[HOT] Request for help/guidance on a project to test diarrheal disease interventions in Kendua Sub-District, Bangladesh.

Wed Feb 4 18:01:53 UTC 2015

Hello all, and thank you all for your enthusiasm in moving this project forward with us. We met here at Stanford, yesterday, and below I am pasting the summary of Eric's Bangla schedule, the job and some of our questions/requests for advice. You can find the original posting for help at the bottom of this thread. We would, of course, like to get a project in Task Manager as soon as possible so we can try to promote it and begin work. 

Stace 

Eric Nelson, the PI on the project in Kendua, will be arriving in Dhaka on Saturday night, the 20th of February. He will be in Dhaka until the 23rd, and would like to meet with HOT/OSM Mappers interested in assisting and coordinating training and the mapping effort, in Kendua. On the 23rd, he will be going to Netrokona/Kendua and will be there until the 27th. We are hoping to have some rough prototype of our workflow (using already existing data, though there is very little) by then, for him to discuss with the health workers in the pilot hospitals/clinics. 

Eric is hoping to meet with OSM/HOT mappers in Dhaka and/or Netrokona/Kendua to develop a plan for trainings with locals who may be able to help with the mapping effort. Obviously, the mapping of some of the features we are interested in (see below) will require local knowledge and on-the-ground efforts. As we are looking to launch the actual pilot study hospitals in early May, this trip will essentially be Eric’s only opportunity to meet and coordinate trainings, efforts before that time. 

What we are interested is the capture of some level of geographic information about the home location of patients admitted for diarrheal disease. We have some up with three possible scenarios for capturing the data we want. Scenario #1 is to use a tilecached version of the OSM map (stored locally in the app) resulting from the work we have been discussing, and have health workers ask patients/family for the home location. The health worker would zoom into the area imputed inside the app (this is actually a very minor component of an app that is primarily geared to assessing and providing guidance on treatment for diarrheal disease), and set the location that way, then move on to assessment of the patient. Scenario 2&3 are essentially the same scenario, but using field-papers.org at two different scales. In scenario 2, we have an “altas” of Kendua printed and laminated on the usual 8.5x11 paper, but arranged to create a giant wall map, hanging in the hospital. The health worker asks patient or family to indicate home location and marks the map with a dry erase marker, then photographs the maps sheet that was marked, embedding the photo in the database for later processing. Scenario 3 is, somehow, printing a VERY large (4x5 feet) field-papers-type wallmap and using essentially the same method as #2. In #2 we are worried about the Field-papers margins causing confusion in the map, at the smaller scale. 

We are likely to adopt scenario #1, with either 2 or 3 as the less tech dependent backup. 

We’ve discussed the level of detail we would like to achieve. Currently, there is NOTHING being recorded for this so, really, any level of geographic tag would be useful to use. Eric indicates that even as coarse as the Union/Paurashava would be useful, though better data is better data. Patient privacy is a concern, we will not be interested in specific household level detail. Indication of home location relative to landmarks like places of worship, schools, clinics, markets, etc… would all be sufficient. The reporting of data will likely be in aggregate, probably at the Union/Paurashava. 

We think that mapping of the following is of most utility: 

    * 

primary, secondary or tertiary for main roads connecting towns 
    * 

residential inside residential boundaries 
    * 

tracks and paths 
    * 

Major hydrologic features 

Secondary to that, mapping: 

    * 

clusters of residential buildings to indicate residential landuse 

Finally, if possible, capturing the specific location of the following will be useful 

    * 

places of worship 
    * 

schools 
    * 

hospitals, clinics & pharmacies 
    * 

markets and other landmarks 
    * 

Drinking Water Sources 

Another point we have concerns about is the use of this type of map with patients and their ability to interpret and locate their home communities. As different cultures typically recognize and use different “topographic handrails” to situate themselves in various urban, rural and wilderness landscapes, we are eager to have input on any examples of use of this type of mapping and the levels of success for the purpose we are proposing. We welcome any suggestions/comments on the possible efficacy of the approach, or the set of landmarks we have chosen to focus upon. 

