Background

Over the past two decades, the rise in transnational trade and travel has the occurence and severity of global illnesses. The impact on tindividuals, communities, countries and continents has been severe. As a result, increased emphasis has been placed on early detection and prevention methods, with newer, affordable technologies driving recent development.

Syndromic surveillance is one such area of particular interest. Broadly defined, syndromic surveillance attempts to collect, analyze, and interpret nonspecific and prediagnostic indicators of disease outbreaks. These may include physical symptoms such as high fevers or diarrhea; activities such as over-the-counter medicine sales; or events such as bird die-off. By rigorously analyzing the data including incidence, prevalence, and temporal-spatial characteristics - we can correctly determine or predict an active or future disease outbreak before it progresses to an epidemic.

Current Situation

In developed countries especially the United States a number of systems exist (e.g. BIOSENSE, RODS, ESSENCE, BIOSENSE, etc) that are resource intensive, using sensors, real-time accumulation of point-of-sales data, or detailed clinical discharge data. Developing countries with limited resources have few systems in place. The grim paradox is that these very countries because of damaged infrastructure, under- funded healthcare systems, and lowered socio-economic conditions are the most impacted by communicable diseases and often the starting point for potential outbreaks.

The Information Gap

While there are systems in place that are either available to or directly target certain developing countries (e.g. ProMed, GPHIN, HealthMap, Voxiva's, etc.), they all suffer from a similar limitation: they require individual lay person reports to come through official sources, including healthcare professionals or news agencies. This can significantly delay the time of reporting between symptom onset and clinical help; potentially leads to an overall increase in the severity of symptoms (and thus community disease progression); and requires symptomatic individuals seek and find official communication channels.

Additionally, lay persons are not easily informed of local disease occurrences in these systems. Such information would provide individuals the knowledge to make informed decisions regarding their own activities, for example changing their immediate behaviors/travels to avoid transmitting or catching a possible infection. The result is a gap in useful disease information and an overall disempowerment of an individual's health status in the wider community.

Closing the Gap: GutFeeling

In order to close this gap, we are developing GutFeeling, a mobile communication system that allows individuals to report disease signs and systems via phone call, SMS, email, or smartphone application, as a well as website. Transmitted data will be collected, de-identified, collated, and mapped, allowing individuals to visualize occurrence of possible disease symptoms in near-real time. Additionally, information regarding occurrences (e.g. incidence, time, and location), will be transmitted back to the individuals. Finally, the de-identified and collated data will be made freely available to the World Health Organization's Global Outbreak Alert and Response Network (GOARN), a system of systems that incorporates syndromic and outbreak data from a number of sources.

GutFeeling thus has 2 primary goals:

• To provide individuals with "situational awareness," or a real-time snapshot of signs, symptoms, and illnesses in their communities that empowers them to alter their behavior to avoid further disease transmission
• And to provide public health agencies (e.g. the World Health Organization) with additional syndromic data and an early overview of community health

GutFeeling Description

Upon feeling ill, individuals will be able to report the symptoms to the GutFeeling database. Information to collect will be based upon signs/symptoms associated with the World Health Organization's (WHO) list of top infectious disease killers, focusing on respiratory ailments (e.g. shortness of breath; bloody sputum); intestinal ailments (e.g. diarrhea); malaria (fever/chills); and measles (including skin manifestation). In addition to symptoms, the system will collect severity (on a 5-point scale); location (geocoded or self-entered); 72-hour travel history; start date (and end date, if applicable) of symptoms; and clinical/treatment status. All data will be "closed ended" and collected via menus to allow for database standardization and simple aggregation and analysis. Traditional outbreak investigations ask who, what, where, and when. GutFeeling data will gather similar data, as follows:

-Who --> Phone number or IP address; last 5 digits of mobile phone (optional identifier)
-What --> One or more related to pre-defined respiratory or intestinal ailments; or those related to malaria or measles; Severity (e.g. 5 pt scale)
-Where --> Geocoded from GPS-enabled phones; manually entered; self-reported 72-hour travel history (e.g. kilometers traveled from current location)
-When --> Self-reported as number of days prior to call (i.e. 0 if today; 1 if yesterday, etc.)
 
Information can be transmitted via:

• Toll-free call menu
• SMSM messaging
• Smartphone application
• Website  

Duplications can be avoided by asking the individual user if she has already reported these symptoms. If "yes," she will be allowed to update her information (e.g. change in severity) or report an additional symptom occurrence.

Because the data will be standardized (based on the system parameters), it can easily be collated and displayed graphically, on both on a map and in simple graphs (accessible via a smartphone and website). Additionally, based on calculated threshold (e.g. incidence of a particular symptom in a given geographic area per population and baseline "denominator" data), a simple notification can be pushed back out to users. Figure 1 summarizes this overview.

Complimentary

GutFeeling is not intended to compete or run in parallel with existing systems, but rather complement and supplement the data already being collected and analyzed. It becomes another piece in the "system of systems'" puzzle. Nor is GutFeeling intended to supplant public health agencies or astute clinical diagnoses. Collected data will be de-identified and made available to international health agencies, including the WHO and the Centers for Disease Control and Prevention (CDC). The WHO and CDC will then perform any and all epidemiological analyses and interpretation, inform local Ministries of Health, and determine any community-wide actions to be taken.

Challenges

As with any self-reporting method, GutFeeling will generate a significant amount of noise (i.e. false positives). However, GutFeeling is not intended to diagnose individual occurrences nor analyze and predict outbreaks (activities better left to clinicians and public health professionals). Our aim is to collect early data that may be of use in conjunction with other system data; and provide individual lay persons a degree of "situational awareness" such that they continue or modify their behavior in such a way to prevent further spread. Additionally, it is our belief that it is worse to prematurely screen or restrict data collection than to collect "too much" data and refine or replace detection algorithms.

Concerns have been raised that unrestricted access to data visualization (such as in Google FluTrends, WhoIsSick.org, and Healthmap.org) could harm   the tourism economy of countries with high counts of possible but unverified or inaccurate symptoms. It is our belief that data restriction and control causes more harm (unintentional or otherwise) than access to information. However, an individual's privacy must be respected especially if the system is to gain users. Consequently, care will be taken to de-identify publically released data. For example, map resolution will be such that an individual reporter's home will not be pinpointed but rather an area of a fixed width (e.g. a circumference of 100 yards).

Marketing the system to the target population will require partnering with grass roots agencies as well as notifying local mobile carriers of its existence. The usefulness of GutFeeling is dependent on its adoption by a large number of users.

Additional Modules

Additionally, GutFeeling can incorporate symptom/disease reporting of animals in a similar format.

Expected Impact

GutFeeling will provide individuals with a mechanism to report illnesses, a updated snapshot of their communities' health, and information enabling them to make immediate behaviorall changes to prevent possible disease spread. It will also provide public health agencies -- in particular the WHO -- additional and missing syndromic data. A final benefit is that individual users will gain a better awareness and understanding of the dynamics of disease spread and take responsibility to protect and ensure their communities' health.

Measures of Success and Benefits

Success will be measured by the number users and disease reports (as compared to historical data); incorporation of data by global public health agencies in surveillance systems; and ultimately by the frequency and severity of disease outbreaks (again, as compared to historical data and anticipated occurrences).

The benefits of GutFeeling are to provide additional data to public health agencies; to provide near real-time public health information to individuals and communities; to arm individuals with the knowledge that may influence behavior in a way that decreases disease