This workaround is asking for trouble. Why can Redcap not implement a better handling of randomization or provide a way for programmers to script it. Manually creating new records and then moving data over is not a great solution for very large and automated studies.
I had completed data entry in "form-1" . Now I created another form called "form-2" where some of the data in "form-2" are the same in "form-1", may I know how I should assign the value from "form-1" to minimize the data entry time? Thx. I tried @DEFAULT but it doesn't seems to work.
Can I ask whether this Randomization module is part of the basic REDCap functionality, or whether this is an external module that would need to be downloaded?
Hi, i need help!! My project is done and i send production. I need add more field. I got to add more field with drafth. But when i clic didn't complete the process
Such a helpful tutorial! I do have two questions, one being an easier answer. :) 1. Where are your "snapshots" stored? 2. When you set up your randomization model, in a needed longitudinal study (3 different treatment groups), it was on the first visit, and the first question. If one has a study where each patient will not get randomized until the 3rd visit (so staff is only "blind" until the 3rd visit, since they would be in direct contact with all the patients and their treatments), could you still set up that model? I'm hoping it would look the same as what you created? I realize they would be re-added to the "correct" arm, you would have to remake the record, but could it be done by downloading the patient file from the "randomization" arm, then uploading and their record? I realize that a REDCap record number can't exist in more than one arm at a time, so it would require either moving the patient completely, or renaming the record? When one of those actions is done, all the previously collected data would be there (if I included the two initial visits in every arm of the study) just in a new arm, is my understanding? I'm so sorry to bother you, but I am really struggling with coming up with a plan for my team, and we have no help/guidance (lack of resources)! Extra info that could be helpful: we don't necessarily need to use REDCap for randomization. I learned everything from awesome tutorials like yours! Thank you in advance for any advice!
Hi, thanks for the comment. 1) you can view snapshot history by clicking on the timestamp of the last snapshot. these are technically stored in backend database of redcap, but im not to familiar with that. 2) You can create 3 arms. 1 arm 'pre-randomization' arm 2 = treatment. arm 3 = control. All patients would start in arm 1. then once they get randomized you would need to re-create that record in either arm 2 or arm 3 depending on treatment or control. its a little tedious, but the best options available at this time. If you turn off auto-numbering, you can maintain their record_id number between arm 1, then 2 and 3. Turn off auto-numbering in project setup menu -> disable auto numbering. Hope this helps!
@@edwardkowalewski4982 Thank you so very much for your time and help with this! It's extremely helpful! We would only be moving patients one at a time between arms, and just the first 2 visits, so minimal downloading, then uploading into their new arms (the instruments have been copied exactly the same to inhibit difficultly during this process). My only concern now is what will put stress on the web server (we don't have any real IT support, just space). I'm not sure what processes actually stress the system, but I'm thinking one or two patient downloads won't?
@@edwardkowalewski4982 Our IT department says that I'm causing lockups when I'm designing/edited instruments, and their best guess was because I'm moving fields around too quickly, or deleting/creating them to quickly. Is there something I can read on what processes you need to slow down for? I'm stumped, honestly (I'm just asking because we literally have no resources). Thank you so much in advance, and I'm sorry for all the questions!
Out of curiosity (just before 43:40) you randomised patient 128 (which was record id 7) into a trial arm... and then added that same data to that trial arm as patient 128 (now record id 8). Is there some reason you rely on the patient 128 being entered and don't just use the record id as the unique id... which in this case would be 8... and record 7 could be just be ignored or discarded?
Great question, the '128' is a hypothetical participant number generated for this example. It can be treated similar to a medical record number (MRN) for a regular patient as a way to identify that specific patient. But as you said, there is no issues with using '8' or whatever the record number is. In REDCap there is also a way to customize the record numbers to be any number you want. This is located in project setup menu -> Additional customizations -> 'Set a Custom Record Label'. This can be useful to visually see patient numbers on the record status dashboard page, instead of auto-numbered records. Hope this helps!
![Summer Walker - Heart Of A Woman [Official Lyric Video]](http://i.ytimg.com/vi/SFddctbZzw8/mqdefault.jpg)
This workaround is asking for trouble. Why can Redcap not implement a better handling of randomization or provide a way for programmers to script it. Manually creating new records and then moving data over is not a great solution for very large and automated studies.
