Loading [MathJax]/jax/output/HTML-CSS/fonts/TeX/fontdata.js

Journal List > J Korean Med Sci > v.40(2) > 1516089436

Hong and Jung: Impact of Government Healthcare Policy Changes on Consumption and Human Movements During COVID-19: An Interrupted Time Series Analysis in Korea

Abstract

Background

The coronavirus disease 2019 (COVID-19) pandemic has altered daily behavioral patterns based on government healthcare policies, including consumption and movement patterns. We aimed to examine the extent to which changes in the government's healthcare policy have affected people's lives, primarily focusing on changes in consumption and population movements.

Methods

We collected consumption data using weekly credit card transaction data from the Hana Card Corporation and population mobility data using mobile phone data from SK Telecom in Seoul, South Korea. Interrupted time-series analysis was used to calculate the relative risk ratio and perform the intervention effects when government healthcare policy changes.

Results

We found that leisure and outside movements were the most immediately affected by changes in government healthcare policies. It took over 2 years and 11 months, respectively, for these sectors to return to their pre-COVID-19 routines.

Conclusion

Enhancing healthcare policies presents advantages and disadvantages. Although such policies help prevent the spread of COVID-19, they also reduce consumption and mobility, extending the time needed to return to pre-COVID-19 levels. Government healthcare policymakers should consider not only disease prevention but also the impact of these policies on social behaviors, economic activity, and mobility.

Graphical Abstract

jkms-40-e6-abf001.jpg

INTRODUCTION

Since the coronavirus disease 2019 (COVID-19) outbreak, many countries have implemented preventive measures. These include social distancing strategies, such as city lockdowns, home quarantines, school closures, bans on mass gatherings, and travel restrictions, in response to the rising number of confirmed cases.12 Other preventive measures involve personal behavioral controls, such as isolation, handwashing, respiratory etiquette, and mask-wearing to reduce COVID-19 transmission. The combination of these interventions has significantly reduced transmission compared to single interventions.34
These preventive policies have successfully suppressed the spread of the virus; however, they have negatively impacted the economy.5 City lockdowns and home quarantines have significantly affected household consumption behavior due to permanent income shocks,6 influenced by various factors such as COVID-19-related policies and changes in consumption patterns.7 Previous studies have shown that COVID-19 has substantially and adversely affected consumption, leading to a significant global decrease in consumption volume.8 Specifically, the travel industry has experienced cancellations as people chose not to travel abroad during cross-border lockdowns.9 Concurrently, the hospitality industry also saw a sharp decline in consumption spending during the lockdown.10
Human movement is crucial in the transmission dynamics of infectious diseases, including COVID-19. To limit the spread of the virus, policymakers should restrict mobility within and between different areas.11 Mobility data can serve as a valuable tool for implementing data-driven preventive policies. With precise location tracking through GPS, mobile phone data can be used to monitor human movement patterns.12 Research on mobility data during the COVID-19 pandemic has primarily focused on public transportation systems, such as buses, subways, and air travel.13 Various countries, including China and Italy, have utilized smartphone tracking data to restrict population mobility, resulting in delays in the epidemic peak and reductions in virus transmission rates.1415 In the United States, measures such as stay-at-home orders and restrictions on restaurant and bar operations significantly reduced human mobility during the early stages of the pandemic.16
The successful implementation of social distancing, quarantine, and mask-wearing policies in South Korea during the COVID-19 pandemic stands as a testament to effective pandemic response.17 Unlike in other countries, South Korea's phased approach—based on the number of confirmed COVID-19 cases, vaccination rates, and the basic reproductive number (R0) of COVID-19—has set a high standard. When the number of confirmed COVID-19 cases and deaths surged, the South Korean government swiftly imposed stringent social distancing measures. For instance, polymerase chain reaction tests were conducted pre-emptively for high-risk groups, such as older adults and children, and large gatherings in facilities were banned in the metropolitan area. The public's strict adherence to these measures has significantly impacted various economic aspects and human movements, including health, education, and the environment.
As COVID-19 shifted from a pandemic to an endemic, previously implemented preventive policies were relaxed, leading to a gradual recovery of the economy and various industries affected by the virus. In the United States, a study estimated the economy's recovery time based on Gross Domestic Product under different scenarios involving social distancing measures and the spread of COVID-19. The study suggested that if a new COVID-19 variant emerged after the relaxation of social distancing policies and new prevention measures were reintroduced, the economy would experience a W-shaped recovery.18 In South Korea, research examined the macroeconomic impact of pandemic shocks using three scenarios and employed a weighted time-series forecasting model to analyze the economy's forecasting and recovery process from the short-term shock. Notably, sectors such as agriculture, culture, and tourism, which suffered significant losses due to the COVID-19 impact, demonstrated lower resilience compared to other sectors.19
Although many studies have explored the economic impacts of the COVID-19 pandemic by assessing damages across various industries and shifts in consumer behavior, there is a notable gap in research regarding the effects of healthcare policy changes. This study aims to analyze how government healthcare policy changes during the pandemic have influenced consumer spending and human mobility and to assess the recovery time to pre-COVID-19 levels in Seoul, South Korea. The findings will contribute to academic understanding of the topic and provide practical insights for policymakers and public health professionals on managing future pandemics.

METHODS

Data

We collected consumption data using weekly credit card transactions and human mobility data using mobile phone records from Hana Card Corporation and SK Telecom in Seoul, South Korea, respectively, between January 1, 2019, and August 31, 2023. The Hana Card Corporation provided more consumption data than other credit card companies through its financial platforms. SK Telecom is the telecommunications company with the highest number of subscribers in South Korea. The consumption data were based on the frequency of weekly credit card payments, encompassing categories such as education, food, health, leisure, shopping, and travel. Human mobility data were based on the frequency of weekly movements inside and outside the region where people live. Specifically, the inside and outside movements were defined based on whether participants visited other areas outside their residences for at least 30 minutes. We selected 12 interventions for interrupted time series (ITS) analysis to evaluate the impact of significant healthcare policies on minimizing the spread of COVID-19 in South Korea. During the study period, South Korea experienced five waves of COVID-19, with the highest peak in confirmed COVID-19 cases occurring within this timeframe.

