epidemiological exposure data and biological samples can be systematically collected and shared rapidly in a format that can be easily aggregated, tabulated and analyzed across many different settings globally for timely estimates of 2019-nCoV infection severity and transmissibility, as well as to inform public health responses and policy decisions.
@LN and I have made the minimum reporting form into a downloadable WHO COVID-19 A0 XLSForm that anyone can use and leave feedback on. This form should be a good model for a number of other applications.
When converting a paper form into an ODK form, there are always many decisions to make about how it should look and feel. Here are some that we made in this case:
Questions are grouped on screens to minimize vertical scrolling and help the data collector focus on what is related and important
Screen breaks are designed to be comfortable on 5-inch 1920x1080 screen (e.g. Pixel 2) or 8-inch 1024x768 screen (e.g. Galaxy Tab A)
Every yes/no/unknown question is horizontal so more questions fit on a screen. They are represented consistently so that data collectors can respond from muscle memory
Select questions are always required
Logic is used to hide questions that aren’t relevant
It’s important to test a form definition and make sure it is appropriate for any new context it is used in. We have added several comments labeled “Review and customize” in the form definition to highlight changes that you should consider.
We will update the form as we get feedback from users. Watch this topic (learn how here) for updates or check the changelog sheet in the Google Sheet.
If you need help using or deploying this form, @LN, myself, and the rest of team at Nafundi can help for free. Fill out the form at https://forms.gle/xMJZPAzKQwgeMzE17.
Looks great. Very intuitive. Some quick observations:
${symtom_existence} should probably be required (oversight?)
any reason not to make ${symtom_existence} and ${onset_known} also columns-pack appearance, to match the rest of yes/no type questions in the form?
in some cases, with dates, there's a "Is x Date known?" followed by dependent mandatory date question, in other cases a dependent show/hide date its not mandatory, ... Might want to go thru all these dates carefully and confirm the required logic
I could see region(s) visited being mandatory but potentially travel may not involve specific cities per se (ie make city questions non-mandatory?).
You could probably add ". < today()" constraint on almost all date questions
Thanks so much for the thorough review, @Xiphware!
Absolutely, thank you! I did a check where I filtered all select columns and checked their required status but this must have slipped.
This is largely a matter of preference. In general, I think vertical choices are easier to scan and so I tend to favor them unless the same options are repeated over and over again. That said, what really pushed me in that direction is that there's a visual bug with the columns-pack appearance in Collect that makes the alignment look strange in that case: https://github.com/opendatakit/collect/issues/3744.
I followed the source form on that one. Some dates have an unknown options while others don't. Certainly that's something folks deploying this form should consider based on the analysis they plan to do.
That's fair.
Done. I used the enumerator-confirmed date of form completion in case the device time is off or the form contents are digitized after the form filling date.
@mathieubossaert got in touch with me about contributing French translations and we just added them. Thanks so much, @mathieubossaert! While I am a native speaker, have reviewed them carefully, and think they are excellent, the French translations have not yet been reviewed by a health professional. If you are a French-speaking health professional, we would deeply appreciate your review. In particular, neither of us is sure how to translate "baseline serum."
I see a lot of activity in the form document. If you are using this form in whole, part, or even just as inspiration, we'd really appreciate hearing about it and learning what additional support we can offer. Are there other COVID-19 forms (e.g. from WHO, CDC, etc) that would be helpful to have ODK versions of?
merci pour le model mais pour les analyses journalière je cherche le statistique covid 19 jour pour jour ou puis je trouver une base de donnée comme ca.
Est-ce que vous cherchez des données déjà collectées? Ce que nous vous proposons ici sont les formulaires pour faire la collecte de données. L'OMS publie chaque jours un rapport avec les statistiques mondiales et beaucoup de pays font la même chose pour leurs régions. John Hopkins essaye de tout mettre ensemble et publie le tout en format facile a analyser. Entre autre, ces données sont utilisées pour la carte à jours de tous les cas:
That sounds great, @paul_macharia. How about making a copy, adding the translation columns and sharing with me when it's ready? I can then copy the translations into the master document. If you have insights into how this form could be or does get used in East Africa, please share.
A large part of East, Central and Southern Africa uses Swahili. At this stage, it's just preparing anything and everything that may be needed depending on how the pandemic turns out in this part of the world.
We tested this questionnaire on our GIC Collect app on the iPhone and it works with some small changes.
We do not manage the variables or formulas in the form data/@version.
We do not manage the last-saved type variables.
In the date of birth and respondent date of birth questions on the iPhone it would be better to use the no-calendar appearance to make it easy to select the year.
Hello everyone, first I should tell you that the work is very helpful!
Also, I can comment that I have been able to download the file and consult it directly from the XLSX format but I am interested to know is there a way to open this form in build.getodk.org to see its programming? My problem is that build.getodk.org only load .odkbuild format.
While I am a native speaker, have reviewed them carefully, and think they are excellent, the French translations have not yet been reviewed by a health professional. If you are a French-speaking health professional, we would deeply appreciate your review. In particular, neither of us is sure how to translate "baseline serum."