DigiActive.org

Campaign: InSTEDD’s Mekong Collaboration Program for Early Detection and Early Response

Description and Background: In 2006, Dr. Larry Brilliant won the TED Prize and called for the development of a technological system that would stop the threat of pandemics and catastrophes. Renowned for his work in the successful, worldwide eradication of smallpox, Dr. Brilliant emphasized the game-changing purpose of this system – early detection and early response.  As such, he declared his TED wish, “I wish that you would help build a global system to detect each new disease or disaster as quickly as it emerges or occurs.”  From this prize and wish emerged InSTEDD (Innovative Support To Emergency Diseases and Disasters), originally titled “International Networked System for Total Early Disease Detection.”  Shortly after TED, Larry Page and Sergey Brin (Google’s founders) appointed Dr. Brilliant the head of Google.org, the organization’s philanthropic arm.

With funding from Google.org, the Rockefeller Foundation, and others, InSTEDD began its social, technological mission with CEO Eric Rasmussen (a former U.S. Navy chief doctor for disaster response in the Pacific), Dr. Dennis Israelski as the Vice President of Global Health, and Eduardo Jezierski as the Vice President of Engineering.  Along with Chief Technology Office Robert Kirkpatrick, Chief Operating Officer Judith Kleinberg, and several other full-time team members, InSTEDD operates very much like a classic startup in which everyone wears many different hats (click here for bios).  In addition to the executive team, InSTEDD has over a dozen dedicated team members on the ground in their Mekong Collaboration Program (MCP).  This program focuses on rapid detection and rapid response in the Mekong Basin of Southeast Asia – Cambodia, Thailand, Lao PDR, Vietnam, Myanmar, and the Yunan province of China – and has been stationed in Phnom Penh, Cambodia since 2008. In the region, InSTEDD partners with the Mekong Basin Disease Surveillance Network (MBDS), a collaboration network of the six countries’ local governments and Ministries of Health that work to improve cross-border disease information sharing.  Not only does InSTEDD work closely with MBDS, but the organization also established a field-based Innovation Lab (iLab) in Phonm Penh in August 2008.  By empowering young Cambodian developers to design technological solutions based on their local expertise and understanding, InSTEDD works to maintain a creative, organic environment in which sustainable innovation can thrive.

Digital Tools Being Developed: GeoChat, Mesh4X, Evolve (tools that incorporate Twitter, RSS feeds, Facebook, SMS text messages, and so on, as applicable)

How These Tools Are Being Designed and Used: As Ed Jezierski explained to me in our Skype interview, GeoChat helps people in the field to share and report data with each other as well as with those at headquarters, amplifying the amount of information flowing between health workers; Mesh4X coordinates this information by synchronizing between various organizations’ virtual databases; Evolve functions as a place to curate the collective information and subsequently analyze it for conclusions.  For more details about InSTEDD’s approach, tools, and impact, read more after the jump.

The key factor that differentiates InSTEDD from other organizations also contributes to their technology’s success – their approach.  Through field observation, team members learned that the technology should be completely controllable from SMS and mobile phones (because SMS channels tend to stay open during disasters), as well as function with low signal or even offline.  In the iLab, InSTEDD’s MCP helps Cambodian developers understand the process of determining local needs and innovating applicable, technological solutions.  The lab fosters their ability to re-evaluate and repurpose new technology in the future.  MCP also sends these members from the lab to the field in different disaster or outbreak response events; the staff observe these efforts and test out technology with locals in order to determine which user interface or SMS they would rather use or receive. And finally, all of InSTEDD’s tools are open-source and free – a feature Dr. Brilliant envisioned that is critical for continual innovation and adaptability.

Though the MCP has ten to fifteen small, open-source projects going on continuously (including hotlines, AIDS patient reminders over SMS, and so on), I spoke with Ed Jezierski about the following three interoperable tools:

GeoChat enables people in the field to communicate via SMS with others using the Internet at an operations center.  Those in the field can text their location and GeoChat will pinpoint them on a map, allowing anyone with a computer to click on that location and engage in a bidirectional conversation across platforms.  It also allows users to subscribe to relevant health alerts so that health volunteers/rapid response teams can receive and report important data with their mobile phones at any time.  Ed outlined the following ways in which GeoChat operates…

• By subscribing people in the Mekong region to GeoChat health alerts, they receive vital information back.  By transforming health bulletins into locations on a map and sending relevant SMS blasts to health volunteers, those in the region can be more aware of disease symptoms and therefore catch the disease before it spreads.
• For cross-border communication, GeoChat provides an “international chatroom” where people on either side of, say, the Lao-Thailand border can quickly inform each other of potential diseases (e.g. children or animals getting sick) by sending local SMS. Regional borders, after all, are prime areas for disease outbreak due to trade between humans and contact with animals.
• GeoChat is always evolving. The team adds features that people request. For instance, if people are using Twitter, they can send a Twitter message that others will see as an SMS message on their mobile phones; these people can reply with an SMS text that will then appear to the Twitter user as a Twitter message.  In another example, GeoChat can incorporate the use of Thuraya satellite phones.  Basically, GeoChat provides social network capacity that can be customized for local needs.

