Caisis 4.0: Re-Designing the Data Supply Chain
Brandon,A,Smith Other; Memorial Sloan-Kettering Cancer Center; Avinash,Chan Other; Memorial Sloan-Kettering Cancer Center; Frank,Sculli MS; Memorial Sloan-Kettering Cancer Center; Paul,G,Alli MD; Memorial Sloan-Kettering Cancer Center; Jason,R,Fajardo Other; Memorial Sloan-Kettering Cancer Center; Kevin,P,Regan ; Memorial Sloan-Kettering Cancer Center; Paul,A,Fearn MD; Memorial Sloan-Kettering Cancer Center; Kinjal,Vora MD; Memorial Sloan-Kettering Cancer Center;
Content:
Caisis is a scalable, web-based information system developed to increase clinical and research productivity, improve data quality, minimize investigator bias, and facilitate research collaboration. It was built around chronological patient histories that can be used in clinic and processed by computer algorithms to generate research datasets. Clinicians and investigators from multiple institutions have contributed to the project. It has been adopted by 15 sites in four countries and is currently used to manage data for over 165,000 patients.
Technology:
Caisis was developed with SQL Server 2000, C#, and ASP.NET 2.0 (Microsoft, Redmond, Washington.) The application is freely distributed under an open-source license (http://caisis.org). Caisis is metadata-driven and easily extensible; it can be configured through XML files and a system-administration module.
Design:
We approach data collection as a supply chain problem, capturing research-quality data directly from clinical staff. Since 2002, we have developed paper and web-based forms to document clinic visits. We are also prototyping web-based operative notes for surgeons. The clinic forms are populated with previously entered data, which allows for corrections and minimizes data entry.
Results:
Populated paper forms are now the predominant documentation used in the urology and genitourinary medicine clinics at our institution. They are billing-compliant substitutes for dictation. The follow-up visit notes are widely accepted because populated data reduces the clinical workload. As some physicians are frustrated with the process of navigating to the right form, we are prototyping a automated email process to send links to the right forms directly to physician inboxes. This process has been rapidly adopted by physicians, but they still desire improvements to the web forms themselves, adding conditional logic and improving layout. Also, tracking the various forms of documentation for every clinic visit is challenging as volume increases.
Conclusion:
The workflow surrounding the web-based forms for follow-up visits and operative notes needs to be improved to be scalable and sustainable. A comprehensive document-tracking framework and improved email form process may help. New-visit documentation may be improved by outsourcing some data entry to patients through web portals they can access before their visit or in the waiting area.
