Impact of the “Class B Infectious Disease Class B Management” Policy on the Passenger Volume of Urban Rail Transit: A Nationwide Interrupted Time Series Study

Abstract

Between 2019 and 2022, passenger volume on China’s urban rail transit system sharply declined due to strict COVID-19 control measures. On 8 January 2023, China implemented the “Class B infectious disease Class B management” policy, marking a significant shift towards a more relaxed approach to epidemic control. The main objective of this study is to evaluate the immediate and lasting effects of this policy on urban rail transit passenger volume. We used interrupted time series (ITS), combined with quasi-Poisson regression models and counterfactual analysis, to analyze monthly urban rail transit operation data covering the period from January 2021 to June 2024 for 42 cities. Our analysis shows that, relative to the expected trend without any intervention, monthly average passenger volume increased by approximately 101.34% after the policy’s implementation, with significant immediate effects observed in 41 cities and significant lasting effects observed in 33 cities. The study concludes that the “Class B infectious disease Class B management” policy has generally promoted the nationwide recovery of urban rail transit passenger volume, although with significant heterogeneity across cities. This result indicates that the reduction in travel restrictions and the restoration of public safety, resulting from the relaxation of COVID-19 prevention and control measures, contributed to the overall recovery of urban rail transit. This study not only provides innovative methodological insights but also offers valuable guidance on developing more effective urban planning strategies and urban rail transit operational measures in the post-pandemic era.

1. Introduction

Coronavirus Disease 2019 (COVID-19) is one of the most severe public health crises globally. In response, countries implemented various non-pharmaceutical interventions such as lockdowns, home isolation, and social distancing measures to curb the spread of the virus [1]. However, these measures significantly affected people’s travel and social activities [2,3]. In the cities most severely impacted by the pandemic, mobility decreased by as much as 90% [4], with urban rail transit being particularly affected [5].

Global statistics indicate that traffic volume in most cities declined significantly during the pandemic, primarily as a result of multiple restrictive policies rather than the direct effects of the virus itself [6]. Due to the high passenger density in narrow spaces, public transportation and stations are considered high-risk areas. Many countries have implemented corresponding restrictive measures to reduce the service of urban rail transit [7]. In China, following the outbreak, many cities intensified entry and exit management, implementing health codes and travel codes while decreasing service frequency to limit mobility. Particularly, lockdown measures led cities like Wuhan and Shanghai to temporarily suspend urban rail transit operations. Furthermore, the pandemic diminished commuting needs and curtailed non-essential shopping, leisure, exercise, and social activities, thereby reducing public reliance on public transportation [8]. Against this backdrop, an increasing number of scholars have begun to examine the pandemic’s impact on urban rail transit passenger volume [9].

On 8 January 2023, the Chinese government announced the adjustment of COVID-19 from “Class B infectious disease Class A management” to “Class B infectious disease Class B management” [10]. “Class B infectious disease Class B management” refers to a special management model that, in accordance with the Law of the People’s Republic of China on the Prevention and Control of Infectious Diseases, belongs to Class B Infectious Diseases (including influenza, AIDS, rabies, etc.), and they are controlled in accordance with Class B management measures (including traffic control, restrictions on congregational activities, etc.). Compared with the “Class B infectious disease Class A management” policy, the “Class B infectious disease Class B management” policy is more lenient in terms of isolation measures, regional management, and transportation and health quarantine measures. This policy change indicated that individuals infected with the virus no longer required isolation, and close contacts were not subject to tracking. Additionally, the distinction between high-risk and low-risk areas was eliminated, with testing strategies shifted to “voluntary testing”. Following this policy adjustment, urban rail transit across various cities announced the cancelation of health code and nucleic acid test verification, and temperature checks were no longer conducted at entry points. These changes signify China’s gradual transition to a new phase of COVID-19 control [11,12].

In public transportation, policy relaxation reduces the cost of passengers’ time, the cost of convenience, and the health risks associated with stringent precautionary measures [13]. When these non-pecuniary costs are reduced, the overall utility of rail transportation for passengers increases, leading to increased demand [14]. Studies indicate that prior to the outbreak, approximately 73% of respondents relied on public transportation as their primary mode of travel. This proportion declined to under 50% during the pandemic, and in the post-epidemic period, about 56% of respondents expressed a willingness to continue using public transportation [15]. The “Class B infectious disease Class B management” policy is more relaxed than earlier epidemic control policies, yet its potential impact on urban rail transit operations remains uncertain.

