A note on how public transport announcements affect the spread of coronavirus

doi:10.21203/rs.3.rs-2187111/v1

Download PDF

Research Article

A note on how public transport announcements affect the spread of coronavirus

https://doi.org/10.21203/rs.3.rs-2187111/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

We investigate the relationship between coronavirus-preventive announcements in public transport hubs and the spread of the coronavirus in Vienna, Austria. Due to its frequent high population densities, the role of public transport is a crucial topic in the field of public health, especially during times of pandemic. Using structural equation modeling, we analyze Google mobility data taken from February to December 2020 and find that changes in the contents of the announcements can influence the pandemic crisis, via their effects on public transport mobility. Interestingly, we find that different announcements affect public transport mobility differently. The impact of these announcements is maintained even after taking the effect of the contemporary government regulations into account. We therefore argue that, in order to fight the spread of the virus, it is important for policymakers to inform the general public about coronavirus-preventive behavior in an appropriate manner. We further suggest that governments take advantage of the large daily flow of people through public transport infrastructure, to disseminate vital information during crises.

Public transport mobility

Public transport announcements

Government policy

Covid-19 preventive strategies

It is widely recognized that ensuring social distancing between people (Greenstone and Nigam 2020) and mask-wearing (Eikenberry et al. 2020; Ngonghala et al. 2020) are the main factors in preventing the spread of the coronavirus. Due to its associated high population densities, the role of transport in public health has attracted massive attention (Buja et al. 2020; Heiler et al. 2020). In Austria, to fight against the coronavirus pandemic, from July 2020 until September 2022 the main Viennese public transport company, Wiener Linien, introduced seven different mask-wearing-related announcements on all public transport and in each station. However, changes in the content of announcements did not happen simultaneously with government regulation updates (see the Austrian government regulations in the Appendix, Table A1). This research note examines the impact of public transport announcements on the spreading of the coronavirus in Vienna in 2020.

During the coronavirus pandemic, policymakers have developed messages to encourage people to take preventive measures. Rooted in behavioral economics, an increasing number of studies show that nudge-based messaging promotes voluntary preventive behaviors (Lunn et al. 2020; Sasaki et al. 2021). However, some studies argue that nudge-based messages increase people’s intention to engage in coronavirus-prevention behavior, but do not necessarily change individual behavior (Falco and Zaccagni 2020; Barari et al. 2020; Everett et al. 2020). Focusing on the clarity of information, Proulx and Sime (2006) conducted a field experiment and found that the clearer the information, the more significantly individuals’ behaviors are affected. In particular, they showed that in the case of a simulated fire in an underground station, an announcement that informs the public of the existence of the fire leads to a quicker evacuation than by using a fire alarm only. Building on these previous works, we use Google mobility data to investigate the correlations between coronavirus-preventive announcements and the spread of coronavirus in Vienna in 2020. We find that changes in the contents of mask-wearing-related announcements affected the number of new coronavirus cases via its impact on public transport mobility. Such impact is sustained even after taking the effect of the contemporary government regulations into consideration.

The remainder of our note is organized as follows: the next section reports details of our data. Section 3 reports empirical analysis. In Section 4, we discuss the relationships between the coronavirus-preventive announcements, public transport mobility, and the pandemic crisis.

We draw the data for the number of new coronavirus infections from the “COVID-19: Timeline of data on Covid-19 cases per province” dataset, which was collected by the Federal Ministry for Social Affairs, Health, Care and Consumer Protection of Austria (2020). We collected the mobility data on public transport from a recent Google mobility report (Google 2022). The Google mobility report contains a series of data that reports changes in the movement of communities, compared with a baseline, across different geographical locations during the pandemic. The baseline value is the “movement median value, for the corresponding day of the week, during the 5-week period Jan 3–Feb 6, 2020” (Google 2022, p.3). The Covid-19-related regulation data was collected from the Austrian Federal Law Gazette, and it was coded into dummy variables government regulation 1 and government regulation 2. The announcement information was collected manually by our group members, and it is coded into dummy variables announcements 1, 2, and 3. We also collect the average temperature of the day as the control variable. Appendix Table A2 reports the descriptive statistics of our data.

