I have added 200K v-safe records to my tool
This includes 130K dependents, and another 70K which had serious health conditions but did not have a "medical care" checkin
Since I am allowed to have up to a million records in my Algolia search index, I have added about 200K v-safe records to my visualization tool.
I have added the following:
About 130K dependents where Relationship = ‘Child or adolescent in my care’ (in the Registrants CSV file) which have at least one checkin.
About 40K of these dependents do not have even a single checkin in the Consolidated Checkins CSV file, which would make it hard for me to add into the timeline tool because it could cause bugs.
I have also added another ~70K records for those who never had a HEALTH_IMPACT value of ‘Get care from a doctor or other healthcare professional‘ in any of their checkins but did experience serious adverse events, including Death.
This is the query I used:
(select registrant_code from covid_meddra_sro
where pt_name in ('Myelitis transverse','Cataplexy','Narcolepsy','Guillain-Barre syndrome',
'Coagulopathy','Death','Kawasaki''s disease','Dyspnoea')
or lower(pt_name) like 'anaphylactic%'
or lower(pt_name) like '%cardi%'
or lower(pt_name) like '%convulsion%'
or lower(pt_name) like '%stroke%'
or lower(pt_name) like '%seizure%'
)
union
(select registrant_code from covid_meddra_sd
where pt_name in ('Myelitis transverse','Cataplexy','Narcolepsy','Guillain-Barre syndrome',
'Coagulopathy','Death','Kawasaki''s disease','Dyspnoea')
or lower(pt_name) like 'anaphylactic%'
or lower(pt_name) like '%cardi%'
or lower(pt_name) like '%convulsion%'
or lower(pt_name) like '%stroke%'
or lower(pt_name) like '%seizure%'
)
union
(select registrant_code from covid_meddra_hcvo
where pt_name in ('Myelitis transverse','Cataplexy','Narcolepsy','Guillain-Barre syndrome',
'Coagulopathy','Death','Kawasaki''s disease','Dyspnoea')
or lower(pt_name) like 'anaphylactic%'
or lower(pt_name) like '%cardi%'
or lower(pt_name) like '%convulsion%'
or lower(pt_name) like '%stroke%'
or lower(pt_name) like '%seizure%'
)
After doing all this, I had about 999600 records (including the original 780K from the previous version of the tool), but of these about 4K was missing some data or the other and does not show up in the final tool.
For now, there is a sum total of about 999200 records in the tool. These records are also available via my Algolia search tool.
Note that I frequently delete and reupload the index in my Algolia search tool so don’t be concerned if the total number of records sometimes shows a zero! The Algolia search tool is used for my personal research and analysis, and I provide no guarantees that it will be up and running when you try and access it.
If you have a specific need and want to “freeze” the results from the Algolia search for your own use, please get in touch with me and I will set up one for you on your own Algolia account for a discounted consulting fee.
The visualization tool is independent of the search tool, and will work even if the Algolia search shows zero records.
Excellent and well done! I'll just sync to your Reg ID#s and fetch the complete data set on my end. Congrats and thank you! I'm thinking I want to model the data sets a bit different in preparation to upload into Tableau and begin a new "mini" dashboard for this project, so that will take me a little time. Hopefully I can get the new dashboard out to the public before the next VAERS drop on Jan 7.