Therefore, further research is necessary to understand the recovery of urban rail transit passenger volume following the effective control of COVID-19, particularly to clarify the effects of public health measures on urban rail transit. Compared to studies conducted during and before the pandemic, there is limited research on the relationship between urban rail transit and prevention policies in the post-epidemic era. To fill this gap, we focus on China, where pandemic control policies have been in place the longest globally, by using the implementation of the “Class B infectious disease Class B management” policy as a timeline marker. We employ interrupted time series analysis to systematically investigate the recovery characteristics of urban rail transit across different cities after the effective control of the pandemic and explore whether and to what extent (short-term or long-term) this policy has influenced the trends of urban rail transit passenger volume.

2. Methods

2.1. Data Sources

The data used in this study include urban population data and monthly data on urban rail transit operations. Urban population data were obtained from the China Urban Statistical Yearbook, and monthly data on urban rail transit operations were obtained from the official website of the Ministry of Transport of the People’s Republic of China (https://www.mot.gov.cn/ (accessed on 20 July 2024)). The data range from January 2021 to June 2024. During this time period, each city has a stable record of rail transit operations, avoiding potential bias due to different construction phases or missing data. During the data processing process, we performed a preliminary check of the data for each city to remove outliers and missing data, as well as to ensure that the data were continuous and consistent in the time dimension.

2.2. Outcomes

In this study, the urban rail transit passenger volume of 42 cities in China that have opened urban rail transit (up to early 2024, there were a total of 53 cities in China that have opened urban rail transit, and to ensure data integrity, the 42 cities that have opened urban rail transit before January 2021 were selected as the research objects) were analyzed for interrupted time series, and the national outcome indicators and the individual outcome indicators of the 42 cities were obtained. Figure 1 shows the distribution of the 42 research areas.

2.3. Intervention

The intervention in this study was defined as the implementation of the “Class B infectious disease Class B management” policy. Consequently, we divided the COVID-19 pandemic into two phases: the “Class B infectious disease Class A management” phase (from January 2021 to December 2022) and the “Class B infectious disease Class B management” phase (from January 2023 to June 2024).

2.4. Interrupted Time Series Analysis

Interrupted time series (ITS) analysis is a quasi-experimental method widely used in fields such as social policy, healthcare policy, and environmental policy. It evaluates the effects of interventions by comparing changes in levels and trends before and after the implementation of the intervention [16]. ITS analysis uses the existing trend from the phase before the intervention as a counterfactual to predict the trend in urban rail transit passenger volume in the absence of the intervention. This expected trend is then compared with the actual observed passenger volume after the intervention, allowing for inferences about the intervention’s effectiveness at a specific point in time. ITS can assess not only the immediate effects of interventions but also analyze their impact on long-term trends. Another significant advantage of ITS is that it allows for comparisons within the same study group before and after the intervention, eliminating the need for a control group [17]. By using the ITS method, researchers can gain a better understanding of the impact of policy interventions on urban rail transit passenger volume and provide strong evidence for addressing similar public health crises.

The basic formula for ITS is as follows:

$$Y_t={\mathsf{\beta }}_0+{\mathsf{\beta }}_1{\mathit{X}}_1+{\mathsf{\beta }}_2{\mathit{X}}_2+{\mathsf{\beta }}_3{\mathit{X}}_3+{\mathsf{\epsilon }}_t$$

(1)

In this equation, $Y_t$ represents the urban rail transit passenger volume; ${\mathit{X}}_1$ is the time indicator variable (monthly); ${\mathit{X}}_2$ is the categorical variable (before Class B management, i.e., pre-intervention = 0; after Class B management, i.e., post-intervention = 1); ${\mathit{X}}_3$ is the time variable after the intervention (pre-intervention = 0 and post-intervention = 1, 2, 3, …); ${\mathsf{ϵ}}_t$ represents the residual term; ${\mathsf{\beta }}_0$ is the intercept, indicating the initial level of urban rail transit passenger volume; ${\mathsf{\beta }}_1$ is the slope before the intervention, indicating the existing trend in passenger volume; ${\mathsf{\beta }}_2$ represents the difference between the actual outcome at the time of the intervention and the counterfactual outcome that would have occurred had the intervention not taken place, reflecting the immediate effect of the intervention. For example, if ${\mathsf{\beta }}_2$ = 0.394, then RR = 1.483, indicating a 48.30% increase in passenger volume in the short term.; and ${\mathsf{\beta }}_3$ is the change in slope before and after the intervention, reflecting the long-term effect of the intervention, with the slope in the post-intervention period being ${\mathsf{\beta }}_1+{\mathsf{\beta }}_3$. For example, if ${\mathsf{\beta }}_1$ = −0.011 and ${\mathsf{\beta }}_3$ = 0.037, then RR = 1.026, indicating a 2.60% increase in passenger volume for each month after the intervention.

