Associations with SARS-CoV-2 Positivity Among People in Refugee Settlements
This study in Ugandan refugee settlements found that while few self-reported symptoms were significantly associated with SARS-CoV-2 positivity, with only unusual fatigue remaining an independent predictor after adjustment, interactive voice response platforms offer a feasible strategy for screening and surveillance in resource-constrained humanitarian settings.
Original paper dedicated to the public domain under CC0 1.0 (https://creativecommons.org/publicdomain/zero/1.0/). This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer
When a virus spreads quickly through a crowded place, health workers face a difficult choice: who should get tested first? In the early days of the global pandemic, testing supplies were scarce, especially in remote areas where many people lived close together. In refugee settlements, where thousands of families share limited resources, the risk of an outbreak was high, but the ability to check everyone for the virus was low. Health officials needed a way to find the people most likely to be infected without testing everyone. The common idea was to look for specific signs of sickness, like a fever or a cough, and assume that people with those signs were the ones carrying the virus. But this approach only works if those signs are reliable. If a sick person feels tired but has no fever, or if a healthy person has a cough from dust, then sorting people by symptoms might miss the real danger or waste precious tests on those who are safe.
To find out if symptoms could truly predict infection in these difficult settings, researchers in Uganda set up a unique system using ordinary mobile phones. They created a free service that anyone could call. When a person dialed a specific number, they did not speak to a human immediately. Instead, they heard a recorded voice in their own language asking a series of questions about how they felt and whether they had been near anyone sick. This technology, known as interactive voice response, allowed the team to reach thousands of people who did not have smartphones or internet access. The goal was simple: listen to what people said, and then invite those who reported symptoms to come to a clinic for a real test. By comparing what people said on the phone with the actual results of their medical tests, the researchers could see if the symptoms people described were actually a sign of the virus.
The study took place between February 2021 and March 2022 across seven refugee settlements in Uganda. During this time, nearly 15,500 people called the service. Of those, about 7,000 completed the full survey. Most of the callers were refugees living in the settlements. When the researchers looked at the data, they found that three out of four people who finished the survey said they had at least one symptom or had been exposed to someone sick. This group was then invited to a health center for testing. Because of distance and resources, only a portion of these callers could make it to the clinic within a week, but 751 people were successfully tested. The results showed that 6.5 percent of those tested were positive for the virus.
The researchers then asked a critical question: could they have predicted who was positive just by listening to the symptoms people reported? They looked closely at the answers given by the 741 people whose test results were available. The most common complaints were general cold symptoms and feeling feverish or having chills. Almost everyone who came in for testing had reported these issues, regardless of whether they actually had the virus. This meant that fever and cough were not good at separating the infected from the uninfected. However, two specific feelings stood out. People who reported severe muscle pain or unusual, deep fatigue were more likely to test positive than those who did not.
When the team adjusted their analysis to account for other factors like age, gender, and whether the person was a refugee, the picture became clearer. The link between muscle pain and the virus became weaker and was no longer a strong predictor on its own. But the feeling of unusual fatigue remained a consistent sign. People who described this specific kind of exhaustion were twice as likely to test positive compared to those who did not feel this way, even after considering other health conditions like high blood pressure or diabetes. Despite this finding, the researchers noted that even with this stronger clue, symptoms alone were not a perfect tool. Many people with the virus did not report these specific feelings, and many people who felt tired did not have the virus.
The study concluded that while listening to symptoms is not a perfect way to decide who needs a test, the phone system itself was a powerful tool for reaching people. The technology successfully gathered information from thousands of individuals in a short time and connected them to the healthcare system. The researchers found that the main challenge was not the ability to talk to people, but the fact that the virus does not always show clear, unique signs. The symptoms of this infection often look like many other common illnesses. This suggests that in places where resources are tight, relying only on how a person feels might miss many cases. Instead, systems like the one used in Uganda are valuable for keeping a broad watch on the community and guiding people to care, even if the symptoms they report cannot pinpoint the virus with absolute certainty. The work highlights that in humanitarian crises, the best approach may be to use technology to stay connected and offer testing to anyone who is concerned, rather than trying to guess who is sick based on a checklist of symptoms.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.