Mesh4X synchronizes data in a seamless way across organization’s differing systems and databases.  It allows everyone to see the same collective information as well as revise the information in one shared virtual database.  Take the example of registering survivors of a natural disaster:

• Various organizations registering survivors can direct their existing applications/databases (e.g. an Excel spreadsheet or an isolated system) to the same Mesh and merge their data using one URL and password. This single database consolidates data from different systems into the same set of columns.
• The database is visible to any field workers with Java-enabled mobile phones, so that anyone who can access the information can simultaneously edit the database online or offline (changes send whenever they find a signal).  Using any application, people can add to or clarify their data so that these changes will appear in everyone else’s spreadsheets/systems.
• I think of Mesh4X along the lines of the The Blind Men and the Elephant parable.  With several different survivor databases, survivors would have to seek out each system and families would have to consult each one to find out if their loved one is still alive.  With Mesh4X, all of the survivor data would be readily available in one master database.  Or, since each village’s experience with one disease results in fragmented data and potentially incorrect diagnoses, Mesh4X would enable everyone to pool their data for a more thorough analysis.

Evolve allows multiple teams around the world to collaborate around streams of information regarding a specific disease or disaster. From news feeds and reports to emails and SMS alerts, Evolve parses huge amounts of information into keywords, locations, and times to form a common shared picture of an event’s progress.  Ed explains the tool in terms of enhancing disease detection by improving timeliness (detecting a disease as early as possible), sensitivity (aggregating information as broadly as possible), and specificity (filtering information as accurately as possible).

• As incoming information (Twitter updates, RSS feeds, SMS messages, etc.) flows in, people collaboratively add tags and comments, “curating the content.”  People use tags to hypothesize what a disease might be. Each item is put on a map for a visualization of updates.
• Evolve learns these tags (according to keywords, location, time) and develops the ability to suggest tags.  When an analyst looks at the information, he/she can thumb up or thumb down the tags to better sort the information that can be confirmed. By making it easier to detect and understand what is going on, decisions can be made more rapidly about how to stop the disease.
• Evolve can be made public in situations like the swine flu outbreak, but can also be used privately for national security and public health issues.

Impact: Thus far,  InSTEDD has seen great local implementation of its tools.  The Cambodian CDC now uses Mesh4X and three countries in the MDBS network registered at last month’s ICT Forum to begin using GeoChat to communicate with each other.  Shannon Oliver, InSTEDD’s Director of Communication for the Mekong region, actually informed me that all the representatives from the six countries in the MBDS network committed to adopting GeoChat at last month’s conference – three countries remain to be registered.  Shannon also explained how, as of two weeks ago, ChangeFusion and OpenDream (two social entrepreneurial companies in Bangkok) began piloting GeoChat with the MBDS/InSTEDD team in Mukdahan, Thailand.  Since Mukdahan borders Savannakhet, Laos, the Thai border region has been vigilant in monitoring and reporting the progress of swine flu, sending out blast messages to people on the status of the disease in the area.  Many other proof of concept pilots have been conducted, in which people use various technologies to see how they work in the field; these pilots have successfully proven that the products are viable and applicable on a local scale.

Moving forward, the Mekong Collaboration Program faces such challenges as maintaining agile learning and quick turnaround with its technology.  They hope to apply their concepts in other regions in the years to come.  In doing so, InSTEDD also wants to transition to a social enterprise model as opposed to relying solely on outside funding and grants.

Ed characterized MCP’s initial challenge in the region as its biggest opportunity – the ability to work with regional governments/Ministries of Health in a neutral way and to build trust with local residents as well as disease response organizations.  This organic process with a locally-focused approach has allowed the MCP to draw from the talented and creative people in the region who want to use technology to reduce human suffering; Ed describes the iLab as “tapping into a latent energy” that far exceeded his expectations. When asked about MCP’s success, both Ed and Shannon emphasized the qualitative nature of InSTEDD’s work.  Rather than trying to quantify lives saved, they look to MCP’s impact on public health.  Both team members were optimistic, as am I, about the broader scalability of InSTEDD’s tools in the ongoing work to improve early detection and rapid response.

Sources: Skype interviews with Eduardo Jezierski (VP of Engineering) and Shannon Oliver (Director of Communication, Mekong Region); InSTEDD website and blogs; Robert Kirkpatrick’s SlideShare presentation; Dr. Larry Brilliant’s TED speech video; BusinessWeek’s InSTEDD interview