According to the timeline of implementation of government regulations and public transport announcements, we analyze our data in six different phases (see Fig. 1 below).

We first run the analysis of variance (ANOVA) and find that the effects of different announcements on public transport mobility are heterogeneous. These heterogeneities are statistically significant at the 5% level. Next, we use structural equation modeling (SEM) to investigate the relationship between announcements, government regulations, public transport mobility, and the spread of coronavirus (see Appendix, Fig. A1 and A2).

During Phase II, the Viennese government imposed the first mask-wearing-related regulation, but no announcement on public transport was implemented. Table 1, model (1) shows the SEM results for Phase II: we find that regulation 1 had no effect on public transport mobility, but it decreased the number of new cases directly. During Phase III, Wiener Linien imposed the first mask-wearing-related announcement on all public transport and in all stations; meanwhile, regulation 1 remained in place. In Phase III, we find that announcement 1 was negatively correlated with transport mobility, and this result is statistically significant at the 1% level. Yet, in Phase III, government regulation 1 had no direct impact on the number of new coronavirus cases (see Table 1, model (2)).

Table 1

Output of the structural models
Structural models
	(1)	(2)	(3)	(4)	(5)
Dependent variable: Number of new coronavirus infection cases
Public transport mobility	23.187*** (6.684)	25.218*** (5.021)	16.520*** (4.302)	88.142** (33.143)	54.771** (17.654)
Temperature	-37.701*** (6.144)	-39.896*** (8.042)	-30.491*** (6.553)	-107.933** (40.202)	-64.894** (22.590)
Regulation 1	-142.272* (85.274)	-146.422 (89.757)	-128.647* (69.929)	-275.002 (282.109)
Regulation 2					520.750**
					(160.465)
Dependent variable: Public transport mobility
Number of new coronavirus cases	-0.086*** (0.031)	-0.096*** (0.022)	-0.054*** (0.018)	-1.114 (1.060)	-0.730 (0.527)
Regulation 1	-6.081 (8.503)	10.413** (4.892)	-2.446 (5.431)	3.079 (32.091)
Regulation 2					48.053
					(59.429)
Announcement 1		-35.627*** (7.375)
Announcement 2			13.387*** (3.724)
Announcement 3				591.530 (560.875)	352.918 (249.570)
Chi-square (p-value)	.	0.928	0.807	.	.
RMSEA	0.000	0.000	0.000	.	.
SRMR	0.000	0.001	0.003	0.024	0.033
CFI	1.000	1.000	1.000	1.000	1.000
NNFI	1.000	1.037	1.047	.	.
p-values in parentheses. ^* p < 0.1, ^ p < 0.05, ^* p < 0.01. Note: The model fit indexes include: chi-square (p > 0.05); Root Mean Square Error of Approximation (RMSEA) (RMSEA < 0.05 or < 0.08); Standardized Root Mean Square Residual (SRMR) (SRMR < 0.05); Comparative Fit Index (CFI) (CFI > 0.90 or > 0.95); and Non-Normed Fit Index (NNFI) (NNFI > 0.90 or > 0.95).

In Phase IV, the content of the announcement changed to announcement 2 while the government regulation stayed the same. We find that announcement 2 increased the mobility on public transport, whereas government regulation 1 reduced the number of new coronavirus infections. However, the government regulation did not affect the mobility on public transport (see Table 1, model (3)). In Phase V, the announcement on public transport switched to announcement 3 while the government regulation stayed at regulation 1. During this phase (V), both the government regulation and the announcement had no influence on the mobility on public transport. Moreover, government regulation 1 failed to lower the number of new coronavirus cases (see Table 1, model (4)). Finally, in Phase VI, announcement 3 remained, whilst the government regulation changed to regulation 2. According to Table 1, model (5), announcement 3 did not have a statistically significant impact on public transport mobility. However, there was a positive correlation between regulation 2 and the number of new coronavirus infection cases. According to the WHO, it can take up to 14 days for an individual to show Covid-19 symptoms (WHO 2020). Hence, it can be argued that there exists a time-lag between new infections and changes in public transport mobility. Adding a lag term of the dependent variable into the structural models, we find that our results are consistent and robust (see Appendix, Table A3).