Given that the results of this study are based on monthly reported urban rail transit passenger volume data and the variance in the data is large, leading to issues of overdispersion, a quasi-Poisson regression model was employed [18]. All analyses were conducted using RStudio (version 4.1.1) and the associated R packages tsModel (version 0.6) and AER (version 1.2–1.4).

3. Results

In this study, the ITS method was used to predict the monthly passenger volume of urban rail transit in 42 cities nationwide from January 2023 to June 2024, assuming that the “Class B infectious disease Class B management” policy had not been implemented. Furthermore, we used September 2021 as a spurious intervention point for further placebo testing, and we compared the results with those obtained from the generalized additive model (GAM). The relevant content can be found in Appendix A.

3.1. Overall Trends for the 42 Cities

The total passenger volume of urban rail transit in the 42 cities increased from 178,519.2 million trips in January 2021 to 262,376.2 million trips in June 2024. Since the implementation of the policy of “Class B infectious disease Class B management”, the average monthly passenger volume in 42 cities nationwide has increased from 179,604.6 million (average monthly passenger volume from January 2021 to December 2022) to 250,485.6 million (average monthly passenger volume from January 2023 to June 2024). The counterfactual analysis and parameter estimation results indicated the following: first, the average monthly passenger volume of urban rail transit increased from 124,409.9 million trips (counterfactual values) to 250,485.6 million trips (actual values) after the intervention, reflecting a growth of 101.34%; second, the intervention had a significant positive immediate effect [${\mathsf{\beta }}_2$ = 0.394 (0.225, 0.563), RR = 1.483 (1.253, 1.756), p < 0.001]; third, the intervention had a significant positive lasting effect [${\mathsf{\beta }}_3$ = 0.028 (0.014, 0.042), RR = 1.028 (1.015, 1.042), p < 0.001] (Figure 2 and

Table 1. Parameter estimation results for the total monthly passenger volume of urban rail transit in 42 cities *.

Parameters	Estimated Value (95%CI)	SE	z	RR (95%CI)	p-Value
${\mathsf{\beta }}_0$	3.861 (3.745, 3.977)	0.059	65.881	47.507 (42.352, 53.290)	<0.001
${\mathsf{\beta }}_1$	−0.017 (−0.025, −0.009)	0.004	−3.975	0.983 (0.975, 0.991)	<0.001
${\mathsf{\beta }}_2$	0.394 (0.225, 0.563)	0.086	4.58	1.483 (1.253, 1.756)	<0.001
${\mathsf{\beta }}_3$	0.028 (0.014, 0.042)	0.007	4.069	1.028 (1.015, 1.042)	<0.001

* ${\mathsf{\beta }}_0$: intercept, ${\mathsf{\beta }}_1$: slope before the intervention, ${\mathsf{\beta }}_2$: immediate effect of the intervention, ${\mathsf{\beta }}_3$: amount of change in slope before and after the intervention, SE: standard error, z: normal distribution statistic, RR: relative risk, CI: confidence interval.

).

3.2. Trends for Each of the 42 Cities

Figure 3 shows the percentage change in the average monthly passenger volume of urban rail transit in 42 cities between counterfactual values and actual values after the implementation of the “Class B infectious disease Class B management” policy. We found that in January 2023, 40 cities experienced an upward trend in passenger volume, with the highest increase of 80.06% in Zhengzhou. Only two cities reported a decrease in passenger volume, with the largest decline of 27.09% in Foshan. Pre-intervention passenger volume and post-intervention changes in 42 cities are shown in Appendix B (

Table A3. Pre-intervention passenger volume and post-intervention changes in 42 cities.