I had completed data entry in "form-1" . Now I created another form called "form-2" where some of the data in "form-2" are the same in "form-1", may I know how I should assign the value from "form-1" to minimize the data entry time? Thx. I tried @DEFAULT but it doesn't seems to work.
Thanks for the tutorial. It doesn't seem that one can use randomization in surveys. Am I right?
Can I ask whether this Randomization module is part of the basic REDCap functionality, or whether this is an external module that would need to be downloaded?
Hi, i need help!!
My project is done and i send production. I need add more field. I got to add more field with drafth. But when i clic didn't complete the process
Such a helpful tutorial! I do have two questions, one being an easier answer. :) 1. Where are your "snapshots" stored? 2. When you set up your randomization model, in a needed longitudinal study (3 different treatment groups), it was on the first visit, and the first question. If one has a study where each patient will not get randomized until the 3rd visit (so staff is only "blind" until the 3rd visit, since they would be in direct contact with all the patients and their treatments), could you still set up that model? I'm hoping it would look the same as what you created? I realize they would be re-added to the "correct" arm, you would have to remake the record, but could it be done by downloading the patient file from the "randomization" arm, then uploading and their record? I realize that a REDCap record number can't exist in more than one arm at a time, so it would require either moving the patient completely, or renaming the record? When one of those actions is done, all the previously collected data would be there (if I included the two initial visits in every arm of the study) just in a new arm, is my understanding? I'm so sorry to bother you, but I am really struggling with coming up with a plan for my team, and we have no help/guidance (lack of resources)! Extra info that could be helpful: we don't necessarily need to use REDCap for randomization. I learned everything from awesome tutorials like yours! Thank you in advance for any advice!
Hi, thanks for the comment. 1) you can view snapshot history by clicking on the timestamp of the last snapshot. these are technically stored in backend database of redcap, but im not to familiar with that.
2) You can create 3 arms. 1 arm 'pre-randomization' arm 2 = treatment. arm 3 = control. All patients would start in arm 1. then once they get randomized you would need to re-create that record in either arm 2 or arm 3 depending on treatment or control. its a little tedious, but the best options available at this time. If you turn off auto-numbering, you can maintain their record_id number between arm 1, then 2 and 3. Turn off auto-numbering in project setup menu -> disable auto numbering.
Hope this helps!
@@edwardkowalewski4982 Thank you so very much for your time and help with this! It's extremely helpful! We would only be moving patients one at a time between arms, and just the first 2 visits, so minimal downloading, then uploading into their new arms (the instruments have been copied exactly the same to inhibit difficultly during this process). My only concern now is what will put stress on the web server (we don't have any real IT support, just space). I'm not sure what processes actually stress the system, but I'm thinking one or two patient downloads won't?
@@otherwise_unspecified nah, that wont cause any issues with the system. We only start to see problems around 10,000 patients.
@@edwardkowalewski4982 Our IT department says that I'm causing lockups when I'm designing/edited instruments, and their best guess was because I'm moving fields around too quickly, or deleting/creating them to quickly. Is there something I can read on what processes you need to slow down for? I'm stumped, honestly (I'm just asking because we literally have no resources). Thank you so much in advance, and I'm sorry for all the questions!
Out of curiosity (just before 43:40) you randomised patient 128 (which was record id 7) into a trial arm... and then added that same data to that trial arm as patient 128 (now record id 8). Is there some reason you rely on the patient 128 being entered and don't just use the record id as the unique id... which in this case would be 8... and record 7 could be just be ignored or discarded?
Great question, the '128' is a hypothetical participant number generated for this example. It can be treated similar to a medical record number (MRN) for a regular patient as a way to identify that specific patient. But as you said, there is no issues with using '8' or whatever the record number is. In REDCap there is also a way to customize the record numbers to be any number you want. This is located in project setup menu -> Additional customizations -> 'Set a Custom Record Label'. This can be useful to visually see patient numbers on the record status dashboard page, instead of auto-numbered records. Hope this helps!