- Email: [email protected]
A microcomputer-based distributed data management system for a large cooperative study of transfusion-associated AIDS
308
Abstracts
respiratory tract. These surgical procedures continue throughout childhood at time intervals varying from weekly to every three months. The assessment procedure is also therapeutic, and the interval between these procedures varies from patient to patient. Patients are followed for 2 years. We present assessment instruments that have been developed to quantify the extent of regrowth. Data collection forms completed at each surgery include a diagram of the larynx defining 28 regions of possible involvement. Methods for calculating a rate of regrowth per month provide a well-defined endpoint for therapeutic evaluation.
The Use of a Distributed Data Entry System in a Multicenter Study L i n d a J. Rings, C h r i s t o p h e r R. Skorlinski, a n d W. M i c h a e l O ' F a l l o n Mayo Clinic, Rochester, Minnesota (P27) A distributed data entry system can be a most effective and efficient way to collect data and monitor study progress in a multicenter study. This collaborative follow-up heart defect study uses a microcomputer (IBM XT with hard disk) with custom software at each of six centers to collect data on 2200 subjects. The software'for each center is preloaded with information about its subjects and the tests required. Several screens and printed reports help the study coordinator keep track of the work remaining to be done for each subject. Other special features include extensive edit checks at time of data entry, and double entry to catch keying errors. Data are thus checked for accuracy, completeness, and careless errors soon after the subject's examination, so that mistakes can be efficiently handled before the data are sent to the coordinating center. Twice a month data are sent by floppy diskette to the coordinating center and loaded into a mainframe computer. The quantity and quality of the data can then be monitored in timely, efficient way, allowing early detection of data problems, and effective study progress monitoring.
A Microcomputer-Based Distributed Data Management System for a Large Cooperative Study of Transfusion-Associated AIDS Joyce N i l a n d - W e i s s a n d S t a n l e y P. A z e n , a n d the T r a n s f u s i o n Safety G r o u p
University of Southern California, Los Angeles, California, and other participating institutions (P28) A microcomputer-based distributed data management system in the acquisition of data collected in a large multicenter cooperative investigation of transfusion-transmitted AIDS is described. Clinical, virological, and immunological data are obtained from six clinical centers and six central laboratories and merged and processed on a mainframe computer (158,000,000 bytes). Details of the forms control, editing procedures, and quality control procedures are presented. Advantages and disadvantages of this system are discussed relative to the dynamics of the study, the magnitude and nature of the database, and the organization of the distributed centers involved. Recommendations for multicenter clinical trials are given. (Supported by Contract No. NO1-HB4-7003 of the NHLBI.)
Distributed System for Treatment Assignment and Limited Data Entry P. W i l k e n s , R. G r o s s , E. Clarke, B. Barton, G. B o w d e n , a n d C. Fiery
The TIMI Study Group, Maryland Medical Research Institute, Baltimore Ma~land (P29) Microcomputers are being used in the Clinical Centers for treatment assignment and limited data entry, in the Thrombolysis in Myocardial Infarction (TIMI) Trial. This design replaces the original design of sealed randomization mailers and several telephone calls to the Coordinating Center to report on the enrollment of each patient and patient status during the period of hospitalization. This data entry system was installed to protect the integrity of the randomization schedules, but it also provided a means of transmitting limited data to the Coordinating Center in a timely manner. This system does not replace the complete data processing system for data collection. The first system was installed in June 1986. System design and development are presented. Some preliminary data on the comparison of the accuracy of data as given by telephone interview versus data keyed in the Clinical Center and the comparison of the time required for telephone interview versus the time for electronic transmission of data are presented.
Abstracts
respiratory tract. These surgical procedures continue throughout childhood at time intervals varying from weekly to every three months. The assessment procedure is also therapeutic, and the interval between these procedures varies from patient to patient. Patients are followed for 2 years. We present assessment instruments that have been developed to quantify the extent of regrowth. Data collection forms completed at each surgery include a diagram of the larynx defining 28 regions of possible involvement. Methods for calculating a rate of regrowth per month provide a well-defined endpoint for therapeutic evaluation.
The Use of a Distributed Data Entry System in a Multicenter Study L i n d a J. Rings, C h r i s t o p h e r R. Skorlinski, a n d W. M i c h a e l O ' F a l l o n Mayo Clinic, Rochester, Minnesota (P27) A distributed data entry system can be a most effective and efficient way to collect data and monitor study progress in a multicenter study. This collaborative follow-up heart defect study uses a microcomputer (IBM XT with hard disk) with custom software at each of six centers to collect data on 2200 subjects. The software'for each center is preloaded with information about its subjects and the tests required. Several screens and printed reports help the study coordinator keep track of the work remaining to be done for each subject. Other special features include extensive edit checks at time of data entry, and double entry to catch keying errors. Data are thus checked for accuracy, completeness, and careless errors soon after the subject's examination, so that mistakes can be efficiently handled before the data are sent to the coordinating center. Twice a month data are sent by floppy diskette to the coordinating center and loaded into a mainframe computer. The quantity and quality of the data can then be monitored in timely, efficient way, allowing early detection of data problems, and effective study progress monitoring.
A Microcomputer-Based Distributed Data Management System for a Large Cooperative Study of Transfusion-Associated AIDS Joyce N i l a n d - W e i s s a n d S t a n l e y P. A z e n , a n d the T r a n s f u s i o n Safety G r o u p
University of Southern California, Los Angeles, California, and other participating institutions (P28) A microcomputer-based distributed data management system in the acquisition of data collected in a large multicenter cooperative investigation of transfusion-transmitted AIDS is described. Clinical, virological, and immunological data are obtained from six clinical centers and six central laboratories and merged and processed on a mainframe computer (158,000,000 bytes). Details of the forms control, editing procedures, and quality control procedures are presented. Advantages and disadvantages of this system are discussed relative to the dynamics of the study, the magnitude and nature of the database, and the organization of the distributed centers involved. Recommendations for multicenter clinical trials are given. (Supported by Contract No. NO1-HB4-7003 of the NHLBI.)
Distributed System for Treatment Assignment and Limited Data Entry P. W i l k e n s , R. G r o s s , E. Clarke, B. Barton, G. B o w d e n , a n d C. Fiery
The TIMI Study Group, Maryland Medical Research Institute, Baltimore Ma~land (P29) Microcomputers are being used in the Clinical Centers for treatment assignment and limited data entry, in the Thrombolysis in Myocardial Infarction (TIMI) Trial. This design replaces the original design of sealed randomization mailers and several telephone calls to the Coordinating Center to report on the enrollment of each patient and patient status during the period of hospitalization. This data entry system was installed to protect the integrity of the randomization schedules, but it also provided a means of transmitting limited data to the Coordinating Center in a timely manner. This system does not replace the complete data processing system for data collection. The first system was installed in June 1986. System design and development are presented. Some preliminary data on the comparison of the accuracy of data as given by telephone interview versus data keyed in the Clinical Center and the comparison of the time required for telephone interview versus the time for electronic transmission of data are presented.