Finally, we would love any input/assistance in creating the Instructions/Description for the Task Manager, and deploying/promoting the project. I’ve been looking at some of the Missing Maps projects and they seem to have a well developed template/protocol for deploying HOT tasks. We’d love to have our project area adopted as a Missing Maps promoted project. We are also looking forward to the trial phase of the study, which will be deployed on a larger region, probably the Netrokona region, as a whole. 

In F,L&T, 
Stace Maples 
Geospatial Manager 
Stanford Geospatial Center 
@mapninja 
staceymaples at G+ 

Skype: stacey.maples 

Get GeoHelp: https://gis.stanford.edu/ 

"I have a map of the United States... actual size. 
It says, "Scale: 1 mile = 1 mile." 
I spent last summer folding it." 
-Steven Wright- 
----- Original Message -----

From: "Ahasanul Hoque" <hoque.ahasan at gmail.com> 
To: "Mikel Maron" <mikel at groundtruth.in> 
Cc: "Stace Maples" <stacemaples at stanford.edu>, "Jorieke Vyncke" <jorieke.vyncke at gmail.com>, "Pete Masters" <pedrito1414 at googlemail.com>, hot at openstreetmap.org, "Eric Jorge Nelson" <eric.j.nelson at stanford.edu>, "Fred Moine" <frmoine at gmail.com>, "Kunce Dale" <dale.kunce at redcross.org>, "Claudia A. Engel" <cengel at stanford.edu> 
Sent: Sunday, February 1, 2015 9:22:39 PM 
Subject: Re: [HOT] Request for help/guidance on a project to test diarrheal disease interventions in Kendua Sub-District, Bangladesh. 

Hi Stace and Mikel, 

FYI, Kendua is a sub district (upazila) of Netrokona District. Kendua also divided in 14 subdivision (13 unions and 1 Paurashava/municipalty). Here I have attached the boundaries and kmz of all for your convenience. I tried to upload in umap but couldnt. Hope Mikel could do it for me. 

Best regards 

Ahasan 

..................................................................................... 
Ahasanul Hoque 
GIS & Environmental Data Mgt Specialist 
WSP, The World Bank. 
MSc in RS and GIS | AIT, Thailand. MSc. in Env. Science| KU, Bangladesh. 
Diploma in Disaster Mgt & Humanitarian Response | 
Uni of Hawai-USA, UNU, Keio& Okayama - Japan; AIT-Thailand . 
Contact: hoque.ahasan at gmail.com ; ahasan_17 at yahoo.com | 
Web: ahasanulhoque.com 
Skype: ahasan4u | Linkedin: http://tinyurl.com/njg3xsp 

On Mon, Feb 2, 2015 at 4:03 AM, Mikel Maron < mikel at groundtruth.in > wrote: 

Stace 

I updated the coordination map of all Bangla projects with the boundary of Kendua 

http://umap.openstreetmap.fr/en/map/bangladesh-mapping-projects_26815#8/23.612/89.742 

-Mikel 
* Mikel Maron * +14152835207 @mikel s:mikelmaron 

On Sunday, February 1, 2015 1:33 PM, Stacey Maples < stacemaples at stanford.edu > wrote: 

<blockquote>

Thanks all. Here is the Umap for our pilot study area: http://umap.openstreetmap.fr/en/map/kendua_27641#11/24.6913/90.7841 , as I understand from Eric, patients arrive at the subdistrict medical center from within the Kendua District, but I wonder if there might be some spillover from adjacent subdistricts (also, please correct my admin boundary terminology, if necessary), based upon travel times. TO account for that, it might make sense to work on a slightly larger envelope than Kendua. 