The government policies

The South Korean government has implemented several significant healthcare policies during the COVID-19 pandemic. The 12 significant healthcare policies implemented over time and chosen for this study are as follows:
1. The first implementation of social distancing: involves reducing interactions within communities with uninfected individuals to reduce the risk of COVID-19 transmission.
2. Implemented three stages of social distancing: restriction of facilities and cancellation of events to avoid mass gatherings, based on the daily number of confirmed COVID-19.
3. Elevating one stage of social distancing: temporarily prohibiting all activities like lockdown.
4. Mitigating one stage of social distancing in the Seoul metropolitan area: unnecessarily refraining from leaving home and attending meetings.
5. A ban on private groups of five or more people in the Seoul metropolitan area: the closure of schools, workplaces, and certain businesses to reduce COVID-19 transmission.
6. Implemented a special quarantine policy in the Seoul metropolitan area: highly restricting the number of individuals or groups depending on vaccination and quarantine.
7. Implemented new social distancing from three to four stages: prohibited nightlife groups and operated multi-use facilities such as theater, pubs, and bars until 22:00.
8. Implemented a phased recovery of daily life: operating 24 h a day in all facilities, except entertainment outlets, regardless of vaccination status.
9. Implemented a new national special quarantine policy: temporarily re-restricted the number of individuals or groups and strengthened the completion of vaccinations.
10. Adjusted quarantine period for confirmed patients of COVID-19 from 14 to 7 days: people who were fully vaccinated based on self-participation and responsibility corresponded, and social distancing policies changed from mandatory to recommended.
11. Lifting social distancing: relaxing or ending social distancing measures while maintaining vaccination.
12. The mandate for wearing masks outdoors was released: wearing masks was left to individual autonomy.
Fig. 1 shows the timeline of healthcare policy changes during the COVID-19 pandemic, presented in a time series format.
Fig. 1

Timeline of healthcare policy changes in South Korea.

COVID-19 = coronavirus disease 2019.
jkms-40-e6-g001

Download Figure

Analytical strategy

An ITS is a statistical method for evaluating intervention or interruption settings in which randomization is not feasible.18 The time series refers to the actual data over the period, whereas the intervention is used to analyze the impact of changes on people’s behavior. Thus, data for the ITS analysis were collected at multiple points in the time series before and after the intervention. Additionally, ITS analysis involves an autoregressive form of segmented regression analysis of the intervention.20 In this study, the formula used for ITS analysis was as follows:
Yt = β0 + β1Timet + β2Interventiont(tT) + β3Time after Interventiont + εt
In the above formula, Yt defines the number of weekly consumptions, including education, food, health, leisure, shopping, and trips, based on the use of credit card data and population movements as mobility, including inside and outside zones based on people’s residential areas using mobile data. Timet is a dummy variable representing the intervention such that pre-intervention = 0 and post-intervention = 1. Time after Interventiont is an interaction term. β0 represents the intercept of the outcome variable (Yt). In the specification of the model in which we have centred t about T in the interaction term. β1 is the slope of the outcome variable until the point of intervention at time t = T. β2 represents the change in the level of the outcome variable that immediately occurs following the point of intervention, including the cumulative COVID-19 vaccination rate. β3 is the difference between the pre-intervention and post-intervention slopes of the outcome variable. Finally, εt is the error terms following a first-order autoregressive process. Thus, we find significant P values in β2 as the intervention effect and β3 indicates the intervention effect over time.
We obtained parameters from the ITS formula and calculated the relative risk (RR) ratios. The RR was calculated as an exponential parameter, and a 95% confidence interval (CI) was used as follows:
Exp (β − 1.96 × Standard Error) − Exp (β + 1.96 × Standard Error)
Statistical analyses were conducted using the R Statistical Software (version 4.3.1; R and R Studio Foundation for Statistical Computing, Vienna, Austria). The results are presented as RR with 95% CI. Statistical P value < 0.05 was considered significant.

Ethics statement

The datasets utilized in this study were de-identified and purchased from a publicly disclosed data platform. Therefore, no approval from the Institutional Review Board was deemed necessary. All studies were carried out based on the Declaration of Helsinki.

RESULTS

From January 1, 2020, to August 31, 2023, the South Korean government implemented 12 social distancing measures to reduce the spread of COVID-19 (Fig. 1). The average daily confirmed COVID-19 cases were 26,191, and the highest daily confirmed COVID-19 cases were 621,035 on March 17, 2022. During the study period, South Koreans spent 2,836,729 ($10,000) on average weekly consumption, including food, health, leisure, shopping, trips, and education. The average weekly mobility was 22,423,981, including inside and outside movements. The study period also covered the fifth wave of the COVID-19 pandemic in South Korea. The first wave was in March 2020, and the second was in September 2020. The third, fourth, and fifth waves occurred in January 2021, December 2021, and March 2022, respectively.

