tag:blogger.com,1999:blog-2395459718714693260.comments2024-03-21T15:02:06.271+00:00Electronic Data Capture - Technology BlogUnknownnoreply@blogger.comBlogger57125tag:blogger.com,1999:blog-2395459718714693260.post-61714396479724003142024-03-21T15:02:06.271+00:002024-03-21T15:02:06.271+00:00Just getting back to you on this comment. 15 year...Just getting back to you on this comment. 15 years may be setting a record, but the feedback was valid and deserves a response. <br /><br />Yes, 6 different topics if not more, but, to get to the 6, you need to start at a top level with enough detail to raise interests. <br /><br />Regarding the age question - ideally, when a piece of logic / trigger is defined, the qualification of what is stated is as specific and non specific as you need. <br /><br />This creates challenges though with triggers. Triggers tend to be 'headless' in that they are not really in the loop of context - so, the trigger engine doesn't know if the age is in any particular group, form or visit. So - that headless mode doesn't really work as we want. Imagine if we had age in 2 different places for example. <br /><br />Instead, we need to think of the execution of triggers as being instead contextual. When a person leaves a field, the reference to the field in the trigger is recognized and the context is compared. If no group, form or field are specified, then the logic engine should assume the context is limited to the triggering operation. This means it qualifies the age with [event].[form].[group].age.[value]. It is almost as if the fully qualified syntax is [current_event].[current_form].[current_group].age.. <br /><br />Now, lets imagine we do not wish to refer to the current, but maybe specifically to a field in a location. lets imagine we want to refer to a date in a particular visit;<br /><br />.*.*.VisitDT.<br /><br />This would say I want to refer to the VISITDT in Visit 1 on any form or group. <br />if I wanted to refer to a specific form, then I would need an alternative syntax to just blank - maybe a ?. <br /><br />By the way, this syntax, or something similar to this was used in DBL Recorder / eDM.Doug Bainhttps://www.blogger.com/profile/08553118142081822231noreply@blogger.comtag:blogger.com,1999:blog-2395459718714693260.post-36668079441206981302010-11-18T00:05:42.005+00:002010-11-18T00:05:42.005+00:00John,
I am not disagreeing with you here. Any iP...John,<br /><br />I am not disagreeing with you here. Any iPad/iPhone app that successfully captures clinical trial data should operate online and offline. When a session goes online, it should intelligently deal with the synchronization of data and metadata. That part is not easy especially if you have offline changes to the metadata and duplication update collisions.<br /><br />Regarding the limitations of Apps - I think these are being removed as the App platform matures. I am not an iP* App pro, but I suspect the platform available today could host both a traditional EDC tool as well as a patient eDiary.EDC Consultanthttps://www.blogger.com/profile/03142353430233190017noreply@blogger.comtag:blogger.com,1999:blog-2395459718714693260.post-10186272015978322812010-11-17T23:55:36.926+00:002010-11-17T23:55:36.926+00:00I disagree that app deployment need be too restric...I disagree that app deployment need be too restrictive versus a browser based deployment. <br />The best implementation of the IPAD in EDC should not depend on uninterrupted WIFI or 3G connectivity to operate; lacking connectivity to operate, a native app is necessary. Yet, once a captured data file is uploaded, there must be a defined process for purging the mobile device. <br />We have implemented it(& more)and it is successfully in place today.Johnhttps://www.blogger.com/profile/15076197529707553907noreply@blogger.comtag:blogger.com,1999:blog-2395459718714693260.post-61019988884148466342010-11-15T11:45:46.600+00:002010-11-15T11:45:46.600+00:00Hi eGirl,
Apologies for not replying earlier.