Yes, I agree on the building footprints being secondary. Our primary objective is to build a map that will provide a familiar enough reference for local health care workers and family members to identify the home village/community of the patients, without being present at the location, as care will be primarily given outside of the home community. Obviously, roads, paths and probably (I am only guessing as I have never been to Bangladesh) water courses would be most important for reference. I have seen some HOT jobs identifying "residential or populated" areas, which might also be useful, short of building footprints. In our discussions, we identified schools, places of worship, markets, etc... as other landmarks that might help users orient. So if we move to creating building footprints, those would be of primary importance. We are also interested in the locations of pharmacies, and clinics/hospitals and other healthcare points of service. 

Finally, and I know this one would require people on the ground with GPS, it would be incredibly useful to identify drinking water facilities/sources. 

Mikel suggested establishing an OSM Bangla Skype Group to coordinate. I've just logged into my Skype account for the first time in years, so it is active. I will make sure I have a Skype client installed on all of my machines by tomorrow. My Skype= stacey.maples 

Again, this response is fantastic. Thanks so much. 

In F,L&T, 
Stace Maples 
Geospatial Manager 
Stanford Geospatial Center 
@mapninja 
staceymaples at G+ 
Skype: stacey.maples 
Get GeoHelp: https://gis.stanford.edu/ 
"I have a map of the United States... actual size. 
It says, "Scale: 1 mile = 1 mile." 
I spent last summer folding it." 
-Steven Wright- 

From: "Jorieke Vyncke" < jorieke.vyncke at gmail.com > 
To: "Pete Masters" < pedrito1414 at googlemail.com > 
Cc: "Stace Maples" < stacemaples at stanford.edu >, hot at openstreetmap.org , "Eric Jorge Nelson" < eric.j.nelson at stanford.edu >, "Fred Moine" < frmoine at gmail.com >, "Kunce Dale" < dale.kunce at redcross.org > 
Sent: Saturday, January 31, 2015 2:05:12 AM 
Subject: Re: [HOT] Request for help/guidance on a project to test diarrheal disease interventions in Kendua Sub-District, Bangladesh. 

Hi Stace and Eric, 
Pete is talking about the same people as I did to you before. Some of our Bangladesh mappers are now also on this list... But I will sent you a follow up mail on this. 
Further I like very much your idea, and would like to give you some input. 
Talking out of my experience; to trace patients, not necessarily all buildings are needed in the first phase. To track patients the main important this is to be able to locate people. So this means collecting locally used neighbourhood names, locally used street names, 
and landmarks used by the people. Buildings are in my view then a second step. 
I don't know how big the area is you're focused on? Maybe you can quickly point it on a Umap for us? Fingers crossed, for good imagery in the area of interest... 
Also I was thinking it might be good to set up an OSM Bangla Skype group to try to coordinate all the upcoming projects a little bit. Lastly there was also interest of Terre des Hommes, the American Red Cross is going to do more things in spring,... So we can coordinate a bit and share resources and thoughts on mapping in the very particular context of Bangladesh. Please let me know if you are interested in this. 

Best greetings, 

Jorieke 

2015-01-31 9:55 GMT+01:00 Pete Masters < pedrito1414 at googlemail.com > : 

<blockquote>

Hi Stace, I have just come back from Dhaka (literally on Thursday), where we were working with the local OSM community to map two areas, Kamrangirchar and Hazaribagh, for the Missing Maps project. We worked with between 10-30 volunteers of varying skills each day for two weeks. They are a smart and enthusiastic bunch and most said they planned to keep mapping anyway. They all have experience in using field papers and surveys and Osmand, and most have at least a days experience using JOSM to edit / upload. 
I have email addresses and phone numbers if you want them or you can contact them via the OpenStreetMap Bangladesh Facebook page. 
There are also a number of very experienced mappers / OSM focused GIS people I can put you in touch with directly. 
Let me know what you think... 
Cheers, 
Pete 
On 30 Jan 2015 21:38, "Stacey Maples" < stacemaples at stanford.edu > wrote: 

<blockquote>

All, 

I'm working with a faculty member studying the efficacy of mobile app based interventions, who needs detailed street and building footprints for his pilot. He is working in the Kendua sub-district of Bangladesh, initially, and needs data for health workers to use to identify cholera patients homes/home village, pharmacies, etc... I've pasted his abstract, below. If he finds efficacy, he will likely expand the project to other sub-districts. We are wondering several things: 

First, what is the process to have a project added to the Task Manager? 