We investigate the relationships between coronavirus-preventive announcements, government regulations, public transport mobility, and the spread of Covid-19 in Vienna. The results show that the mobility on public transport is strongly correlated with the daily increases in coronavirus cases. We also find that mask-wearing-related underground announcements can affect public transport mobility in Vienna. Both of these effects are statistically significant even if we take into account the impact of government regulations. Taken as a whole, these results suggest that coronavirus-preventive announcements can indirectly influence the spread of the coronavirus.

From 9th April until 31st October 2020, the government regulation remained the same while the contents of announcements changed three times. Interestingly, we find that these different announcements influenced public transport mobility differently. The first announcement lowered public transport mobility. The second announcement increased the mobility on public transport; compared with the first, this announcement had an extra notice that passengers should cover their noses when wearing masks. We argue that this extra reminder may have encouraged correct mask-wearing behavior. The third announcement did not affect the mobility on public transport. Based on our findings, we suggest that the manner in which coronavirus-prevention behavior is communicated to the general public is crucial.

Although we find a positive correlation between the second government regulation and the number of new coronavirus cases in Phase VI, we argue that the second regulation may not have exacerbated the spread of the coronavirus. It rather reflects the dramatic surge in mobility during the relaxation of the lockdown for Christmas shopping (Pollak et al. 2021). Despite the fact that we do not directly capture the effects of different announcements on passengers’ usage of masks, our work contributes to the identification of more effective coronavirus-preventive messages.

Data availability The public transport mobility data used in this study is available to the public at Google COVID-19 Community Mobility Reports: https://www.google.com/covid19/mobility/. The dataset for the number of coronavirus cases is available to the public at Federal Ministry for Social Affairs, Health, Care and Consumer Protection of Austria (BMSGPK): https://www.data.gv.at/katalog/dataset/covid-19-zeitliche-darstellung-von-daten-zu-covid19-fallen-je-bundesland

Conflicts of interest The authors declare that they have no conflicts of interest.

Acknowledgement The authors would like to thank Kexin Wang and Professor Oliver Fabel for useful discussions.

Authors’ contributions Study concept and design: Heluo; Data collection: Heluo; Data analysis: Heluo and Robson; Preparation of draft manuscript: Heluo and Robson. All authors reviewed the results and approved the final version of the script.

Yuxi Heluo is a doctoral student in the Department of Business Decisions and Analytics at University of Vienna, Austria. She earned her master’s degree from Heriot-Watt University, UK in 2019. Her research interests include behavioral economics, personnel economics and macroeconomic policies.

Dr. Charles W. Robson is a visiting researcher in the Physics Unit at Tampere University, Finland. He earned his PhD degree in Physics from Heriot-Watt University, UK in 2019. His research interests include nonlinear systems and gravitation.