Interrupted time series on consumption and human movement

We used ITS analysis to examine the impact of interventions on consumption levels and human mobility rates, as well as the changes in healthcare policies. Consumption was categorized into six groups, whereas human mobility was divided into two categories. Table 1 presents the ITS results for consumption, focusing on changes in healthcare policies, including the cumulative COVID-19 vaccination rate since February 2021. Our leisure data revealed that, except for the 12th healthcare policy, which did not mandate wearing masks outdoors, most healthcare policies had significant intervention effects, with protective effects (1 – RR) ranging from 41.0% to 57.0%. Specifically, the sixth healthcare policy, which implemented a special quarantine in the Seoul metropolitan area, resulted in a 57.0% decrease in leisure consumption, with effects sustained over time. This policy was the most robust, restricting group sizes based on vaccination and quarantine status. Following this policy, it took over 2 years to return to pre-COVID-19 routines. This strict policy significantly affected activities such as shopping, traveling, dining out, and healthcare services, with reductions of approximately 28.2%, 30.4%, 37.4%, and 44.3%, respectively, in the Seoul metropolitan area. These sectors took an average of 1 to 2 months to recover. However, the healthcare policies did not significantly impact the education sector (Fig. 2). In Table 2, the results of the ITS analysis for human mobility indicate that most healthcare policies had significant effects on inside and outside movements. The fifth healthcare policy, which involved the closure of schools and workplaces, and reduced operating hours for public venues such as restaurants, pubs, and bars, resulted in an 11.0% decrease in inside movement and a 39.1% decrease in outside movement (Fig. 3). It took an average of 3 months for inside movement and 11 months for outside movement to return to pre-COVID-19 levels. However, the effects of these interventions were not sustained over time, suggesting that healthcare policies have a greater impact on outside movements than inside movements. Further details on the effects of these interventions can be found in the Supplementary Figs. 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22.
Fig. 2

High intervention effects of consumption on the sixth healthcare policy (2020–2023) in South Korea. (A) Leisure. (B) Shopping. (C) Health. (D) Food. (E) Trips. (F) Education.

The red dotted line indicates the starting point of the sixth healthcare policy to prevent COVID-19. The blue line represented the interrupted time series before and after the red dotted line.
jkms-40-e6-g002

Download Figure

Fig. 3

High intervention effects of human mobility on the fifth healthcare policy (2019–2023) in South Korea. (A) Inside movement. (B) Outside movement.

The red dotted line indicates the starting point of the fifth healthcare policy to prevent COVID-19. The blue line represented the interrupted time series before and after the red dotted line.
jkms-40-e6-g003