T...Hi eGirl,<br /><br />Apologies for not replying earlier. <br /><br />This may be too late, but here is my perspective based on past expereince here. <br /><br />Data entry is often carried out by study nurses from Source documents. This is common. What is slightly less common is that these source documents are transferred to another site, and then keyed. I can see how this might make your sponsor nervous. <br /><br />However, ultimately, the Investigator should be signing off on the eCRF - your chosen EDC system should support Investigator signature. <br /><br />What I would suggest is that the system is configured to ensure each page is signed by the investigator in the EDC system. This will help assure that the values are correct.<br /><br />I would very strongly advise attempting to do Double Data Entry with an EDC system that does not support it. It cannot do it. It will be highly unsatisfactory for everyone involved. <br /><br />Contact me privately at the email address at the foot of the blog page and I can provide further advise on how you might resolve issues here.EDC Consultanthttps://www.blogger.com/profile/03142353430233190017noreply@blogger.comtag:blogger.com,1999:blog-2395459718714693260.post-82095598339953020832010-10-14T08:41:10.236+01:002010-10-14T08:41:10.236+01:00I have a study where the primary site will be the ...I have a study where the primary site will be the 'data-entry hub' for all nearby smaller satellite sites. The problem is, the EDC system doesnt allow for double data entry, but the sponsor is insisting on this. <br />From one point I understand as this central site data entry person is entering from someone else's source notes, but on the other hand EDC has done away with DDE.<br />What is your experience with this situation?<br />Do you have any documentation I can use to convince the sponsor ?e-Girlhttps://www.blogger.com/profile/14912628364730201898noreply@blogger.comtag:blogger.com,1999:blog-2395459718714693260.post-58112896368821620512010-09-09T09:51:05.197+01:002010-09-09T09:51:05.197+01:00This comment has been hidden from the blog.EDC Consultanthttps://www.blogger.com/profile/03142353430233190017noreply@blogger.comtag:blogger.com,1999:blog-2395459718714693260.post-18932763813945013272010-09-08T14:26:54.305+01:002010-09-08T14:26:54.305+01:00This comment has been hidden from the blog.Unknownhttps://www.blogger.com/profile/11949035837974812617noreply@blogger.comtag:blogger.com,1999:blog-2395459718714693260.post-81727195211000895402010-04-05T15:24:09.969+01:002010-04-05T15:24:09.969+01:00You mentioned me (XML4Pharma), so here I am!
The C...You mentioned me (XML4Pharma), so here I am!<br />The CDISC ODM/define.xml team was already working on such an extension three years ago, but was stopped by the decision of the FDA that SAS Transport (the current transport format for SDTM datasets) would be replaced by HL7-XML, and not by an ODM-extension! At that time, at least 3 vendors had already an intermediate format that is an ODM-extension to exactly carry SDTM data (only in the last step, their XML is "downgraded" into SAS Transport), so it would only be a matter of aligning these 3 “dialects”, and add some additional extension elements/attributes in cooperation with the FDA.<br />So at that time, the FDA “killed” our project.<br />The recent decision of the FDA to "put the HL7-XML on ice" (see http://www.cdisc.org/fda-cber-cder) and the recent comment that "We may need to go back to the ODM piloting that we were doing" (see http://www.cdisc.org/fda-town-hall) gives us a window of opportunity to take this work up again and try to convince both CDISC and the FDA that an ODM extension for carrying SDTM data is the right way to go.<br />This would also perfectly fit with define.xml, which is the ODM-extension for carrying the metadata of the submission. <br />It would also allow us to define the SDTM rules in a machine-readable format (e.g. using Schematron).<br />As you say, the step going from operational clinical data (e.g. as ODM) to SDTM is not an easy step. This is however only partially due to the technology (SAS Transport is 30 years old and makes things unnecessary difficult), but mostly due to the fact that going to SDTM is an <b>interpretation</b> and <b>categorization</b> step. Tools like SDTM-ETL(R) help a lot, but one still needs to understand the clinical data as well as have (a lot of) experience with SDTM.<br />Your suggestion to allow a “linking” in the SDTM-ODM to the original ODM data is a very good one (which we will pick up), but this presumes that the FDA reviewers have tools to inspect ODM files including the audit trails, which they currently do not have at all.<br />I am thinking about submitting an abstract for the automn CDISC Interchange in Washington, so that I can give a presentation on this topic, with the goal that we can (re)start the discussion within CDISC and with the FDA. <br />It's only so unfortunate that we lost three years due to some stupid (political?) decisions within the FDA.XML4Pharmahttps://www.blogger.com/profile/05523910878300565120noreply@blogger.comtag:blogger.com,1999:blog-2395459718714693260.post-47688577612798279302010-04-01T10:48:11.565+01:002010-04-01T10:48:11.565+01:00Many thanks for your comments.
