photo du big data

Big Data in times of pandemic

With a significant number of active cases worldwide, the novel coronavirus represents an extreme public health challenge. The outbreak of the pandemic has led to strict travel restrictions, making data collection very difficult for the development research community.

Traditional data collection methods that require field visits can be risky in these difficult times. In this context, if appropriate privacy and ethical safeguards are put in place, big data is both relevant and useful, now more than ever. For example, global positioning system coordinates obtained from cell phone records can be useful for tracking people’s movements. During the pandemic, there is immense potential to use this data to predict hotspots and stop the spread of the virus. Another example: sentiment analysis by leveraging social media data could provide useful insights to help design appropriate health messages for the public.

A new systematic map supported by CEDIL and developed by 3ie brings together a unique and comprehensive collection of studies that use big data to measure or evaluate development outcomes. The map covers impact evaluations that use big data to assess development outcomes, systematic reviews of big data impact evaluations, and other measurement studies that have innovatively used big data to measure and validate any development results. This blog is an attempt to highlight the role that big data can play in solving public health problems. We provide an overview of our gap map results and then discuss the potential use of big data in healthcare.

What did the map reveal about Big Data in healthcare?

Of the 437 studies included in the map, 63 examined health-related developmental outcomes. Twenty-eight studies examined interventions aimed at reducing mortality, and another 28 evaluated interventions aimed at ending the epidemic of a communicable disease. There is, however, a lack of assessment of the impact of an epidemic, both in small units like districts and in large units like a state or a country.

Satellite data was used in 29 studies and was the most frequently used source of important data. For example, one of the included studies examined vaccine coverage against measles outbreaks in Niger. The study merged satellite-derived population distribution measurements with high-resolution measles cases reported in the country. This study was closely followed by cell phone call detail recording (CDR), which was used in 27 studies. A study conducted in Haiti aimed to evaluate whether CDR could predict the early spatial evolution of the cholera epidemic. Additionally, our map results show that the greatest number of large data studies related to outbreaks have been conducted in sub-Saharan Africa, with the fewest in the Middle East and North Africa.

The map highlights a significant lack of data when it comes to studies related to outbreaks in fragile contexts. Regarding overall data gaps on other health outcomes, deaths from road traffic accidents, substance abuse, and sexual and reproductive health services have rarely been studied.

A lire également  iCloud: What is iCloud and how does it work?

Big Data big data health table during the pandemic Covid-19

With the number of coronavirus cases increasing exponentially, important data can help detect outbreaks. By bringing together data from a variety of sources, we can use algorithms to analyze medical records and trace patient contact history to help identify patterns of virus spread. These apps can help delineate not only current areas with high case numbers, but also predict future outbreaks through travel and contact tracing.

An artificial intelligence technique called natural language processing (NLP) is worth mentioning here. By analyzing regular human interactions in the form of text and speech, NLP can help make human communication more meaningful. It can be used to analyze social media links and online newsletters, which can potentially raise an alarm when there are new developments related to COVID-19 around the world. NLP and other big data techniques can also be used in incident detection so that in future, such health emergencies are dealt with quickly.

Despite significant progress in recent years, these technologies are still new and many implementation challenges, such as information overload and data ambiguities, remain.

Australia COVIDSafe
Israel Hamagen
Singapore TraceTogether
France StopCovid
Germany Corona-Warn
India Aarogya Setu

Many countries around the world are trying to flatten the pandemic curve using smartphone apps (see the table showing examples of mobile phone apps). These apps monitor people’s movements to determine if they are in high-risk locations or have been in contact with high-risk people.

The increasing use of big data, however, has raised ethical concerns and posed legal challenges. These mobile phone applications have access to a significant amount of personal information. Ethical issues include privacy, lack of personal autonomy, and the public’s demand for transparency and fairness when using big data. It is therefore very important to carefully consider and implement privacy policies when using big data.

Despite privacy concerns, big data has a promising future in healthcare. As travel restrictions persist in many countries, there may be increasing opportunities to use large data to compensate for the lack of in-person data collection. However, to achieve this, financial investments are necessary. If healthcare organizations want to use big data, they will need to invest in the necessary technology, infrastructure and staff training. They would need smartphone apps with strong privacy protections and the IT infrastructure needed to work securely with large amounts of data. Most importantly, they will need to train their staff in data analysis techniques. To explore the potential of big data in healthcare, a more systematic evaluation of available methodologies by the research community is needed. Given the current public health crisis, it will be useful to see if big data can be used to predict possible outbreaks in the future.