Download Figure

Table 1

Consumption of interrupted time series results

jkms-40-e6-i001
Intervention points Changes in healthcare policies
1 2 3 4 5 6 7 8 9 10 11 12
Coeff. RR (95% CI) Coeff. RR (95% CI) Coeff. RR (95% CI) Coeff. RR (95% CI) Coeff. RR (95% CI) Coeff. RR (95% CI) Coeff. RR (95% CI) Coeff. RR (95% CI) Coeff. RR (95% CI) Coeff. RR (95% CI) Coeff. RR (95% CI) Coeff. RR (95% CI)
Leisure
Intercept β 0 12.059 - 11.878 - 11.858 - 11.771 - 11.764 - 11.734 - 11.715 - 11.645 - 11.641 - 11.624 - 11.579 - 11.585 -
Baseline trend β 1 −0.051 0.950 (0.867–1.042) −0.021 0.979 (0.961–0.997) −0.020 0.981 (0.968–0.994) −0.012 0.989 (0.978–0.999) −0.011 0.989 (0.982–0.995) −0.009 0.991 (0.986–0.996) −0.009 0.991 (0.988–0.995) −0.006 0.994 (0.991–0.997) −0.006 0.994 (0.992–0.997) −0.005 0.995 (0.992–0.997) −0.004 0.996 (0.994–0.998) −0.004 0.996 (0.994–0.998)
Level change β 2 −0.813 0.444 (0.281–0.701) −0.743 0.476 (0.356–0.636) −0.747 0.474 (0.361–0.623) −0.730 0.482 (0.369–0.629) −0.784 0.456 (0.352–0.592) −0.845 0.430 (0.326–0.566) −0.803 0.448 (0.297–0.676) −1.219 0.496 (0.343–0.613) −1.325 0.466 (0.341–0.501) −1.169 0.511 (0.367–0.677) −1.238 0.590 (0.350–0.862) −1.117 0.627 (0.367–0.939)
Trend change β 3 0.052 1.054 (0.961–1.155) 0.023 1.024 (1.005–1.043) 0.022 1.022 (1.009–1.035) 0.014 1.014 (1.003–1.025) 0.014 1.014 (1.007–1.021) 0.012 1.013 (1.007–1.018) 0.012 1.012 (1.008–1.016) 0.012 1.012 (1.008–1.016) 0.012 1.012 (1.008–1.017) 0.012 1.012 (1.007–1.016) 0.012 1.012 (1.007–1.016) 0.011 1.006 (1.006–1.016)
Shopping
Intercept β 0 13.489 - 13.486 - 13.515 - 13.461 - 13.494 - 13.512 - 13.588 - 13.564 - 13.561 - 13.557 - 13.540 - 13.550 -
Baseline trend β 1 0.000 1.000 (0.902–1.109) 0.002 1.002 (0.984–1.021) −0.001 0.999 (0.986–1.012) 0.003 1.003 (0.993–1.014) 0.001 1.001 (0.995–1.007) 0.000 1.000 (0.995–1.005) −0.003 0.997 (0.993–1.000) −0.003 0.997 (0.995–1.000) −0.002 0.998 (0.995–1.000) −0.002 0.998 (0.996–1.000) −0.002 0.998 (0.996–1.000) −0.002 0.998 (0.996–1.000)
Level change β 2 −0.036 0.965 (0.567–1.642) −0.070 0.932 (0.690–1.260) −0.104 0.901 (0.685–1.187) −0.097 0.907 (0.695–1.183) −0.205 0.815 (0.634–1.047) −0.331 0.718 (0.554–0.931) −0.282 0.754 (0.520–1.093) −0.077 0.725 (0.469–1.025) −0.475 0.722 (0.344–1.125) −0.440 0.744 (0.459–1.156) −0.509 0.801 (0.520–1.129) −0.509 0.801 (0.521–1.130)
Trend change β 3 −0.001 0.999 (0.901–1.108) −0.002 0.998 (0.980–1.017) 0.001 1.001 (0.989–1.014) −0.003 0.997 (0.987–1.008) 0.000 1.000 (0.994–1.006) 0.001 1.001 (0.996–1.006) 0.004 1.004 (1.001–1.008) 0.003 1.003 (0.999–1.007) 0.004 1.004 (1.000–1.008) 0.004 1.004 (1.000–1.008) 0.004 1.004 (1.000–1.009) 0.004 1.004 (1.001–1.009)
Trips
Intercept β 0 13.955 - 13.865 - 13.844 - 13.774 - 13.770 - 13.774 - 13.783 - 13.746 - 13.741 - 13.737 - 13.714 - 13.724 -
Baseline trend β 1 −0.016 0.984 (0.881–1.098) −0.009 0.991 (0.970–1.012) −0.007 0.993 (0.978–1.008) −0.001 0.999 (0.987–1.011) −0.001 0.999 (0.992–1.006) −0.001 0.999 (0.993–1.004) −0.002 0.998 (0.995–1.002) 0.000 1.000 (0.997–1.002) 0.000 1.000 (0.997–1.002) 0.000 1.000 (0.997–1.002) 0.000 1.000 (0.998–1.002) 0.000 1.000 (0.998–1.002)
Level change β 2 −0.387 0.679 (0.388–1.188) −0.323 0.724 (0.518–1.013) −0.306 0.736 (0.541–1.002) −0.276 0.758 (0.562–1.023) −0.308 0.735 (0.555–0.973) −0.363 0.696 (0.522–0.927) −0.321 0.725 (0.483–1.091) −0.358 0.699 (0.333–1.166) −0.546 0.697 (0.305–1.102) −0.426 0.753 (0.549–1.220) −0.426 0.753 (0.549–1.220) −0.298 0.742 (0.580–1.450)
Trend change β 3 0.019 1.019 (0.913–1.138) 0.012 1.012 (0.991–1.034) 0.010 1.010 (0.995–1.025) 0.004 1.004 (0.992–1.017) 0.005 1.005 (0.997–1.012) 0.004 1.004 (0.999–1.010) 0.005 1.005 (1.001–1.009) 0.004 1.004 (1.000–1.009) 0.004 1.005 (1.001–1.010) 0.005 1.005 (1.000–1.009) 0.005 1.005 (1.000–1.009) 0.004 1.004 (0.999–1.009)
Food
Intercept β 0 13.721 - 13.681 - 13.704 - 13.644 - 13.697 - 13.717 - 13.744 - 13.695 - 13.691 - 13.684 - 13.657 - 13.666 -
Baseline trend β 1 −0.006 0.994 (0.897–1.101) −0.001 0.999 (0.980–1.017) −0.004 0.996 (0.983–1.009) 0.001 1.001 (0.991–1.012) −0.003 0.997 (0.991–1.004) −0.004 0.996 (0.992–1.001) −0.005 0.995 (0.992–0.998) −0.003 0.997 (0.994–0.999) −0.003 0.997 (0.994–0.999) −0.003 0.997 (0.995–0.999) −0.002 0.998 (0.996–1.000) −0.003 0.997 (0.996–0.999)
Level change β 2 −0.200 0.819 (0.485–1.384) −0.227 0.797 (0.588–1.081) −0.268 0.765 (0.579–1.012) −0.270 0.764 (0.583–1.000) −0.379 0.684 (0.529–0.885) −0.469 0.626 (0.478–0.819) −0.369 0.691 (0.470–1.016) −0.581 0.660 (0.278–1.127) −0.588 0.655 (0.447–0.839) −0.701 0.696 (0.471–0.908) −0.801 0.649 (0.324–0.861) −0.686 0.704 (0.360–0.977)
Trend change β 3 0.006 1.006 (0.909–1.115) 0.002 1.002 (0.983–1.021) 0.004 1.004 (0.992–1.018) 0.000 1.000 (0.989–1.010) 0.004 1.004 (0.998–1.011) 0.005 1.005 (1.000–1.010) 0.006 1.006 (1.003–1.010) 0.006 1.006 (1.002–1.010) 0.007 1.007 (1.003–1.011) 0.007 1.007 (1.002–1.011) 0.007 1.007 (1.002–1.012) 0.006 1.006 (1.002–1.011)
Health
Intercept β 0 12.011 - 11.984 - 12.066 - 12.041 - 12.080 - 12.059 - 12.099 - 12.058 - 12.054 - 12.040 - 12.001 - 12.008 -
Baseline trend β 1 −0.005 0.995 (0.890–1.112) 0.004 1.004 (0.985–1.024) −0.004 0.996 (0.982–1.009) −0.002 0.998 (0.987–1.009) −0.005 0.995 (0.988–1.002) −0.004 0.996 (0.991–1.001) −0.006 0.994 (0.990–0.998) −0.005 0.995 (0.993–0.998) −0.004 0.996 (0.993–0.998) −0.004 0.996 (0.994–0.998) −0.003 0.997 (0.995–0.999) −0.003 0.997 (0.995–0.999)
Level change β 2 −0.177 0.838 (0.473–1.484) −0.275 0.759 (0.551–1.046) −0.374 0.688 (0.513–0.992) −0.394 0.674 (0.508–0.895) −0.479 0.619 (0.472–0.813) −0.585 0.557 (0.418–0.743) −0.574 0.563 (0.370–0.858) −0.500 0.606 (0.286–1.287) −0.656 0.569 (0.244–0.902) −0.651 0.621 (0.275–0.987) −0.636 0.579 (0.241–0.952) −0.616 0.560 (0.269–1.083)
Trend change β 3 0.005 1.005 (0.899–1.124) −0.003 0.997 (0.978–1.016) 0.005 1.005 (0.992–1.019) 0.003 1.003 (0.992–1.015) 0.007 1.007 (1.000–1.014) 0.006 1.006 (1.000–1.011) 0.008 1.008 (1.004–1.012) 0.007 1.007 (1.003–1.011) 0.008 1.008 (1.003–1.012) 0.007 1.007 (1.003–1.012) 0.007 1.007 (1.002–1.012) 0.007 1.007 (1.002–1.012)
Education
Intercept β 0 10.157 - 10.077 - 10.147 - 10.081 - 10.103 - 10.103 - 10.168 - 10.141 - 10.151 - 10.140 - 10.132 - 10.155 -
Baseline trend β 1 −0.021 0.979 (0.760–1.262) 0.007 1.007 (0.965–1.052) −0.001 0.999 (0.969–1.030) 0.005 1.005 (0.980–1.030) 0.002 1.002 (0.987–1.017) 0.002 1.002 (0.991–1.013) −0.001 0.999 (0.992–1.007) 0.000 1.000 (0.995–1.006) 0.000 1.000 (0.994–1.005) 0.000 1.000 (0.995–1.005) 0.000 1.000 (0.996–1.005) 0.000 1.000 (0.996–1.004)
Level change β 2 0.001 1.001 (0.279–3.592) 0.050 1.051 (0.514–2.150) 0.007 1.007 (0.526–1.928) 0.039 1.040 (0.554–1.952) 0.018 1.018 (0.563–1.840) −0.044 0.957 (0.520–1.761) 0.057 1.059 (0.457–2.451) 0.126 1.135 (0.239–3.379) −0.068 0.935 (0.237–2.686) −0.206 0.814 (0.209–2.166) −0.206 0.814 (0.209–2.166) −0.075 0.928 (0.208–3.131)
Trend change β 3 0.021 1.021 (0.792–1.315) −0.007 0.993 (0.951–1.037) 0.001 1.001 (0.971–1.032) −0.004 0.996 (0.971–1.021) −0.002 0.998 (0.983–1.013) −0.002 0.998 (0.986–1.010) 0.001 1.001 (0.992–1.009) 0.000 1.000 (0.990–1.009) 0.000 1.000 (0.991–1.010) 0.001 1.001 (0.991–1.011) 0.001 1.001 (0.991–1.011) 0.001 1.001 (0.990–1.011)
Coeff. = coefficient, RR = relative risk, CI = confidence interval.

