Simulation model for COVID-19 prevention in nursing homes

Agent-based prevention model proves effective in real life

A detailed epidemiological model for the spread of the coronavirus, developed by scientists working at the CSH, enables optimal prevention strategies in nursing homes. Practical experience in nursing homes run by CARITAS Vienna show: the model works! 

Since the beginning of the coronavirus pandemic, residents of nursing homes have been particularly at risk. The spatial proximity to each other favors local transmission chains, and old age combined with pre-existing conditions can lead to severe courses of disease and an increased mortality rate. Political leaders and agencies responded with far-reaching measures, such as bans on visits or restricted community activities. By strictly isolating the homes, the virus was intended to be “locked out.”

“But at the same time, that meant ‘locking people in'”, says complexity researcher Jana Lasser from Graz Technological University who worked at the Hub until March 2021 and is now a member of our Associate Faculty. Through her grandparents, Jana experienced how tense the situation in nursing homes was, especially at the beginning of the pandemic. She also saw how the lack of social contacts endangers the physical and mental health of older people.

Agent-based simulation minimizes risk of infection

Almost two years after the onset of the pandemic, testing facilities and vaccinations now make it possible to control the spread of infections in nursing homes and at the same time relax isolation measures.

Especially at the beginning of a new wave of infections, the optimal combination and timing of individual measures is crucial. Here, a new simulation tool now facilitates the decision for or against individual measures.

Based on information from nursing homes run by CARITAS Vienna—an aid and social organization engaged in social work, nursing, care for people with disabilities, education, youth work and disaster relief—Jana and a team of scientists led by Peter Klimek have developed an agent-based simulation of residents and staff in a nursing home.

Agent-based means that individual entities—in this case people—have been identified and their behaviors modeled using several parameters from empirical observations. Analogous to the behavior of people in a real care home—including everyday encounters and interactions—connections are established between these units that map potential infection pathways. The interactions of many individuals make the dynamics of the system as a whole.

The simulation tool, which allows different scenarios in residential care homes to be studied and results relevant to everyday life to be obtained, is now published in the Journal of the Royal Society Interface.

Balancing safety and quality of life

The scientists developed the model after a request of CARITAS Vienna. “The Caritas—one of the largest operators of nursing homes in Austria—provided us with insights into its pandemic management,” Jana recalls. Based on this information, the scientists developed a detailed model of a nursing home, which was calibrated by using Covid-19 outbreak data of the CARITAS homes.

Two questions were at the centre of the study:

What measures can be taken to optimally protect people in residential care homes from a SARS-CoV-2 infection?

And how can measures be chosen in such a way that the quality of life of the residents is maintained?

“The virus poses a particular threat to older adults,” says Thomas Wochele-Thoma, medical director of CARITAS Vienna. “Especially at the beginning of the pandemic, it was difficult to balance the safety of those affected on the one hand and their freedom on the other. As a consequence of the strict measures, older adults were faced with fears of isolation and experiences of loneliness during this period.”

The jointly developed test strategy was enormously helpful for the pandemic management in CARITAS homes, Wochele-Thoma says. “At a time when case numbers were rising in many residential care homes across Austria, infections were the exception in the homes where we piloted and implemented the testing strategy—at a time when the vaccination was not even available.”

Scenario “all vaccinated” or “all unvaccinated”

Specifically, the work plays out two scenarios: one without vaccinations and one with a high vaccination rate in the homes.

“It is likely that there will be infectious diseases again in the future for which vaccines are not available in the early days, as was the case with Covid-19,” Jana explains. “For this scenario, our study provides a decision-making basis for organizing preventive measures, such as a testing strategy, in nursing homes.”

The simulations were strongly oriented towards everyday work and were directly relevant to CARITAS’ decision-making processes during the pandemic from autumn 2020 onwards.

Vaccination rate decisive

On the one hand, the study shows that the time gap between sample collection and the test result as well as the accuracy of Covid-19 test results have a great influence on the incidence of infection in nursing homes. Thus, in the scenario of a completely unvaccinated home population, twice-weekly PCR tests of staff and, in the case of positive test results, strict quarantine measures are sufficient to prevent large outbreaks.

In turn, with a current vaccination rate of well over 80 per cent among residents, simulation results indicate that major coronavirus disease outbreaks are no longer to be expected, even if all other measures are relaxed or discontinued. However, the prerequisite is a vaccination protection against infection of at least 60 per cent. “More research is definitely needed here to better assess the effectiveness of vaccinations for older age groups,” Jana emphasizes. It is therefore recommended that the testing infrastructure in the houses is maintained, voluntary tests carried out on a regular basis and the virus genomes sequenced so that new variants of concern can be detected at an early stage.

CSH model the basis for testing strategy in Viennese nursing homes

Based on these simulations, the CARITAS developed its own testing strategy, starting in autumn 2020, and introduced it in its homes. “The fact that nursing homes have largely vanished from public attention shows what a great job our colleagues at the CARITAS have done,” says Peter, the scientific project manager of this publication. “This work also shows that only several measures in combination lead to the desired prevention success.”

Currently, measures in the homes are being adapted to the threat assessment against the background of the high vaccination rates in such a way that the quality of life of the people is restricted as little as possible.

“This work is a wonderful example of how helpful reliable simulation tools are, especially in critical phases. The simulation tool we developed facilitates evidence-based decisions. You can play with different variables and immediately see the effects of each measure. It is desirable to use the results of our and other simulation studies for pandemic-related decisions as early as possible,” conclude the project participants.

The work Agent-based simulations for protecting nursing homes with prevention and vaccination strategies by Jana Lasser, Johannes Zuber, Johannes Sorger, Elma Dervic, Katharina Ledebur, Simon David Lindner, Elisabeth Klager, Maria Kletečka-Pulker, Harald Willschke, Katrin Stangl, Sarah Stadtmann, Christian Haslinger, Peter Klimek, Thomas Wochele-Thoma appeared in the Dec. 22 issue of the Journal of the Royal Society Interface.

Text adapted from a joint press release of TU Graz and CSH.