Pre-Intervention Passenger Volume Ranking	City	Rate of Change in Passenger Volume After Intervention	Ranking of Passengers Volume Change Rates After Intervention
1	Shanghai	179.95%	8
2	Beijing	119.67%	15
3	Guangzhou	71.04%	33
4	Shenzhen	107.96%	20
5	Chengdu	58.44%	35
6	Wuhan	84.42%	29
7	Hangzhou	36.77%	39
8	Chongqing	95.06%	25
9	Xian	136.29%	12
10	Nanjing	66.04%	34
11	Changsha	90.62%	27
12	Zhengzhou	219.77%	5
13	Tianjin	176.56%	9
14	Shenyang	168.70%	11
15	Suzhou	114.86%	16
16	Qingdao	52.90%	36
17	Hefei	78.68%	31
18	Nanchang	96.45%	24
19	Ningbo	49.65%	37
20	Nanning	41.91%	38
21	Harbin	13.54%	40
22	Kunming	103.07%	21
23	Fuzhou	114.40%	18
24	Xiamen	6.32%	41
25	Dalian	114.27%	19
26	Changchun	261.68%	4
27	Wuxi	125.85%	14
28	Shijiazhuang	114.72%	17
29	Guiyang	73.88%	32
30	Foshan	−5.13%	42
31	Lanzhou	569.86%	1
32	Jinan	100.63%	23
33	Xuzhou	101.97%	22
34	Changzhou	86.44%	28
35	Hohhot	298.49%	3
36	Dongguan	92.17%	26
37	Taiyuan	131.17%	13
38	Urumqi	311.03%	2
39	Wenzhou	173.23%	10
40	Huaian	84.36%	30
41	Sanya	214.28%	6
42	Tianshui	209.11%	7

). The counterfactual analysis and parameter estimation results showed the following: first, the five cities with the highest changes in average monthly passenger volume after the intervention were Lanzhou (569.86%), Urumqi (311.03%), Hohhot (298.49%), Changchun (261.68%), and Zhengzhou (219.77%); second, the five cities with the lowest changes in average monthly passenger volume after the intervention were Foshan (−5.13%), Xiamen (6.32%), Harbin (13.54%), Hangzhou (36.77%), and Nanning (41.91%); third, the five cities with the highest passenger volume were Shanghai, Beijing, Guangzhou, Shenzhen, and Chengdu, with their respective changes being 179.95% (8th), 119.67% (15th), 71.04% (33rd), 107.96% (20th), and 58.44% (35th); forth, 41 cities exhibited a significant positive immediate effect following the intervention, while Xiamen did not show a significant immediate effect [${\mathsf{\beta }}_2$ = 0.091 (−0.072, 2.254), RR = 1.096 (0.931, 1.289), p = 0.272]; fifth, 33 cities exhibited a significant positive lasting effect following the intervention, while Hangzhou [${\mathsf{\beta }}_3$ = 0.006 (−0.006, 0.018), RR = 1.006 (0.994, 1.019), p = 0.347], Qingdao [${\mathsf{\beta }}_3$ = 0.007 (−0.009, 0.023), RR = 1.007 (0.992, 1.022), p = 0.372], Ningbo [${\mathsf{\beta }}_3$ = 0.011 (−0.001, 0.023), RR = 1.011 (0.999, 1.023), p = 0.066], Nanning [${\mathsf{\beta }}_3$ = 0.010 (−0.002, 0.022), RR = 1.010 (0.998, 1.023), p = 0.114], Xiamen [${\mathsf{\beta }}_3$ = −0.002 (−0.016, 0.012), RR = 0.998 (0.984, 1.011), p = 0.746], Shijiazhuang [${\mathsf{\beta }}_3$ = 0.032 (−0.001, 0.065), RR = 1.032 (0.998, 1.067), p = 0.063], Jinan [${\mathsf{\beta }}_3$ = 0.017 (−0.012, 0.046), RR = 1.018 (0.988, 1.048), p = 0.239], Changzhou [${\mathsf{\beta }}_3$ = 0.011 (−0.011, 0.033), RR = 1.011 (0.990, 1.033), p = 0.291], and Harbin [${\mathsf{\beta }}_3$ = −0.010 (−0.039, 0.019), RR = 0.990 (0.961, 1.021), p= 0.530] did not show a significant lasting effect (Figure 4 and

Table 2. Parameter estimation results for the monthly passenger volume of urban rail transit in different cities *.