Download Table

Table 2

Human mobility of interrupted time series results

jkms-40-e6-i002
Intervention points Changes in healthcare policies
1 2 3 4 5 6 7 8 9 10 11 12
Coeff. RR (95% CI) Coeff. RR (95% CI) Coeff. RR (95% CI) Coeff. RR (95% CI) Coeff. RR (95% CI) Coeff. RR (95% CI) Coeff. RR (95% CI) Coeff. RR (95% CI) Coeff. RR (95% CI) Coeff. RR (95% CI) Coeff. RR (95% CI) Coeff. RR (95% CI)
Inside
Intercept β0 16.469 - 10.476 - 16.473 - 16.476 - 16.473 - 16.477 - 16.464 - 16.458 - 16.458 - 16.456 - 16.454 - 16.453 -
Baseline trend β1 0.000 1.000 (0.999–1.000) −0.001 0.999 (0.999–1.000) −0.001 0.999 (0.999–1.000) −0.001 0.999 (0.999–1.000) −0.001 0.999 (0.999–1.000) −0.001 0.999 (0.999–0.999) 0.000 1.000 (0.999–1.000) 0.000 1.000 (1.000–1.001) 0.000 1.000 (1.000–1.000) 0.000 1.000 (1.000–1.000) 0.000 1.000 (1.000–1.000) 0.000 1.000 (1.000–1.000)
Level change β2 −0.093 0.911 (0.892–0.930) −0.104 0.902 (0.882–0.922) −0.110 0.895 (0.875–0.916) −0.110 0.896 (0.875–0.917) −0.116 0.890 (0.867–0.914) −0.098 0.907 (0.865–0.950) −0.144 0.865 (0.820–0.914) −0.118 0.889 (0.836–0.945) −0.118 0.889 (0.832–0.949) −0.092 0.912 (0.848–0.982) −0.033 0.968 (0.888–1.054) −0.029 0.971 (0.888–1.062)
Trend change β3 0.001 1.001 (1.000–1.001) 0.001 1.001 (1.001–1.002) 0.001 1.001 (1.001–1.001) 0.001 1.001 (1.001–1.001) 0.001 1.001 (1.001–1.001) 0.001 1.001 (1.001–1.002) 0.001 1.001 (1.001–1.002) 0.001 1.001 (1.001–1.001) 0.001 1.001 (1.001–1.001) 0.001 1.001 (1.000–1.001) 0.000 1.000 (1.000–1.001) 0.000 1.000 (1.000–1.001)
Outside
Intercept β0 16.080 - 16.128 - 16.122 - 16.132 - 16.128 - 16.137 - 16.107 - 16.092 - 16.090 - 16.080 - 16.075 - 16.072 -
Baseline trend β1 0.000 1.000 (0.998–1.001) −0.003 0.997 (0.996–0.998) −0.002 0.998 (0.997–0.998) −0.003 0.997 (0.996–0.998) −0.003 0.997 (0.997–0.998) −0.003 0.997 (0.997–0.997) −0.002 0.998 (0.997–0.998) −0.002 0.998 (0.998–0.998) −0.002 0.998 (0.998–0.998) −0.002 0.998 (0.998–0.999) −0.002 0.998 (0.998–0.999) −0.002 0.998 (0.998–0.999)
Level change β2 −0.346 0.708 (0.671–0.746) −0.417 0.659 (0.621–0.700) −0.452 0.636 (0.599–0.676) −0.456 0.634 (0.595–0.675) −0.495 0.609 (0.569–0.653) −0.422 0.655 (0.578–0.743) −0.544 0.580 (0.503–0.670) −0.423 0.655 (0.558–0.769) −0.418 0.659 (0.553–0.784) −0.323 0.724 (0.596–0.880) −0.016 0.984 (0.793–1.222) 0.032 1.033 (0.824–1.296)
Trend change β3 0.002 1.002 (1.001–1.003) 0.004 1.004 (1.003–1.005) 0.004 1.004 (1.003–1.005) 0.004 1.004 (1.004–1.005) 0.005 1.005 (1.004–1.005) 0.004 1.004 (1.003–1.005) 0.004 1.004 (1.004–1.005) 0.004 1.004 (1.003–1.005) 0.004 1.004 (1.002–1.005) 0.003 1.003 (1.002–1.004) 0.001 1.001 (1.000–1.002) 0.001 1.001 (1.000–1.002)
Coeff. = coefficient, RR = relative risk, CI = confidence interval.