Could I ask for cl...Many thanks for your comments.<br /><br />Could I ask for clarification on your point that data entered at time of availability is not 'eSource'. If a Doctor takes notes during a assessment of a patient, and then records these notes into an electronic Patient Record system, is the Doctor required to file the paper notes in a binder for the patient?<br /><br />If the Doctor - instead - uses an iPad to take notes, and then transposes these notes into the EPR, is that not eSource then?<br /><br />The Practice of Medicine is the most critical aspect - above and beyond the study of drugs. However, provided the practice of medicine is not compromised and the data can be provided more rapidly, accurately and effectively for the study of drugs, is this not a better overall sitation for the patients in general? Are the GCP guidelines, and SOP instructions really doing their job in helping patients?Doug Bainhttps://www.blogger.com/profile/08553118142081822231noreply@blogger.comtag:blogger.com,1999:blog-2395459718714693260.post-21819062266357519812010-03-25T16:17:36.346+00:002010-03-25T16:17:36.346+00:00I too have been intrigued since the release of the...I too have been intrigued since the release of the iPad in this arena, however it still doesn't answer the question of if an eCRF can be considered source? I can understand how it will allow the site to capture the eCRF data at time of subject visit, however its still not eSource. GCP's and SOP's still drive the "practice of medicine"; the protocol and the data being collected through the CRF is still a byproduct of source collected for the visit. The iPad may be a unique tool that gets us closer but i still feel that we are missing a lot before we classify it as eSource.eSource Consultanthttps://www.blogger.com/profile/17515161131656609364noreply@blogger.comtag:blogger.com,1999:blog-2395459718714693260.post-50103634908958881312010-02-15T14:10:20.557+00:002010-02-15T14:10:20.557+00:00I'm in general agreement with you. There must ...I'm in general agreement with you. There must be a way to execute immediate edit checks in batch for situations where you create a new edit check or want to modify an edit check and re-run it across all the data.<br /><br />I think we sometimes have a situation where the data is collected and checked in EDC with online validations. The data is then exported as SAS and the SAS programmer creates all kinds of new validations (which by SAS's nature are batch). Some of those might be clever statistical sampling checks, difficult to write in an EDC system but I suspect most are not.<br /><br />Maybe we can convince Data Management to trust the checks but we'll probably have to beat this habit out of the SAS programmers. They have SAS and they are going to use it!Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2395459718714693260.post-69164046883126223952010-02-06T08:34:18.487+00:002010-02-06T08:34:18.487+00:00This comment has been removed by a blog administrator.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2395459718714693260.post-35722089562874818442010-01-17T11:11:42.521+00:002010-01-17T11:11:42.521+00:00This comment has been removed by a blog administrator.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2395459718714693260.post-77860144810972114812009-10-31T15:51:25.271+00:002009-10-31T15:51:25.271+00:00Personally (so NOT as one of the developers of the...Personally (so NOT as one of the developers of the CDISC ODM standard) I agree with you.<br />Like always in standards development, the problem is to find consensus.<br />For example, my favorite choice for the rules is to use XPath and XQuery, but this will probably be totally unacceptable for most of the vendors that use relational databases.<br />I remember to have looked into RuleML some time ago, and I will certainly revisit it. For those wanting to go with RuleML, they can easily already do so by using the ODM extension mechanism. I think that mechanism is a very good way of trying something out - when e.g. RuleML is successfully used by several vendors, it may thus become "defacto" or "good practice" which means that the ODM team will surely consider bringing it into the formal standard.<br /><br />By the way, the ODM team always is looking for new volunteers that already know the standard well, like you - so why not joining us?XML4Pharmahttps://www.blogger.com/profile/05523910878300565120noreply@blogger.comtag:blogger.com,1999:blog-2395459718714693260.post-17079878517775398922009-10-21T09:05:34.913+01:002009-10-21T09:05:34.913+01:00A good point X. Something like PEDC (Paper to Ele...A good point X. Something like PEDC (Paper to Electronic Data Capture) However, I don't see many vendors volunteering to call them themselves something considered less than EDC. <br /><br />Also, the decision to go paper first is often either a sponsor decision or a site decision. I recall as instance were I was bemoaning the lack of source for EDC for a particular study. It was explained to me that as the study drug was for treating psychosis, and the subjects were interviewed in a locked room, the presence of a laptop being used to record responses might create an aggressive response. [Could we consider throwing of laptop across interview room as an adverse reaction?..]<br /><br />ClinicalInk is an interesting approach. Oracle followed the 'like paper' way using PDF's, but failed to realise the heavy client, and lack of interactivity as a boundary. Provided you are able to make the technology pervasive and can ensure the experience is 'Paper+' then you might make some headway here for particular types of studies. Being hardware based, I am not sure how far it might go though. The old school RDE systems from the 90's failed due in part to the need to manage the physical distribution of hardware. <br /><br />Maybe as a bolt on to a main stream pure web based EDC product, it would work. Good luck!Doug Bainhttps://www.blogger.com/profile/08553118142081822231noreply@blogger.comtag:blogger.com,1999:blog-2395459718714693260.post-44973746959093651332009-10-19T18:26:37.680+01:002009-10-19T18:26:37.680+01:00I think we should stop calling data entry that was...I think we should stop calling data entry that was first captured on paper and then typed over into a computer "EDC".