City	Parameters	Estimated Value (95%CI)	SE	z	RR (95%CI)	p-Value
Shanghai	${\mathsf{\beta }}_0$	2.628 (2.395, 2.861)	0.119	22.046	13.849 (10.964, 17.495)	<0.001
	${\mathsf{\beta }}_1$	−0.029 (−0.047, −0.011)	0.009	−3.199	0.971 (0.954, 0.989)	0.001
	${\mathsf{\beta }}_2$	0.537 (0.167, 0.907)	0.189	2.838	1.711 (1.181, 2.479)	0.005
	${\mathsf{\beta }}_3$	0.037 (0.008, 0.066)	0.015	2.451	1.038 (1.007, 1.069)	0.014
Beijing	${\mathsf{\beta }}_0$	2.584 (2.413, 2.755)	0.087	29.576	13.251 (11.165, 15.725)	<0.001
	${\mathsf{\beta }}_1$	−0.022 (−0.036, −0.008)	0.007	−3.392	0.978 (0.965, 0.991)	0.001
	${\mathsf{\beta }}_2$	0.469 (0.206, 0.732)	0.134	3.502	1.599 (1.229, 2.079)	<0.001
	${\mathsf{\beta }}_3$	0.030 (0.008, 0.052)	0.011	2.830	1.031 (1.009, 1.053)	0.005
Chengdu	${\mathsf{\beta }}_0$	2.028 (1.906, 2.150)	0.062	32.74	7.595 (6.727, 8.575)	<0.001
	${\mathsf{\beta }}_1$	−0.011 (−0.021, −0.001)	0.005	−2.500	0.989 (0.980, 0.998)	0.012
	${\mathsf{\beta }}_2$	0.319 (0.139, 0.499)	0.092	3.485	1.376 (1.150, 1.646)	<0.001
	${\mathsf{\beta }}_3$	0.016 (0.002, 0.030)	0.007	2.203	1.016 (1.002, 1.031)	0.028
Guangzhou	${\mathsf{\beta }}_0$	2.627 (2.474, 2.780)	0.078	33.716	13.833 (11.874, 16.116)	<0.001
	${\mathsf{\beta }}_1$	−0.016 (−0.028, −0.004)	0.006	−2.817	0.984 (0.973, 0.995)	0.005
	${\mathsf{\beta }}_2$	0.326 (0.093, 0.559)	0.119	2.731	1.385 (1.096, 1.750)	0.006
	${\mathsf{\beta }}_3$	0.023 (0.003, 0.043)	0.010	2.350	1.023 (1.004, 1.042)	0.019
Shenzhen	${\mathsf{\beta }}_0$	2.394 (2.265, 2.523)	0.066	36.186	10.959 (9.626, 12.477)	<0.001
	${\mathsf{\beta }}_1$	−0.014 (−0.024, −0.004)	0.005	−2.804	0.987 (0.977, 0.996)	0.005
	${\mathsf{\beta }}_2$	0.403 (0.215, 0.591)	0.096	4.174	1.496 (1.238, 1.807)	<0.001
	${\mathsf{\beta }}_3$	0.025 (0.009, 0.041)	0.008	3.290	1.026 (1.010, 1.041)	0.001
Wuhan	${\mathsf{\beta }}_0$	1.894 (1.743, 2.045)	0.077	24.462	6.645 (5.709, 7.734)	<0.001
	${\mathsf{\beta }}_1$	−0.011 (−0.023, 0.001)	0.006	−1.992	0.989 (0.978, 1.000)	0.046
	${\mathsf{\beta }}_2$	0.405 (0.187, 0.623)	0.111	3.648	1.499 (1.206, 1.863)	<0.001
	${\mathsf{\beta }}_3$	0.021 (0.003, 0.039)	0.009	2.404	1.022 (1.004, 1.039)	0.016
Nanjing	${\mathsf{\beta }}_0$	2.143 (1.996, 2.290)	0.075	28.566	8.528 (7.362, 9.879)	<0.001
	${\mathsf{\beta }}_1$	−0.013 (−0.023, −0.003)	0.005	−2.378	0.987 (0.976, 0.998)	0.017
	${\mathsf{\beta }}_2$	0.275 (0.055, 0.495)	0.112	2.441	1.316 (1.056, 1.640)	0.015
	${\mathsf{\beta }}_3$	0.023 (0.005, 0.041)	0.009	2.563	1.023 (1.005, 1.042)	0.010