Download Table

DISCUSSION

This study examined the impact of changes in governmental healthcare policies during the COVID-19 era on consumption and human movement in South Korea. Among the consumption categories (including shopping, trips, food, health, and education), leisure was greatly impacted by rigorous governmental healthcare policies. Conversely, they did not affect education. Compared to pre-COVID-19 levels, consumption took over 2 years to return to its previous levels. Outside movements were more affected by government healthcare policies than inside movements, taking 11 months for recovery. When the Korean government implemented its sixth special quarantine policy in the Seoul metropolitan area, Koreans felt that this healthcare policy was the strictest. Therefore, highly restricted government policies affected recovery time.
Abbas et al. 21 observed that during the COVID-19 era, the travel and leisure industries experienced a steep decline in revenue, amounting to 2.86 trillion US dollars, compared with other damaged global industries. These associations between industry and disease have been observed during the peak of various diseases, such as the SARS epidemic in 2003, seasonal influenza, and other global pandemics.22 People’s food consumption patterns changed during the COVID-19 pandemic as a result of their concern about infection and affecting supermarket employees. Asian countries observed a shift in eating out and shopping behaviors to embracing online shopping and food delivery.23 As people spend more time staying at home, the frequency of eating out is reduced, and online grocery shopping to cook for themselves increases.24 For consumption associated with health, an increasing number of people acutely infected with COVID-19 incurred higher direct medical costs than those with other common infectious diseases in the United States.25 In contrast, the South Korean government decided to pay most of the medical costs related to COVID-19 from the National Health Insurance (NHI) during the COVID-19 pandemic.26 Focusing on the confirmed patients with COVID-19, the average medical costs increased not only among the older patients owing longer hospital stays but also among the younger patients who have NHI and stay longer in the hospitals. Although the overall medical costs increased because of the NHI, personal medical costs were not as high.27 The COVID-19 pandemic affected the educational systems of several countries. Most students and their parents who experienced COVID-19 worried about being compared with those who studied before the COVID-19 pandemic.28 According to the South Korean Statistical Information Service of 2020, approximately 74.5% of all K-12 students experienced shadow education in 2019.29 Additionally, South Korea has the highest spending on education among the Organization for Economic Cooperation and Development (OECD) countries.30 Thus, even with the changes in government healthcare policies, education was not significantly impacted.
Most countries implemented strict restrictions, including lockdowns and working from home, according to the COVID-19 impact.31 However, the South Korean government has focused on social distancing policies, such as restricting private meetings with acquaintances and reducing store working hours instead of lockdowns and border closures. Koreans followed non-lockdown social distancing measures to avoid mass gatherings, facilitate contact tracing, and quarantine.3233 These results aligned with the observation that outside movements responded in a timely and appropriate manner to changes in government healthcare policies compared to inside movements, likely because of the impact of COVID-19 vaccinations, which began in March 2021, further supporting the results of the ITS analysis. The overall trend exhibits a V-shaped pattern over the analytical period, suggesting that the mid-time point of the intervention reflects the most significant changes, particularly in leisure and outside movements.
The strength of this study lies in the analysis of consumption and human mobility during the COVID-19 era based on changes in government healthcare policies in South Korea using real data such as credit card and mobile phone data. This study investigated intervention effects as a protective effect of policies on time series using ITS when government healthcare policies changed. The ITS is generally unaffected by typical confounding factors, such as socioeconomic status, which changes slowly over time. However, seasonality in consumption and human movements can be considered a time-varying confounder. Therefore, we considered seasonality in our ITS analysis.
This study had a few limitations. First, we could not account for potential time-varying confounders in the ITS analysis. Future studies should include unmeasured or unknown variables to examine intervention effects using a multiple baseline design. Second, the count data assumes that the Poisson distribution equals the variance. However, the variance tends to be greater, leading to overdispersion in the real data analysis when credit card and mobile phone data are used. Third, we could not distinguish between offline and online consumption credit card use, and all real-world data were limited to Seoul, Korea. Fourth, it is important to interpret our results carefully, considering that the increase in the COVID-19 vaccination rate may have influenced consumption and population movement. Fifth, the range of design ITS has to be equal between the pre- and post-period on the intervention point; however, early healthcare policies were restricted due to data availability issues. Lastly, the sustained transmission of severe acute respiratory syndrome coronavirus 2, driven by continuous antigenic drift of new variants and waning immunity, as well as the calculation of recovery time to pre-COVID-19 routines, may impact the results of our ITS model.
In summary, the leisure sector was significantly affected by changes in government policies, whereas the education sector was not. This study's findings contribute to the growing body of evidence suggesting that government policy changes have a more pronounced effect on outdoor activities than on indoor activities. Consequently, modifications in government healthcare policies had an immediate effect on leisure and outdoor activities, with recovery taking, on average, more than 2 years and 11 months compared to pre-COVID-19 levels. These results suggest that strengthening healthcare policies has pros and cons. Although it prevents the spread of COVID-19, it also reduces people from consuming and moving and requires more recovery time to return to the routine of pre-COVID-19.