<br />If something has been put on paper first, I do not consider it EDC anymore. Or should we use another word for those applications where the investigator directly enters the data using any electronic device (such as tablet, PDA, e-paper,...)?XML4Pharmahttps://www.blogger.com/profile/05523910878300565120noreply@blogger.comtag:blogger.com,1999:blog-2395459718714693260.post-69107247890687213752009-10-12T15:52:21.270+01:002009-10-12T15:52:21.270+01:00I completely agree. EDC has shifted data entry re...I completely agree. EDC has shifted data entry responsibility to clinical site staff, but in the majority of cases still requires true data capture to be done on paper source. That means that we ask sites to perform double data entry (once on paper and once on a keyboard) and to do that within 48 hours of the patient visit. No wonder sites are challenged to meet those requirements. <br /><br />Traditional EDC applications can be used to capture trial data as e-Source but I argue that using them does not fit the business process used at clinical sites. Transferring those EDC screens to a tablet device carried to the patient visit would not make them any more efficient because the are really focused the sponsors needs for EDC not e-Source and the needs of site personnel. Using e-Source documents which compliment clinical site business processes requires different thinking and different technology to be successful. I am associated with a company (www.clinicalink.com) which hopes to capitalize on this route. We invite you to view a short demo on our web site; all comments are welcome.EDC Wizardhttps://www.blogger.com/profile/13553100580106198201noreply@blogger.comtag:blogger.com,1999:blog-2395459718714693260.post-44612017608343140242009-09-07T06:03:35.810+01:002009-09-07T06:03:35.810+01:00This comment has been removed by a blog administrator.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2395459718714693260.post-71216548207261334442009-07-23T13:19:21.839+01:002009-07-23T13:19:21.839+01:00Very interesting article. I agree entirely. We'...Very interesting article. I agree entirely. We're seeing an increase in the need for data on real world patients for a range of purposes from safety to outcomes measurement. This isn't always easy to get through non-interventional studies, especially in primary care, and we've been talking to a couple of clients about rolling out these sort of programmes - as you say, it must be a win/ win.Dominic Farmerhttps://www.blogger.com/profile/06713033028783670433noreply@blogger.comtag:blogger.com,1999:blog-2395459718714693260.post-59376273444082153992009-07-23T00:48:19.723+01:002009-07-23T00:48:19.723+01:00Just one of these spam blocking mechanisms that cl...Just one of these spam blocking mechanisms that clearly is not working as it should. <br /><br />I can be reached via edcconsultant0 at gmail.comDoug Bainhttps://www.blogger.com/profile/08553118142081822231noreply@blogger.comtag:blogger.com,1999:blog-2395459718714693260.post-50160108480914580292009-07-22T21:25:51.856+01:002009-07-22T21:25:51.856+01:00I did not see a valid email address at the end of ...I did not see a valid email address at the end of the blog. All it says is;<br />"Private Email<br />hidden-email.com"<br />I would appreciate it if you could include a email address in your reply.Vichttps://www.blogger.com/profile/06403874798048101607noreply@blogger.comtag:blogger.com,1999:blog-2395459718714693260.post-83659093214178142009-07-21T21:26:51.848+01:002009-07-21T21:26:51.848+01:00I can be reached at the email address posted at th...I can be reached at the email address posted at the foot of the blog.Doug Bainhttps://www.blogger.com/profile/08553118142081822231noreply@blogger.comtag:blogger.com,1999:blog-2395459718714693260.post-28098978634247789792009-07-21T19:11:27.874+01:002009-07-21T19:11:27.874+01:00Hi, I would like to get in touch with you, please ...Hi, I would like to get in touch with you, please tell me the best way to reach you. Thanks - VicVichttps://www.blogger.com/profile/06403874798048101607noreply@blogger.comtag:blogger.com,1999:blog-2395459718714693260.post-73849994056267452432009-05-12T14:53:00.000+01:002009-05-12T14:53:00.000+01:00The problem I have with save-as-you-go and with fo...The problem I have with save-as-you-go and with form dynamics that occur before you have saved the data you have is that, if it requires a round-trip to the server to save or run that dynamic code, you just can't say how long it will take.<br /><br />Jakob Nielsen says usability is all about tasks that take less than 1 second to perform. You will get into a race between the app and the user as the user enters "female" and moves onto the next part of the form (or the next form). In the background the system is running code on the server to evaluate whether the "Of childbearing potential?" field should appear but the user may move off the page before the response comes back.<br /><br />It's a cool idea but I don't think you really need it.Ecohttps://www.blogger.com/profile/00870871808842593297noreply@blogger.comtag:blogger.com,1999:blog-2395459718714693260.post-57275973184748931862009-05-11T20:54:00.000+01:002009-05-11T20:54:00.000+01:00I've seen at least one EDC system with an in-built...I've seen at least one EDC system with an in-built "mailbox" concept. Notifictions that are sent to the mailbox can be routed as emails external to the system. That's useful.<br /><br />RSS or Atom feeds? Not so much. I don't think it's because the developers of EDC systems are ignorant of technologies like RSS, Atom and iCalendar but rather because there is no demand from the client base. <br /><br />By writing about these things and discussing what they are good for we may help to stimulate that demand. Let's hope.Ecohttps://www.blogger.com/profile/00870871808842593297noreply@blogger.com