Vanderbilt Research Data Management Discussion Group

Goals

To evaluate data entry and management tools that could be useful for internal clinical trials, multi-center clinical trials, non-experimental clinical research, and epidemiologic research. As the Department of Biostatistics creates a generic Clinical Trials Coordinating Center, we would like to find the right set of tools that will allow us to to acquire high-quality data in a flexible yet secure way. It is possible that we will find one tool that is best for "quick start" databases that are for internal projects (and for adminstrative databases), and another system for multi-center clinical trials that may not even involve data from Vanderbilt subjects.
Review of PhOSCo | Open Infrastructure for Outcomes | Archive.MayPole | Ruby on Rails | ActiveGrid | caBIG

June 2005: We have begun a sub-topic to discuss requirements and evaluation results for various research database tools. See ResearchDatabaseGoals.

Meetings

24Mar04
Demonstration of PercipEnz Oncore cancer clinical trials database system by David Browning; Room 569 (5th Floor Conference Room), Preston Research Building, 3:30-5:00pm
List of persons who will attend: FrankHarrell, PaulHarris, Browning, JaneyWang, LianhongTang, SorenaNadaf, DalePlummer, YuweiZhu, ColeBeck, TomCaldwell, ChangYu, JulieAnnHardin, RichardUrbano, JeffreyHorner, SvetlanaEden . Review here.
28Apr04
Demonstration of Metaclinic Database System by Thomas Caldwell; Room A3210 (GCRC Conference Room), Medical Center North, 3:30-5:00pm
List of persons who will attend: FrankHarrell, PaulHarris, RichardUrbano, JeffreyHorner, SvetlanaEden, DalePlummer, SorenaNadaf, ColeBeck, YuweiZhu, ReneeStiles, StephenDeppen . Review here.
19May04
Paul Harris will discuss various web-based database tools and methods used in GCRC projects. These methods use existing resources available at VUMC (mySQL, PHP, VUNet and E-Pass authentication). There will also be a general discussion of tools and database models; Room A3210 (GCRC Conference Room), Medical Center North, 3:30-5:00pm
List of persons who will attend: FrankHarrell, PaulHarris, SorenaNadaf, ColeBeck, JeffreyHorner. Review here.
23Aug2005
Met with Paul Harris to have another look at his tools set. It has matured in the year since we last had a demonstration and looks nice. Attendees: PaulHarris, ColeBeck, DalePlummer, JeremyStephens, CharlesDupont. Review here.

Interested persons:

E-mail addresses of all potentially interested persons:

f.harrell@vanderbilt.edu, paul.a.harris@vanderbilt.edu, david.browning@vanderbilt.edu, ed.shultz@vanderbilt.edu, trent.rosenbloom@Vanderbilt.Edu, susan.r.conner@Vanderbilt.Edu, ayumi.shintani@Vanderbilt.Edu, patricia.harrison@Vanderbilt.Edu, tebeb.gebretsadik@Vanderbilt.Edu, renee.stiles@Vanderbilt.Edu, cara.hanby@Vanderbilt.Edu, julie.a.hardin@Vanderbilt.Edu, charles.matthews@Vanderbilt.Edu, daniel.byrne@Vanderbilt.Edu, zhi.chen@Vanderbilt.Edu, hui.cai@Vanderbilt.Edu, wande.guo@Vanderbilt.Edu, stephanie.michon@Vanderbilt.Edu, jenni.blackford@vanderbilt.edu, s.nadaf@vanderbilt.edu, mary.edgerton@vanderbilt.edu, jason.moore@vanderbilt.edu, chun.li@vanderbilt.edu, yuwei.zhu@vanderbilt.edu, chang.yu@vanderbilt.edu, wang@chgr.mc.vanderbilt.edu, lynda.lane@vanderbilt.edu, robin.ginn@Vanderbilt.Edu, dale.plummer@vanderbilt.edu, jeff.horner@vanderbilt.edu, cole.beck@vanderbilt.edu, svetlana.eden@vanderbilt.edu, ted.speroff@vanderbilt.edu, huiming@phg.mc.vanderbilt.edu, l.tang@vanderbilt.edu, tom.caldwell@vanderbilt.edu, stanley.higgins@vanderbilt.edu
Notes from Ed Shultz (ed.shultz@vanderbilt.edu) 12Mar04
  • "Research database system", and "multi center clinical trials" are suggestive but not specific enough for me to know exactly what you want, but I do know that generalized research data collection needs have never really been met successfully, not that efforts haven't been made. I think looking at Percipenz is a good idea. A synopsis of other tools, and time to entry:
  • Potential Fits
    • Percipenz: oriented to Oncology trials. May be difficult to generalize beyond that. Commercial. I am not up to speed on the Cancer Center impression of this product, so I would encourage you to contact them directly.
    • ACT/DB: Pros: Generalized data collection tool, trials oriented, minimal data validation on entry, web based data collection, relatively easy to add ad hoc elements. Complete study design can be completed in an couple hours by spreadsheet level person. 5 to 10 clinical trails have been captures using this too Cons: much of the code was inherited in the form of an Microsoft Access Data base. Although it now lives in Oracle, the basic design tools are still in Access and fragile. Bottom line is "what you see is what you get", since adding new features would be very costly. We can get you a demo if you're interested. Wendy Smith at the Cancer Center was the major expert and moved to Percipenz as they feel they will offer a better path for evolutionary new features, which is probably true.
    • StarForms: Pros: Storage of categorical data, (XML format if that helps) in individual forms which are stored StarChart. web based data collection. New forms can be constructed by spreadsheet level personnel in a few hours. You can re-use specific categorical items, but you can create new ones also. Patient oriented. You can have drop down selections, but I don't think data validation is normally included. Cons: data will not reside in a relational databases. Hence, customized extraction to your analysis tools will need to be done. Data normally resides in the clinical record, so if that is not desirable, special arrangements would have to be made. Since it's forms based, the concept of clinical trials is not present. Contact Susan Conner for more information.
  • Less Likely Fits:
    • StarNotes: Primarily free text based. Probably not as useful for your task as StarForms. Contact Trent Rosenbloom for more information.
    • Quill: Pros: optimized for high speed entry with branching logic. Template driven, but very clinical in orientation. Stored in Oracle in awkward format for your purposes, but still could be used. Cons: oriented to clinical notes capture, ie hx, physical, hpi, etc. Would have to be reoriented to succeed with your project, Adding new categorical data items goes through QC process that makes it slower than you wish, typically days. You would have to invest to make this work in your environment.

-- FrankHarrell - 14 Mar 2004; Modified 20 May 2004

Topic attachments
I Attachment Action Size Date Who Comment
TheMetaclinicDatabaseSystem.htmhtm TheMetaclinicDatabaseSystem.htm manage 49.2 K 22 Apr 2004 - 14:57 TomCaldwell Revised version of Metaclinic Database System pape
Topic revision: r48 - 29 Jul 2008, JohnHarrell
 

This site is powered by FoswikiCopyright © 2013-2017 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding Vanderbilt Biostatistics Wiki? Send feedback