Second, do you happen to currently have mappers in this area who could work on this? 

Finally, we may be able to obtain gps traces from food delivery drivers to upload to OSM. It would be great to have a training for them if there are mappers in the area, or in Dhaka who would be willing to travel. Wondering who to contact about the possibility of that (I know bulk uploads are frowned upon unless coordinated with OSM). 

Thanks in advance for your time, I've pasted the abstract for the project, below my signature. 

In F,L&T, 
Stace Maples 
Geospatial Manager 
Stanford Geospatial Center 
@mapninja 
staceymaples at G + 
Get GeoHelp: https://gis.stanford.edu/ 
"I have a map of the United States... actual size. 
It says, "Scale: 1 mile = 1 mile." 
I spent last summer folding it." 
-Steven Wright- 

Leveraging mobile technology to improve clinical outcomes and scientific research of the second leading cause of childhood death: diarrheal disease 

Abstract 
Diarrheal disease is the second leading cause of death among children under 5 years of age globally. We are specifically interested in the diarrheal disease cholera because of the devastating impact the disease has on at-risk populations and the emerging opportunities to leverage mobile technology to overcome fundamental clinical, epidemiologic, and scientific challenges. Despite effective treatments and advances in provider education, cholera case fatality rates remain unacceptably high. Conventional methods have been unable to overcome barriers to provide patients timely access to care in resource-poor settings. This is especially true early in outbreaks because response teams are slow to mobilize and cholera can infect, transmit and kill in less than 20 hours. Our research challenge is to take an unconventional approach to develop a new method using mobile technology to identify outbreak clusters early, improve care, and advance our basic understanding of the disease. The specific aims of this project are to (i) develop mobile technology for clinical decision support and real-time epidemiology, (ii) test the mobile-technology and determine microbial correlates to disease progression at the hospital level, and (iii) test the mobile-technology and determine microbial correlates to disease progression at the community level. We chose to develop and test this strategy in partnership with the Ministry of Health of Bangladesh at a site with high cholera morbidity and relatively high mortality. We anticipate this NIH funded research will provide an exciting cross-departmental forum for collaboration and training, as well as a pathway to discovery that will directly benefit populations inflicted with diseases like cholera. 

Eric Jorge Nelson, MD PhD 
Pediatric Global Health Physician Scientist Instructor, 
Division of Infectious Diseases Department of Pediatrics, 
Stanford University School of Medicine 
Email: eric.nelson.mdphd at gmail.com 
Telephone: (857)-492-2174 
Address: Beckman B241, School of Medicine, Stanford, California 94305-5323 

In F,L&T, 
Stace Maples 
Geospatial Manager 
Stanford Geospatial Center 
@mapninja 
staceymaples at G+ 
Get GeoHelp: https://gis.stanford.edu/ 
"I have a map of the United States... actual size. 
It says, "Scale: 1 mile = 1 mile." 
I spent last summer folding it." 
-Steven Wright- 

_______________________________________________ 
HOT mailing list 
HOT at openstreetmap.org 
https://lists.openstreetmap.org/listinfo/hot 

_______________________________________________ 
HOT mailing list 
HOT at openstreetmap.org 
https://lists.openstreetmap.org/listinfo/hot 

</blockquote>

</blockquote>

</blockquote>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/hot/attachments/20150204/9a80f837/attachment.html>