Chongqing	${\mathsf{\beta }}_0$	1.165 (1.022, 1.308)	0.073	15.891	3.206 (2.777, 3.701)	<0.001
	${\mathsf{\beta }}_1$	−0.018 (−0.028, −0.008)	0.005	−3.257	0.982 (0.972, 0.993)	0.001
	${\mathsf{\beta }}_2$	0.412 (0.196, 0.628)	0.110	3.755	1.510 (1.218, 1.872)	<0.001
	${\mathsf{\beta }}_3$	0.029 (0.011, 0.047)	0.009	3.342	1.029 (1.012, 1.047)	0.001
Hangzhou	${\mathsf{\beta }}_0$	1.768 (1.654, 1.882)	0.058	30.515	5.856 (5.228, 6.560)	<0.001
	${\mathsf{\beta }}_1$	0.005 (−0.003, 0.013)	0.004	1.331	1.005 (0.997, 1.013)	0.183
	${\mathsf{\beta }}_2$	0.225 (0.074, 0.376)	0.077	2.935	1.253 (1.078, 1.457)	0.003
	${\mathsf{\beta }}_3$	0.006 (−0.006, 0.018)	0.006	0.940	1.006 (0.994, 1.019)	0.347
Qingdao	${\mathsf{\beta }}_0$	0.655 (0.512, 0.798)	0.073	8.972	1.926 (1.669, 2.222)	<0.001
	${\mathsf{\beta }}_1$	0.008 (−0.002, 0.018)	0.005	1.609	1.008 (0.998, 1.018)	0.108
	${\mathsf{\beta }}_2$	0.368 (0.188, 0.548)	0.092	4.022	1.445 (1.208, 1.729)	<0.001
	${\mathsf{\beta }}_3$	0.007 (−0.009, 0.023)	0.008	0.892	1.007 (0.992, 1.022)	0.372
Xian	${\mathsf{\beta }}_0$	2.005 (1.852, 2.158)	0.078	25.743	7.424 (6.373, 8.648)	<0.001
	${\mathsf{\beta }}_1$	−0.021 (−0.033, −0.009)	0.006	−3.676	0.979 (0.968, 0.990)	<0.001
	${\mathsf{\beta }}_2$	0.573 (0.348, 0.798)	0.115	4.980	1.774 (1.416, 2.224)	<0.001
	${\mathsf{\beta }}_3$	0.028 (0.010, 0.046)	0.009	3.055	1.028 (1.010, 1.047)	0.002
Tianjin	${\mathsf{\beta }}_0$	1.155 (1.024, 1.286)	0.067	17.226	3.175 (2.784, 3.622)	<0.001
	${\mathsf{\beta }}_1$	−0.024 (−0.034, −0.014)	0.005	−4.731	0.976 (0.967, 0.986)	<0.001
	${\mathsf{\beta }}_2$	0.587 (0.391, 0.783)	0.100	5.852	1.799 (1.478, 2.190)	<0.001
	${\mathsf{\beta }}_3$	0.036 (0.020, 0.052)	0.008	4.602	1.037 (1.021, 1.053)	<0.001
Shenyang	${\mathsf{\beta }}_0$	1.357 (1.159, 1.555)	0.101	13.406	3.886 (3.186, 4.739)	<0.001
	${\mathsf{\beta }}_1$	−0.018 (−0.032, −0.004)	0.007	−2.428	0.982 (0.968, 0.997)	0.015
	${\mathsf{\beta }}_2$	0.402 (0.114, 0.690)	0.147	2.730	1.496 (1.120, 1.997)	0.006
	${\mathsf{\beta }}_3$	0.049 (0.025, 0.073)	0.012	4.189	1.050 (1.026, 1.075)	<0.001
Suzhou	${\mathsf{\beta }}_0$	1.036 (0.877, 1.195)	0.081	12.797	2.818 (2.404, 3.302)	<0.001
	${\mathsf{\beta }}_1$	−0.013 (−0.025, −0.001)	0.006	−2.190	0.987 (0.976, 0.999)	0.029
	${\mathsf{\beta }}_2$	0.376 (0.147, 0.605)	0.117	3.206	1.457 (1.158, 1.834)	0.001
	${\mathsf{\beta }}_3$	0.028 (0.010, 0.046)	0.009	2.918	1.028 (1.009, 1.047)	0.004
