A personalized risk assessment tool for COVID-19 and other respiratory illnesses
- Gives a personalized assessment of your risk of contracting respiratory diseases, including COVID-19, updated daily.
- Helps you keep yourself and your community safe by keeping you informed about the spread of respiratory disease.
- Helps you make proactive and intelligent decisions about your life by helping you understand how likely you are to get sick.
- Hunala leverages cutting edge network science and machine learning (and uses both crowdsourced and public information) to determine your likelihood of contracting the flu or respiratory illnesses.
- Using the information you provide about your location and your social interactions, Hunala maps the distribution of symptoms and can offer information you can decide how to use with respect to avoiding illness. By updating information daily, you have a powerful tool for monitoring your risk of exposure.
- Hunala assesses what is happening where you live and in your extended social network, similar to apps that crowdsource information about traffic conditions.
- Hunala ensures your privacy and does not share your personal information with other users or with third parties.
- Fill out a brief daily survey (which is just a couple of yes/no questions if nothing is happening, and otherwise takes about a minute) to self-report any symptoms you may have
- Tell Hunala where you live and who you spend time with (once, and thereafter updated as you wish)
- Tell Hunala other information related to your risk factors (once, and thereafter updated as you wish)
Hunala relies on quite a number of methods from network science and machine learning to process data and develop forecasts. Some of these ideas are illustrated by the below links.
- Watch our TED talk
Read some published work on network sensor technology:
Read some published work on machine learning and privacy:
Hunala’s network sensing technology was developed in the lab of Professor Nicholas A. Christakis, MD, PhD, and his Human Nature Lab team at Yale University. For more information, please visit http://humannaturelab.net
The development of the machine learning component of Hunala was spearheaded in the lab of Professor Amin Karbasi, PhD, and his Inference, Information, and Decision Systems Group at Yale University. For more information, please visit http://iid.yale.edu
Relative risk compares the current risk in each county to all of the other counties in the country.
Absolute risk is the order of magnitude of cases per county.
Historical risk shows the current risk relative to a historical baseline of respiratory disease.
June 5, 2020, Yale News, "Yale app Hunala aims to be ‘Waze for coronavirus’"
May 23, 2020, a16z Podcast, "Pandemics — Early Detection, Networks, Spreaders with Nicholas Christakis and Jorge Conde"
April 27, 2020, NPR Morning Edition ‘Health’ by Allison Aubry, “New Normal: How Will Things Change In Post Pandemic World”
April 22, 2020, The Atlantic, “Would You Sacrifice Your Privacy to Get Out of Quarantine?”
If you deny Hunala access to your contacts and want to grant it permission later, you can do so in your phone's settings. In Android, go to Settings, Apps & notifications, Hunala, Permissions, Contacts, and select the "Allow" option. In iOS, go to Settings, Hunala, and enable the switch next to "Contacts".
We pronounce it who-NA-la.
We tried dozens of names and then just used an acronym for our lab, the Human Nature Lab.
Hunala ensures your privacy and does not share your personal information with other users or with third parties. We will first ask you basic questions about who you are: for example, your age and any health conditions you might have. On a daily basis, we will ask you whether you have been experiencing any symptoms of a respiratory or other illness, and how you may have responded to that. We will combine that information with data available from USAFacts.org and the CDC on the number of infections in your area and will give you estimates of your risks of contracting a respiratory illness. We will ask you to connect your friends to the app to more accurately (and anonymously) calculate your risk. The more people who use the app, the more accurate the estimates become, similar to apps that track traffic conditions in an area.
Our contact with other people, and the people they in turn have contact with, directly affects our risk of exposure to viral illnesses. The Hunala team recognizes this, and Hunala uses this (anonymized) information to inform you about your own risks. We will ask you to connect your friends to the app to more accurately calculate everyone’s risk. The more people who use the app, the more accurate the estimates are, like traffic apps that grow in accuracy as more people in an area use them.
The Hunala team uses principles of network science for the early detection of outbreaks of respiratory disease. To construct a scheme of the social network that extends around you, we will ask you to nominate a number of your friends and family members. We will only use the phone number of the people you nominate for identification. The app will request permission to temporarily display your contact list so you can more easily select these people. Those people will be invited to also use Hunala. We will not pester them to join, and we do not tell other users who suggested them.
You will receive two risk calculations: individual risk and regional risk.
The individual risk score is based on information about your health, your social behaviors, and your extended social network and what is happening in it (for instance, whether your contacts’ contacts reported respiratory symptoms in the recent past). The mathematical models and procedures we use for this may be frequently updated as more people use the app.
Your regional risk is calculated by comparing the current number of COVID19 cases to a regionally tuned baseline condition where there is not a lot of “Influenza-Like Illness” (ILI). We look at historical data for respiratory illness in your region for this time of year, and compute a score based on how extreme the current COVID19 cases are compared to this baseline.
The regional risk score is computed using data provided by USAFacts.
The risk level the app provides is a statistical calculation. It does not replace the advice and risk assessment made by your health care provider. In addition, the risk determination does not decrease your responsibility to comply with recommendations or requirements imposed by the Centers for Disease Control (CDC) and other federal, state, and local authorities established to reduce the spread of disease.
Hunala provides you with information about the relative risk for you and your community of contracting an influenza-like illness, including COVID-19. Researchers will use the information collected to study how viruses spread.
Your data will be used to:
- provide you with the app's services of helping you track your health;
- inform future research, subject to approval of the University ethics board;
- assist with public health surveillance and pandemic planning;
- assist with the examination or forecasting of other trends related to topics we might ask you about in the future;
- monitor and improve the functionality of the app and services provided;
- provide you with information about other projects we are developing that may be of interest to you.
We carefully protect your privacy. We will remove any identifiers from the data prior to using or sharing the data with researchers, where possible. Identified data you provide or which we collect will never be sold or shared with advertisers without your explicit permission. Hunala ensures your privacy and does not share your personal information with other users or with third parties.
You do not get notified about particular contacts. Likewise, we do not tell your friends who use Hunala any information about you as an individual. If someone in your network reports a diagnosis or other risk factors, your risk index will be adjusted accordingly.
Your physical location is important to your risk of contracting viral illnesses. This information will be used in our statistical calculations to provide your risk assessment
You may withdraw your consent at any time by emailing firstname.lastname@example.org
Hunala is the only app that (to our knowledge) gives users feedback about their risk of exposure to, or getting sick from, COVID-19 and other communicable respiratory diseases. Most of the available symptom-reporting apps allow users to contribute to collective efforts to fight the coronavirus pandemic. Hunala does this too, but, unlike other apps, Hunala also gives users feedback and information they can use to better understand their own relative risk. Hunala is also not a contact tracing app, which are used to locate and alert people who have been in contact with someone who later tests positive. Hunala aggregates anonymized information and does not notify anyone about their direct exposure to particular individuals, keeping everyone’s privacy intact while letting the user know if there is need for concern. In this regard, it’s like popular crowd-sourcing apps that alert users to traffic problems road hazards but do not inform you about who else is on the road or who reported a given problem. Users get the information they need to enable better decision making without giving up their privacy.