[aprssig] Citizen Health Observer Program (CHOP)

Steve Dimse steve at dimse.com
Mon Jun 22 19:19:54 EDT 2009


Very interesting idea. I'm not sure CWOP should be your model for data  
flow, nor that the APRS IS should be the transport medium.

When we created CWOP, we were creating a system to get time-sensitive  
data to the NWS, so we leveraged the real-time capabilities of APRS.  
We also had the existing amateur radio weather stations as an initial  
test group and a pool of easy-to-recruit participants. The key  
difference between your project and CWOP is the latter sends dozens of  
daily, automated reports from each user. APRS was perfect for the CWOP  
transport medium.

Your project is more like a much older citizen science project, in  
fact the first large scale citizen project ever mounted, the  
Cooperative Observer Program created in 1890. These are stations that  
take observations once a day; some COOP sites have continuous data  
stretching back more than a century. Initially these were forwarded to  
the NWS as monthly written logs. The touch-tone phone led to an  
automated phone system, which was replaced with a web-based system  
only two years ago.

I think COOP is a much better model for your data collection. By using  
a web based page, you do not limit your participants to hams, or  
Windows computers if you write a Windows client, or anything else. All  
you need is web access, and that is close to universal. People go to a  
form page (user name and password saved by cookie to increase  
convenience). Use pop-ups or text fields to enter the data, click  
send. Data goes straight into a database. COOP has a confirmation  
field where you must review the data for correctness, you may not need  
to do this. There is also automated validity checking, for example if  
your minimum is higher than your current temp you get an error message.

If you still want to use APRS, adding support to the clients is not  
easy. Every client author would have to add a new menu item and dialog  
to collect the data. You could have the user send an APRS message to  
some call with a specified format, but each user would need to  
manually format each message, asking for trouble.

HIPPA aside, I think as designed you will have problems with  
recruitment and retention. I don't want to ask 10 of my friends every  
day if they have any of a list of symptoms. If a friend brought the  
idea up to me, I'd feel uncomfortable being asked those questions  
daily. This is also a lot of work to lay on the observers. Even the  
most ardent recruits may get tired by month 2.

See

http://www.birds.cornell.edu/citscitoolkit

for tips on recruitment and retention, the biggest problem with  
Citizen Science.

Steve K4HG


On Jun 22, 2009, at 6:20 PM, Wm Seffens wrote:

> Citizen Health Observer Program (CHOP)
>
> I am writing a grant proposal for the CDC concerning collecting Flu  
> symptom
> data from volunteers in a manner very similar to the weather program  
> CWOP.
> Volunteers would register with CHOP, get CH0xx ID numbers, receive  
> training
> on flu medical symptoms, and submit the number of friends or  
> coworkers the
> volunteer has daily knowledge of health symptoms. Once a day, the  
> volunteer
> would submit a short packet of coded symptom data for all of the  
> anonymous
> friends. Hence there is a multiplier effect with this data, one  
> volunteer
> could be supplying data on 10-20 persons. This has been cleared with  
> the
> Atlanta office of DHHS for the HIPAA rules on privacy (details and  
> waivers
> will be supplied later at a CHOP website).
>
> This data will be used to monitor flu and H1N1 influenza virus for  
> very
> early detection of epidemics. The data would be appearing 1-2 weeks  
> before
> the people see a doctor and trigger the CDC Sentinel. Like the storm
> spotters, you can be a hero and help the nation.
>
> So, how easy would it be to include this packet into APRS  
> applications (and
> transmitted by RF and/or internet), and how many amateurs would you  
> estimate
> would volunteer? The symptom data field could be expandable to include
> injury data for natural disasters, SAR, etc.
>
> Wm Seffens
> KG4EYO
>
>
>
> _______________________________________________
> aprssig mailing list
> aprssig at tapr.org
> https://www.tapr.org/cgi-bin/mailman/listinfo/aprssig
>





More information about the aprssig mailing list