Zhengzhou	${\mathsf{\beta }}_0$	1.241 (0.963, 1.519)	0.142	8.733	3.461 (2.619, 4.573)	<0.001
	${\mathsf{\beta }}_1$	−0.034 (−0.056, −0.012)	0.011	−3.057	0.967 (0.946, 0.988)	0.002
	${\mathsf{\beta }}_2$	0.741 (0.316, 1.166)	0.217	3.409	2.099 (1.370, 3.214)	0.001
	${\mathsf{\beta }}_3$	0.055 (0.024, 0.086)	0.016	3.324	1.056 (1.023, 1.091)	0.001
Changsha	${\mathsf{\beta }}_0$	1.620 (1.479, 1.761)	0.072	22.375	5.052 (4.384, 5.823)	<0.001
	${\mathsf{\beta }}_1$	−0.004 (−0.014, 0.006)	0.005	−0.862	0.996 (0.986, 1.006)	0.389
	${\mathsf{\beta }}_2$	0.426 (0.234, 0.618)	0.098	4.355	1.531 (1.264, 1.855)	<0.001
	${\mathsf{\beta }}_3$	0.016 (0.000, 0.032)	0.008	1.963	1.016 (1.002, 1.032)	0.048
Dalian	${\mathsf{\beta }}_0$	0.629 (0.451, 0.807)	0.091	6.944	1.876 (1.571, 2.241)	<0.001
	${\mathsf{\beta }}_1$	−0.012 (−0.026, 0.002)	0.007	−1.849	0.988 (0.975, 1.001)	0.064
	${\mathsf{\beta }}_2$	0.545 (0.298, 0.792)	0.126	4.342	1.725 (1.349, 2.206)	<0.001
	${\mathsf{\beta }}_3$	0.025 (0.005, 0.045)	0.010	2.465	1.025 (1.005, 1.045)	0.014
Ningbo	${\mathsf{\beta }}_0$	0.832 (0.728, 0.936)	0.053	15.583	2.299 (2.070, 2.553)	<0.001
	${\mathsf{\beta }}_1$	−0.002 (−0.010, 0.006)	0.004	−0.590	0.998 (0.990, 1.005)	0.555
	${\mathsf{\beta }}_2$	0.302 (0.157, 0.447)	0.074	4.083	1.352 (1.170, 1.563)	<0.001
	${\mathsf{\beta }}_3$	0.011 (−0.001, 0.023)	0.006	1.838	1.011 (0.999, 1.023)	0.066
Kunming	${\mathsf{\beta }}_0$	0.857 (0.73, 0.984)	0.065	13.130	2.355 (2.072, 2.676)	<0.001
	${\mathsf{\beta }}_1$	−0.014 (−0.024, −0.004)	0.005	−2.990	0.986 (0.977, 0.995)	0.003
	${\mathsf{\beta }}_2$	0.463 (0.279, 0.647)	0.094	4.913	1.588 (1.321, 1.910)	<0.001
	${\mathsf{\beta }}_3$	0.023 (0.007, 0.039)	0.008	3.036	1.023 (1.008, 1.038)	0.002
Hefei	${\mathsf{\beta }}_0$	0.896 (0.753, 1.039)	0.073	12.224	2.451 (2.123, 2.830)	<0.001
	${\mathsf{\beta }}_1$	−0.004 (−0.014, 0.006)	0.005	−0.688	0.996 (0.986, 1.007)	0.491
	${\mathsf{\beta }}_2$	0.26 (0.064, 0.456)	0.100	2.598	1.297 (1.066, 1.579)	0.009
	${\mathsf{\beta }}_3$	0.032 (0.016, 0.048)	0.008	4.013	1.033 (1.017, 1.049)	<0.001
Nanning	${\mathsf{\beta }}_0$	1.021 (0.913, 1.129)	0.055	18.477	2.777 (2.492, 3.094)	<0.001
	${\mathsf{\beta }}_1$	−0.004 (−0.012, 0.004)	0.004	−1.081	0.996 (0.988, 1.003)	0.280
	${\mathsf{\beta }}_2$	0.224 (0.067, 0.381)	0.080	2.814	1.251 (1.070, 1.462)	0.005
	${\mathsf{\beta }}_3$	0.01 (−0.002, 0.022)	0.006	1.580	1.010 (0.998, 1.023)	0.114
Changchun	${\mathsf{\beta }}_0$	0.758 (0.495, 1.021)	0.134	5.676	2.135 (1.643, 2.773)	<0.001