Notes

Funding: The authors disclosed receipt of the following financial support: Jinwook Hong and Jaehun Jung received grant of the project for Infectious Disease Medical Safety, funded by the Ministry of Health & Welfare, Republic of Korea (grant number: HG22C0094).

Disclosure: The authors have no potential conflicts of interest to disclose.

Data Availability Statement: The data for this study is available for purchase from Hana Card Corporation and SK Telecom.

Author Contributions:

  • Conceptualization: Jung J.

  • Data curation: Hong J.

  • Formal analysis: Hong J.

  • Funding acquisition: Jung J, Hong J.

  • Methodology: Jung J, Hong J.

  • Project administration: Jung J.

  • Writing - original draft: Hong J.

  • Writing - review & editing: Jung J.

References

1. Yan B, Zhang X, Wu L, Zhu H, Chen B. Why do countries respond differently to COVID-19? A comparative study of Sweden, China, France, and Japan. Am Rev Public Adm. 2020; 50(6-7):762–769.
2. Jang WM, Jang DH, Lee JY. Social distancing and transmission-reducing practices during the 2019 coronavirus disease and 2015 Middle East respiratory syndrome coronavirus outbreaks in Korea. J Korean Med Sci. 2020; 35(23):e220. PMID: 32537955.
3. Ferguson NM, Laydon D, Nedjati Gilani G, Imai N, Ainslie K, Baguelin M, et al. Impact of non-pharmaceutical interventions (NPIs) to reduce COVID-19 mortality and healthcare demand. Updated 2020. Accessed February, 2024. http://hdl.handle.net/10044/1/77482 .
4. Park IN, Yum HK. Stepwise strategy of social distancing in Korea. J Korean Med Sci. 2020; 35(28):e264. PMID: 32686376.
5. Liu K. COVID-19 and the Chinese economy: impacts, policy responses and implications. Int Rev Appl Econ. 2021; 35(2):308–330.
6. D’Ambrosio C, Rohde N. The distribution of economic insecurity: Italy and the U.S. over the great recession. Rev Income Wealth. 2014; 60(S1):S33–S52.
7. Cohen MJ. Does the COVID-19 outbreak mark the onset of a sustainable consumption transition? Sustain: Sci, Pract, Policy. 2020; 16(1):1–3.
8. Gordon-Wilson S. Consumption practices during the COVID-19 crisis. Int J Consum Stud. 2022; 46(2):575–588. PMID: 34220342.
9. Mertzanis C, Papastathopoulos A, Koutra C, Marashdeh H. COVID-19, mobility, leisure, and travel around the world. J Leis Res. 2023; 54(2):227–249.
10. Hall MC, Prayag G, Fieger P, Dyason D. Beyond panic buying: consumption displacement and COVID-19. J Serv Manag. 2021; 32(1):113–128.
11. Hu T, Wang S, She B, Zhang M, Huang X, Cui Y, et al. Human mobility data in the COVID-19 pandemic: characteristics, applications, and challenges. Int J Digit Earth. 2021; 14(9):1126–1147.
12. Meloni S, Perra N, Arenas A, Gómez S, Moreno Y, Vespignani A. Modeling human mobility responses to the large-scale spreading of infectious diseases. Sci Rep. 2011; 1(1):62. PMID: 22355581.
13. Watts A, Au NH, Thomas-Bachli A, Forsyth J, Mayah O, Popescu S, et al. Potential for inter-state spread of COVID-19 from Arizona, USA: analysis of mobile device location and commercial flight data. J Travel Med. 2020; 27(8):taaa136. PMID: 32822473.
14. Kraemer MUG, Yang CH, Gutierrez B, Wu CH, Klein B, Pigott DM, et al. The effect of human mobility and control measures on the COVID-19 epidemic in China. Science. 2020; 368(6490):493–497. PMID: 32213647.
15. Pepe E, Bajardi P, Gauvin L, Privitera F, Lake B, Cattuto C, et al. COVID-19 outbreak response, a dataset to assess mobility changes in Italy following national lockdown. Sci Data. 2020; 7(1):230. PMID: 32641758.
16. Abouk R, Heydari B. The immediate effect of COVID-19 policies on social-distancing behavior in the United States. Public Health Rep. 2021; 136(2):245–252. PMID: 33400622.
17. De Oliveira Neto L, Elsangedy HM, de Oliveira Tavares VD, Teixeira CV, Behm DG, Da Silva-Grigoletto ME. Training in home-training at home during the COVID-19 (SARS-CoV-2) pandemic: physical exercise and behavior-based approach. Braz J Exerc Physiol. 2020; 19(2):9–19.
18. Sheiner L, Yilla K. The ABCs of the post-COVID economic recovery. Updated 2020. Accessed July, 2024. https://www.brookings.edu/articles/the-abcs-of-the-post-covid-economic-recovery/ .
19. Park C, Park J. COVID-19 and the Korean economy: when, how, and what changes? Asian Soc Innov Policy. 2020; 9(2):187–206.
20. Rozario NL, Moore CG, McWilliams A. Interrupted time series power calculation using DO Loop simulations. Accessed February, 2024. http://support.sas.com/resources/papers/proceedings17/1339-2017.pdf .
21. Abbas J, Mubeen R, Iorember PT, Raza S, Mamirkulova G. Exploring the impact of COVID-19 on tourism: transformational potential and implications for a sustainable recovery of the travel and leisure industry. Curr Res Behav Sci. 2021; 2:100033. PMID: 38620720.
22. Lee CC, Chen MP. Ecological footprint, tourism development, and country risk: International evidence. J Clean Prod. 2021; 279:123671.
23. Siksna I, Uturyte L, Karu K, Vanhalst J. The COVID-19 Pandemic and Food Consumption Patterns. Copenhagen, Denmark: Nordic Council of Ministers;2021. p. 1–28.
24. Ben Hassen T, El Bilali H, Allahyari MS. Impact of COVID-19 on food behavior and consumption in Qatar. Sustainability (Switzerland). 2020; 12(17):1–18.
25. Bartsch SM, Ferguson MC, McKinnell JA, O’Shea KJ, Wedlock PT, Siegmund SS, et al. The potential health care costs and resource use associated with COVID-19 in the United States. Health Aff (Millwood). 2020; 39(6):927–935. PMID: 32324428.
26. Choi YJ, Sohn J, Kim TH. Changes in expenditures of the National Health Insurance of Korea during the COVID-19 pandemic and the financial implications thereof. Yonsei Med J. 2023; 64(1):71–75. PMID: 36579382.
27. Jang SY, Seon JY, Yoon SJ, Park SY, Lee SH, Oh IH. Comorbidities and factors determining medical expenses and length of stay for admitted COVID-19 patients in Korea. Risk Manag Healthc Policy. 2021; 14:2021–2033. PMID: 34040465.
28. Daniel SJ. Education and the COVID-19 pandemic. Prospects. 2020; 49(1-2):91–96. PMID: 32313309.
29. KOSIS. Shadow education participation rate by school level (Korean, author’s translation). Accessed February, 2024. https://kosis.kr/statHtml/statHtml.do?orgId=101&tblId=DT_1PE301&vw_cd=MT_ZTITLE&list_id=O153&seqNo=&lang_mode=ko&language=kor&obj_var_id=&itm_id=&conn_path=MT_ZTITLE .
30. OECD. Private spending on education. Accessed February, 2024. https://data.oecd.org/eduresource/private-spending-on-education.htm .
31. Chakraborty I, Maity P. COVID-19 outbreak: migration, effects on society, global environment and prevention. Sci Total Environ. 2020; 728:138882. PMID: 32335410.
32. Ryu S, Ali ST, Jang C, Kim B, Cowling BJ. Effect of nonpharmaceutical interventions on transmission of severe acute respiratory syndrome coronavirus 2, South Korea, 2020. Emerg Infect Dis. 2020; 26(10):2406–2410. PMID: 32487283.
33. Park YJ, Choe YJ, Park O, Park SY, Kim YM, Kim J, et al. Contact tracing during coronavirus disease outbreak, South Korea, 2020. Emerg Infect Dis. 2020; 26(10):2465–2468. PMID: 32673193.