Barari, S., Caria, S., Davola, A., Falco, P., Fetzer, T., Fiorin, S., Hensel, L., Ivchenko, A., Jachimowicz, J., King, G., Kraft-Todd, G., Ledda, A., MacLennan, M., Mutoi, L., Pagani, C., Reutskaja, E., Roth, C., Slepoi, F.R.: Evaluating COVID-19 public health messaging in Italy: Self-reported compliance and growing mental health concerns. MedRxiv. (2020). https://doi.org/10.1101/2020.03.27.20042820
Buja, A., Paganini, M., Cocchio, S., Scioni, M., Rebba, V., Baldo, V.: Demographic and socio-economic factors, and healthcare resource indicators associated with the rapid spread of COVID-19 in Northern Italy: An ecological study. PLoS One. 15(12) (2020). https://doi.org/10.1371/journal.pone.0244535
Eikenberry, S.E., Mancuso, M., Iboi, E., Phan, T., Eikenberry, K., Kuang, Y., Kostelich, E., Gumel, A.B.: To mask or not to mask: Modeling the potential for face mask use by the general public to curtail the COVID-19 pandemic. Infect. disease modelling. 5, 293–308 (2020). https://doi.org/10.1016/j.idm.2020.04.001
Everett, J.A., Colombatto, C., Chituc, V., Brady, W.J., Crockett, M.: The effectiveness of moral messages on public health behavioral intentions during the COVID-19 pandemic. PsyArXiv Preprints. (2020). https://doi.org/10.31234/osf.io/9yqs8
Falco, P., Zaccagni, S.: Promoting social distancing in a pandemic: Beyond good intentions. PloS one. 16(12) (2021). https://doi.org/10.1371/journal.pone.0260457
Federal Ministry for Social Affairs, Health, Care and Consumer Protection of Austria (BMSGPK):, “COVID-19: Timeline of data on Covid19 cases per province,” (2020). https://www.data.gv.at/katalog/dataset/covid-19-zeitliche-darstellung-von-daten-zu-covid19-fallen-je-bundesland
Google: COVID-19 Community Mobility Reports: Vienna mobility changes. Google. (2022). https://www.google.com/covid19/mobility/ Accessed 13 October 2022
Greenstone, M., Nigam, V.: Does social distancing matter? University of Chicago, Becker Friedman Institute for economics working paper No. 2020-26. SSRN Electronic J. (2020). http://dx.doi.org/10.2139/ssrn.3561244
Heiler, G., Reisch, T., Hurt, J., Forghani, M., Omani, A., Hanbury, A., Karimipour, F.: Country-Wide Mobility Changes Observed Using Mobile Phone Data during COVID-19 Pandemic. In IEEE International Conference on Big Data. IEEE 2021 (2021). (2020). http://dx.doi.org/10.1109/BigData50022.2020.9378374
Lunn, P.D., Timmons, S., Belton, C.A., Barjaková, M., Julienne, H., Lavin, C.: Motivating social distancing during the Covid-19 pandemic: An online experiment. Social Sci. Med. 265, 113478 (2020). http://dx.doi.org/10.1016/j.socscimed.2020.113478
Pollak, M., Kowarz, N., Partheymüller, J.: Chronology of the Corona Crisis in Austria - Part 4: Lockdowns, mass testing and the launch of the vaccination campaign. Corona-Blog. (2021). https://viecer.univie.ac.at/en/projects-and-cooperations/austrian-corona-panel-project/corona-blog/corona-blog-beitraege/blog100-en/ Accessed 13 October 2022
Ngonghala, C.N., Iboi, E., Eikenberry, S., Scotch, M., MacIntyre, C.R., Bonds, M.H., Gumel, A.B.: Mathematical assessment of the impact of non-pharmaceutical interventions on curtailing the 2019 novel Coronavirus. Math. Biosci. 325 (2020). http://dx.doi.org/10.1016/j.mbs.2020.108364
Proulx, G., Sime, J.D.: To prevent ‘panic’ in an underground emergency: why not tell people the truth? In Fire Safety Science—Proceedings of the Third International Symposium, pp. 843–852. Routledge (2006)
Sasaki, S., Kurokawa, H., Ohtake, F.: Effective but fragile? Responses to repeated nudge-based messages for preventing the spread of COVID-19 infection. The Japanese Economic Review. 72(3), 371–408 (2021)
World Health Organization (WHO): How long does it take to develop symptoms? Coronavirus disease (COVID-19). World Health Organization. (2020). https://www.who.int/emergencies/diseases/novel-coronavirus-2019/question-and-answers-hub/q-a-detail/coronavirus-disease-covid-19 Accessed 13 October 2022

No competing interests reported.

Appendix.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

A note on how public transport announcements affect the spread of coronavirus

Status:

Version 1

Abstract

Figures

Introduction

Methods

Results

Discussion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1