By Dr. Vikram Venkateswaran
India is perhaps the only country in the world (of its size) where addressing preventable diseases continues to drain public resources. The Delhi dengue outbreak is a case in point. This is the third or fourth time in a decade (the first being in 2006) that the city has failed to prevent the spread of this disease. Close to 3,500 cases of dengue have been reported in Delhi over the last month, of which nearly all cases required hospitalization.
If the government wants to prevent such large scale outbreak of disease in the future, it needs to consider leveraging disease surveillance systems that rely on digital/social media to inform, educate and warn citizens.
How a digital disease surveillance system can help
A digital disease surveillance system collects data from multiple sources (in real time or otherwise) and can help indicate the potential of disease outbreak in certain locations, thereby enabling the government to take steps to control the situation.
Mature economies have been doing this for some time now and rely significantly on digital and social media to prevent and manage outbreaks. For instance, the US was able to identify cases of Ebola around 9 days before the World Health Organization declared the Ebola epidemic using a software that mines social media. This software, part of an infectious disease surveillance system, was able to pick up instances of a âmystery hemorrhagic feverâ from among the various entries listed on government websites, local news sites, and social networks, correlating it with Ebola.
Mamadouba ContĂ©, working in Ebola lab at Donka Hospital; Pic by S. Hawkey/WHO
In other cases, it looked at the social media feed from users who indicated their ailments â directly (by looking at feed that saw people discussing their illness or indicating that they were unwell on social media) or indirectly (by analyzing the number of people searching websites like Google, looking to investigate their symptoms and seek cure). This could have easily been applied to India, where we often self-medicate and hesitate to seek out professional medical help, unless oneâs health significantly deteriorates. Especially in cases like dengue, symptoms arenât unique enough to arouse suspicions from patients or their families to immediately seek medical help.
Another example is that of a disease detection app called Flu Near You. This app helps predict outbreaks of the flu in real time. Users self-report symptoms in a weekly survey, which the app then analyzes and maps to show where pockets of influenza-like illness are located. Although there is an element of potential inaccurate reporting by citizens and the real possibility of symptoms not correlating with any specific disease (such as cold or headache may not directly help identify a serious condition), it is still a model that India could use in some areas where care is difficult to administer.
A fever clinic especially set up to cater to those suffering from fever, one of the main symptoms of several mosquito-borne diseases, in Delhi; PTI
Integrating weather into disease surveillance
Some developed nations are also working on a more comprehensive disease monitoring mechanism that involves use of weather data. After all weather influences our health and any seasonal variations tend to bring about the bulk of diseases in the world.
Germany has been sharing âhealth-weatherâ forecasting for three decades now. Newspapers not only carry the weather data, but also include information such as likely diseases accompanying the weather (upto 40 different diseases are covered) as well as symptoms (such as irritation, insomnia). On TV, varying degrees of symptoms and diseases are also provided. This website provides flu activity, pollen allergy and UV burns related information or Dachau in Germany. Media company AccuWeather now provides such health forecast for other countries too, like this one focused on allergies in New York city.
Japan undertook health-weather forecasting by including Ultra Violet (UV) forecasting, adjusted to skin type as part of its surveillance mechanism in 2008. The country has been divided into 39 regions based on significant differences in terms of lifestyles, clothing, and sensibilities of residents. Each region sees certain variable such as heat, air temperature and humidity dominating public health. For each region, these unique parameters are considered before the weather-health forecast is put out. These forecasts can be obtained by any citizen with a mobile phones. Those visiting the website where these forecasts are made, have the option of sharing feedback which is then used to validate the forecast.
Britainâs NHS piloted a health forecasting unit in 2001 that integrated weather data with infectious disease surveillance data in real time from five locations to help generate workloads prediction for those areas. For instance, flash warnings were given to ambulance services and accidents and emergency services units if the weather snowed or ice fell in anticipation of increased trauma and falls. Today this system has evolved to such an extent that hospitals prepare themselves for treating patients of heart and lung conditions that can worsen during cold weather. A comprehensive alert system has been set up and cold weather alerts are provided between November and April. A similar heat wave related alert system is also in place.
Considering the dengue outbreak in India first started in 2006 and has been recurring over the last 3-4 years involving similar casualties, the government could have used historical data to identify vulnerable areas and taken extra efforts to mitigate the risks this time.
Social media in disease surveillance
The Chicago Department of Public Health uses Twitter to identify cases of foodborne disease outbreak that analyzed tweets that referred to food poisoning in the area. Using an app called Smart Chicago, the local government aimed to identify hotels and food joints that could be operating under unhygienic conditions (thereby flouting norms) where patrons were falling prey to food borne diseases. These establishments were then inspected. The New York City Department of Health has tied up with Yelp (a company that offers restaurant reviews, similar to Indiaâs Zomato) to create a similar app to help detect food borne diseases.
In the context of the Delhi dengue, a similar app may have been able to point at potential sources where such mosquitos could be breeding. After all most food joints and street food hawkers in India tend to store fresh water for their patrons consumption and this is breeding ground for the dengue mosquito.
To ensure real time detection of flu in the US (which is the most common ailment and the reason for student/employee absenteeism in most cases), the government is collaborating with external research groups to forecast seasonal flu outbreaks. The current domestic influenza surveillance system informs public health decision-making, but it lags behind real-time flu activity. The goal of the new infectious disease forecasting is to provide a more-timely and forward-looking tool that predicts rather than monitors flu activity so that health officials can be more proactive in their prevention and mitigation responses.
For example, if it were possible to predict when and where flu activity will peak, health officials, health care providers and other partners could plan in advance to optimize the timing of flu vaccination clinics and communications outreach efforts around vaccination, as well as distribution of flu antiviral medications, according to a press release.
In the Delhi dengue context, if hospitals and healthcare professionals had used digital/social media to indicate the number of dengue cases that was being treated every hour, the severity of the situation would have come to light much earlier. The government could have responded faster with measures such as fumigation, distribution of vaccines, and even undertaking preventive checks in vulnerable localities.
A municipal worker fumigates a residential area in New Delhi; PTI
Indiaâs population is not slowing down any time soon and healthcare funding as a percentage of GDP continues to shrink. Under such circumstances, digital disease surveillance can provide a cost effective way to manage disease and improve health care outcomes. It can bring together various stakeholders like healthcare professionals (who on their own donât tend to use digital media for creating awareness on diseases), patients (who need authentic medical information online), and healthcare providers (like hospitals and clinics who can share disease information) to comprehensively tackle disease and its symptoms.
The author is Founder and Editor, Healthcare in India, an organization that provides advice to healthcare organizations on their digital, marketing and social media strategies.
(Note: The views expressed here are the personal opinions of the author.)