feedback reports based on CT Scan data were generated and mailed to the Clinical Centers for distribution to the patients. New data values were received from the CT Reading Unit based on adjustments to existing data. SAS Software was used to incorporate the new data values into the Master File for CT Scan data. Invoices were generated for payment of CT Scans along with several reports tracking payments for CT Scans. All software used for data entry and data editing for EDIC and Genetic Studies was designed and developed internally, by the DCC EDIC Database Management staff and are not available commercially. DATA MANAGEMENT Over the past 20 years, the DCC has developed an extensive and well-established central data entry system and quality control procedures for receiving, checking, entering and verifying data to be entered into its central data entry system from the 28 EDIC clinical centers. Forms are mailed from each of the 28 clinic sites in monthly mailings. The procedures for receiving and processing the data from these forms are summarized below: Each of the 28 EDIC clinic centers mail all completed forms once a month on the Thursday of the week designated for each clinic. When forms arrive at the Data Coordinating Center, the number and types of forms received is compared to the number and types of forms sent by the Study Coordinator as listed on EDIC Form 101, Forms Mailing List, and EDIC Form 100, Clinic Forms Inventory. A trace is initiated if there are any discrepancies, and the Study Coordinator is contacted so that copies of any missing forms may be mailed with the next month?s batch. Data entry specialists then key the data using SAS programs. There is a separate FSEDIT data entry program for each EDIC Form. FoxPro Software is used to perform data entry and initial editing of the EDIC Genetics Studies data. Keyed data is verified by the execution of another SAS program. Keyed batch files are transferred from each data entry PC to the mainframe for further processing on a weekly basis. EDIC Genetics Studies data is uploaded to the mainframe from each PC on a bi-weekly basis. A SAS program is executed to produce a listing of keyed batches in preparation for the weekly production run. The report lists the number of Forms in each batch, the date that the batch file was transmitted to the mainframe, and the date that the batch file was processed. The production program is executed on a weekly basis. This program adds keyed batch files for each EDIC Form to the corresponding Master File. The DCC maintains a separate Master File for each EDIC Form. SAS Software is used to convert the FoxPro data into a SAS database Master File for each Genetics Studies form, separately. A separate edit program that searches for missing data, inconsistencies, and values out of range edits each form. If any are found, an error notice (which lists all the errors detected for a given form) is generated and mailed to the originating Clinical Center for recording the correct information. Corrected notices are returned to the Data Coordinating Center in the next month?s mailing. A SAS FSEDIT program is used to key data received via EDIC Edit Message Reports. Batches containing keyed EDIC Edit Message Report data are transferred from each PC hard-drive to the mainframe for further processing. Another SAS program is run to add the Edit Message data to a transaction file. Finally, another SAS program is executed to apply the edits to the appropriate Master File. The DCC generates clinical and feedback reports for each participant. The purpose of the Patient Clinical Report/Feedback Report is to provide the patients and their physicians with the centrally determined results of his/her examination each year. The EDIC Patient Clinical/Feedback Report that the patient receives includes exam results for all DCCT and EDIC years. The report provides the following results for the examinations: HbA1c Measurement (%) Glomerular Filtration Rate (Creatinine Clearance Lothalamate) Serum Creatinine and Urine Albumin Excretion Lipids (Cholesterol, Triglycerides, HDL, and LDL) Retinopathy Level as measured with the ETDRS grading system Visual Acuity (Snellen units) Macular Edema Status Intraocular Pressure (mmHg) Ankle/Arm Doppler ECG Yearly gifts and holiday cards are also circulated from the DCC to the EDIC clinics to be distributed to the participants. The DCC also generates reports for the clinics and various monitoring committees associated with the study. These include: a) one annual report for clinical centers; b) one quarterly report for clinical centers; c) nine reports for the Data Quality Assurance Subcommittee; d) eight quarterly reports for the Adherence Monitoring Subcommittee; e) six monthly reports for the Genetic Studies Subcommittee; f) 15 monthly reports for the CT Scan Subcommittee; and g) ten annual reports for the External Advisory Committee. A summary list of these reports and their content is attached to this document (see Attachment 6). STATISTICAL ANALYSIS - The statistical staff of the DCC has extensive experience with the various types of data that are recorded and analyzed in studies of Type 1 diabetes, and with the statistical methods that are employed. In some cases, these are highly specialized. These include the analysis of ordinal longitudinal retinopathy data using multivariate rank tests, the analysis of coronary calcification measures with a lower limit of quantification using Tobit regression models, and the analysis of interval censored event time data, among others. The Biostatistics Center has numerous software packages and products available to its statisticians and programmers (SAS and related SAS products, S-Plus, StatXact, LogXact, NQuery and various others) for use in conducting analyses and running periodic, ad hoc and interim study reports. WEBSITE MANAGEMENT The DCC maintains and updates the secured, password protected EDIC website where current versions of EDIC and EDIC Genetics forms, manuals of operations and protocols, and directories can be found. Meeting materials and minutes, as well as information regarding EDIC and its ancillary studies can also be downloaded from this site. Recently, the DCC has begun to incorporate interactive web forms that allow the clinical centers and central units to update certain EDIC Genetics Studies data remotely. Future plans to accommodate evolving data management methods for the DCC includes moving towards a paperless, web-based data entry system where the clinics can conduct the yearly visits with patients and the data would be immediately available to the coordinating center. COLLABORATIONS One important function of the DCC is to facilitate scientific collaborations utilizing this unique cohort. Extensive knowledge of the DCCT/EDIC database allows the DCC to function as an investigative partner. Currently, the following institutions are collaborators with DCCT/EDIC Study Group on ongoing research projects: The Medical University of South Carolina is studying candidate genes for vascular disease complications in the DCCT/EDIC and determining associations of a detailed serum lipoprotein profile with microvascular disease status by performing a cross sectional study of subjects from DCCT/EDIC. The University of Washington's School of Medicine is evaluating specific common polymorphisms in candidate genes related to insulin resistance and central adiposity. The University of California, San Francisco is studying the natural history of Urologic complications in adults with Type 1 Diabetes. The Joslin Diabetes Center and University of Pittsburgh are working jointly with the DCC in evaluating the effects of chronic hyperglycemia and episodic severe hypoglycemia on cognitive function. Finally, the University of Toronto is assessing familial aggregation of DNA variation in candidate genes within the cohort and their diabetic and non-diabetic family members in order to assess associations with diabetic complications. The DCC staff is also making preparations for impending analyses utilizing the extensive DCCT/EDIC data, including: retinopathy, visual acuity, albumin:creatinine ratio, carotid stenosis, urologic complications, obesity and overweight, cardiovascular and sub-clinical events, quality of life and SCL 90, dietary data, family clustering of diabetes, heart rate, and mortality. The Data Coordinating Center contract was competitively awarded to The George Washington University on February 8, 1982 through March 7, 1987, and extended non-competitively three times through February 2007.Data analysis and reporting in DCCT/EDIC is ongoing. Determination to use other than full and open competition is made in accordance with 41 U.S.C. 253 (C)(1). See Note 22. Interested parties may submit.
- Place of Performance
- Address: The George Washington University, Washington, DC
- Zip Code: 20052
- Country: USA
- Zip Code: 20052
- Record
- SN00718448-W 20041210/041208211709 (fbodaily.com)
- Source
-
FedBizOpps.gov Link to This Notice
(may not be valid after Archive Date)
| FSG Index | This Issue's Index | Today's FBO Daily Index Page |