Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Time series anomaly detection in helpline call trends for early detection of COVID-19 spread across Sweden, 2020

Hashemi, Atiye Sadat LU ; Dietler, Dominik LU orcid ; Fall, Tove ; Inghammar, Malin LU ; Johansson, Anders F ; Bonander, Carl ; Ohlsson, Mattias LU orcid and Björk, Jonas LU orcid (2025) In Scientific Reports 15(1).
Abstract

Timely detection and surveillance of disease community spread is a potent tool for implementing effective public health interventions. This study investigates the National Telehealth Service (1177 helpline) across 18 regions in Sweden in 2020 to identify early signals of community transmission of COVID-19 at the beginning of the pandemic. Focusing on calls related to key COVID-19 symptoms (cough, fever, and breathing difficulties in adults), we analyze their frequency and distribution across referral categories, comparing them to 2019 data. We employ an explainable time series anomaly detection algorithm using daily call data to identify the first collective anomalies across regions. The results show that anomalies in call data were... (More)

Timely detection and surveillance of disease community spread is a potent tool for implementing effective public health interventions. This study investigates the National Telehealth Service (1177 helpline) across 18 regions in Sweden in 2020 to identify early signals of community transmission of COVID-19 at the beginning of the pandemic. Focusing on calls related to key COVID-19 symptoms (cough, fever, and breathing difficulties in adults), we analyze their frequency and distribution across referral categories, comparing them to 2019 data. We employ an explainable time series anomaly detection algorithm using daily call data to identify the first collective anomalies across regions. The results show that anomalies in call data were correlated with, but preceded, the first confirmed case infected in Sweden by a median of 7 days (IQR: 2.5–10.5) and the first hospitalized case infected in Sweden by a median of 13 days (IQR: 7.25–16). They also preceded the estimated onset of community spread, indicated by the absolute confirmed cases (median: 24.5, IQR: 18.25-32.5), and severe outcomes defined by hospitalizations (median: 33, IQR: 27.25-44). These findings showcase how helpline call monitoring, using time series anomaly detection, can aid early outbreak detection.

(Less)
Please use this url to cite or link to this publication:
author
; ; ; ; ; ; and
organization
publishing date
type
Contribution to journal
publication status
published
subject
keywords
Humans, Sweden/epidemiology, COVID-19/epidemiology, Early Diagnosis, SARS-CoV-2/isolation & purification, Telemedicine, Hotlines, Adult, Algorithms, Pandemics
in
Scientific Reports
volume
15
issue
1
article number
32701
publisher
Nature Publishing Group
external identifiers
  • scopus:105016908829
  • pmid:40993291
  • pmid:40993291
ISSN
2045-2322
DOI
10.1038/s41598-025-20641-2
project
eSSENCE@LU 10:6 - Pandemic preparedness in the era of big data: Disease surveillance tools using individual-level register data and novel mobility data
Explainable and Just AI in Data-Driven Disease Surveillance
Improved preparedness for future pandemics and other health crises through large-scale disease surveillance
language
English
LU publication?
yes
additional info
© 2025. The Author(s).
id
c322c5fe-8c0b-491e-bcdf-9b3804f8d230
date added to LUP
2025-09-29 09:20:07
date last changed
2025-10-17 11:07:36
@article{c322c5fe-8c0b-491e-bcdf-9b3804f8d230,
  abstract     = {{<p>Timely detection and surveillance of disease community spread is a potent tool for implementing effective public health interventions. This study investigates the National Telehealth Service (1177 helpline) across 18 regions in Sweden in 2020 to identify early signals of community transmission of COVID-19 at the beginning of the pandemic. Focusing on calls related to key COVID-19 symptoms (cough, fever, and breathing difficulties in adults), we analyze their frequency and distribution across referral categories, comparing them to 2019 data. We employ an explainable time series anomaly detection algorithm using daily call data to identify the first collective anomalies across regions. The results show that anomalies in call data were correlated with, but preceded, the first confirmed case infected in Sweden by a median of 7 days (IQR: 2.5–10.5) and the first hospitalized case infected in Sweden by a median of 13 days (IQR: 7.25–16). They also preceded the estimated onset of community spread, indicated by the absolute confirmed cases (median: 24.5, IQR: 18.25-32.5), and severe outcomes defined by hospitalizations (median: 33, IQR: 27.25-44). These findings showcase how helpline call monitoring, using time series anomaly detection, can aid early outbreak detection.</p>}},
  author       = {{Hashemi, Atiye Sadat and Dietler, Dominik and Fall, Tove and Inghammar, Malin and Johansson, Anders F and Bonander, Carl and Ohlsson, Mattias and Björk, Jonas}},
  issn         = {{2045-2322}},
  keywords     = {{Humans; Sweden/epidemiology; COVID-19/epidemiology; Early Diagnosis; SARS-CoV-2/isolation & purification; Telemedicine; Hotlines; Adult; Algorithms; Pandemics}},
  language     = {{eng}},
  number       = {{1}},
  publisher    = {{Nature Publishing Group}},
  series       = {{Scientific Reports}},
  title        = {{Time series anomaly detection in helpline call trends for early detection of COVID-19 spread across Sweden, 2020}},
  url          = {{http://dx.doi.org/10.1038/s41598-025-20641-2}},
  doi          = {{10.1038/s41598-025-20641-2}},
  volume       = {{15}},
  year         = {{2025}},
}