SUPPLEMENTARY MATERIALS

Supplementary Fig. 1

Intervention effects of consumption on the first healthcare policy between 2019 and 2023 in South Korea.
jkms-40-e6-s001.doc

Supplementary Fig. 2

Intervention effects of human mobility on the first healthcare policy between 2019 and 2023 in South Korea.
jkms-40-e6-s002.doc

Supplementary Fig. 3

Intervention effects of consumption on the second healthcare policy between 2019 and 2023 in South Korea.
jkms-40-e6-s003.doc

Supplementary Fig. 4

Intervention effects of human mobility on the second healthcare policy between 2019 and 2023 in South Korea.
jkms-40-e6-s004.doc

Supplementary Fig. 5

Intervention effects of consumption on the third healthcare policy between 2019 and 2023 in South Korea.
jkms-40-e6-s005.doc

Supplementary Fig. 6

Intervention effects of human mobility on the third healthcare policy between 2019 and 2023 in South Korea.
jkms-40-e6-s006.doc

Supplementary Fig. 7

Intervention effects of consumption on the fourth healthcare policy between 2019 and 2023 in South Korea.
jkms-40-e6-s007.doc

Supplementary Fig. 8

Intervention effects of human mobility on the fourth healthcare policy between 2019 and 2023 in South Korea.
jkms-40-e6-s008.doc

Supplementary Fig. 9

Intervention effects of consumption on the fifth healthcare policy between 2019 and 2023 in South Korea.
jkms-40-e6-s009.doc

Supplementary Fig. 10

Intervention effects of human mobility on the sixth healthcare policy between 2019 and 2023 in South Korea.
jkms-40-e6-s010.doc

Supplementary Fig. 11

Intervention effects of consumption on the seventh healthcare policy between 2019 and 2023 in South Korea.
jkms-40-e6-s011.doc

Supplementary Fig. 12

Intervention effects of human mobility on the seventh healthcare policy between 2019 and 2023 in South Korea.
jkms-40-e6-s012.doc

Supplementary Fig. 13

Intervention effects of consumption on the eighth healthcare policy between 2019 and 2023 in South Korea.
jkms-40-e6-s013.doc

Supplementary Fig. 14

Intervention effects of human mobility on the eighth healthcare policy between 2019 and 2023 in South Korea.
jkms-40-e6-s014.doc

Supplementary Fig. 15

Intervention effects of consumption on the ninth healthcare policy between 2019 and 2023 in South Korea.
jkms-40-e6-s015.doc

Supplementary Fig. 16

Intervention effects of human mobility on the ninth healthcare policy between 2019 and 2023 in South Korea.
jkms-40-e6-s016.doc

Supplementary Fig. 17

Intervention effects of consumption on the tenth healthcare policy between 2019 and 2023 in South Korea.
jkms-40-e6-s017.doc

Supplementary Fig. 18

Intervention effects of human mobility on the tenth healthcare policy between 2019 and 2023 in South Korea.
jkms-40-e6-s018.doc

Supplementary Fig. 19

Intervention effects of consumption on the eleventh healthcare policy between 2019 and 2023 in South Korea.
jkms-40-e6-s019.doc

Supplementary Fig. 20

Intervention effects of human mobility on the eleventh healthcare policy between 2019 and 2023 in South Korea.
jkms-40-e6-s020.doc

Supplementary Fig. 21

Intervention effects of consumption on the twelfth healthcare policy between 2019 and 2023 in South Korea.
jkms-40-e6-s021.doc

Supplementary Fig. 22

Intervention effects of human mobility on the twelfth healthcare policy between 2019 and 2023 in South Korea.
jkms-40-e6-s022.doc
TOOLS
Similar articles