	${\mathsf{\beta }}_1$	−0.032 (−0.052, −0.012)	0.010	−3.169	0.968 (0.949, 0.988)	0.002
	${\mathsf{\beta }}_2$	0.562 (0.154, 0.97)	0.208	2.699	1.754 (1.166, 2.638)	0.007
	${\mathsf{\beta }}_3$	0.056 (0.023, 0.089)	0.017	3.363	1.057 (1.023, 1.092)	0.001
Nanchang	${\mathsf{\beta }}_0$	1.267 (1.108, 1.426)	0.081	15.607	3.551 (3.029, 4.164)	<0.001
	${\mathsf{\beta }}_1$	−0.008 (−0.020, 0.004)	0.006	−1.426	0.992 (0.980, 1.003)	0.154
	${\mathsf{\beta }}_2$	0.392 (0.171, 0.613)	0.113	3.464	1.479 (1.185, 1.846)	0.001
	${\mathsf{\beta }}_3$	0.026 (0.008, 0.044)	0.009	2.823	1.026 (1.008, 1.044)	0.005
Wuxi	${\mathsf{\beta }}_0$	0.58 (0.441, 0.719)	0.071	8.125	1.786 (1.553, 2.054)	<0.001
	${\mathsf{\beta }}_1$	−0.016 (−0.026, −0.006)	0.005	−3.041	0.984 (0.974, 0.994)	0.002
	${\mathsf{\beta }}_2$	0.375 (0.169, 0.581)	0.105	3.569	1.455 (1.184, 1.787)	<0.001
	${\mathsf{\beta }}_3$	0.040 (0.024, 0.056)	0.008	4.856	1.041 (1.024, 1.058)	<0.001
Xiamen	${\mathsf{\beta }}_0$	0.914 (0.792, 1.036)	0.062	14.655	2.493 (2.206, 2.817)	<0.001
	${\mathsf{\beta }}_1$	0.011 (0.003, 0.019)	0.004	2.672	1.011 (1.003, 1.020)	0.008
	${\mathsf{\beta }}_2$	0.091 (−0.072, 0.254)	0.083	1.099	1.096 (0.931, 1.289)	0.272
	${\mathsf{\beta }}_3$	−0.002 (−0.016, 0.012)	0.007	−0.324	0.998 (0.984, 1.011)	0.746
Shijiazhuang	${\mathsf{\beta }}_0$	−0.328 (−0.647, −0.009)	0.163	−2.009	0.721 (0.523, 0.992)	0.045
	${\mathsf{\beta }}_1$	−0.007 (−0.031, 0.017)	0.012	−0.588	0.993 (0.971, 1.016)	0.556
	${\mathsf{\beta }}_2$	0.539 (0.116, 0.962)	0.216	2.495	1.714 (1.123, 2.618)	0.013
	${\mathsf{\beta }}_3$	0.032 (−0.001, 0.065)	0.017	1.858	1.032 (0.998, 1.067)	0.063
Fuzhou	${\mathsf{\beta }}_0$	0.191 (0.001, 0.381)	0.097	1.976	1.211 (1.002, 1.464)	0.048
	${\mathsf{\beta }}_1$	−0.002 (−0.016, 0.012)	0.007	−0.284	0.998 (0.985, 1.012)	0.777
	${\mathsf{\beta }}_2$	0.337 (0.090, 0.584)	0.126	2.676	1.401 (1.094, 1.793)	0.007
	${\mathsf{\beta }}_3$	0.047 (0.027, 0.067)	0.010	4.694	1.048 (1.028, 1.069)	<0.001
Wenzhou	${\mathsf{\beta }}_0$	−2.395 (−2.522, −2.268)	0.065	−36.944	0.091 (0.080, 0.104)	<0.001
	${\mathsf{\beta }}_1$	−0.016 (−0.026, −0.006)	0.005	−3.451	0.984 (0.975, 0.993)	0.001
	${\mathsf{\beta }}_2$	0.481 (0.305, 0.657)	0.090	5.360	1.618 (1.357, 1.929)	<0.001
	${\mathsf{\beta }}_3$	0.057 (0.043, 0.071)	0.007	8.025	1.059 (1.044, 1.074)	<0.001
Hohhot	${\mathsf{\beta }}_0$	0.506 (0.230, 0.782)	0.141	3.592	1.659 (1.258, 2.186)	<0.001
	${\mathsf{\beta }}_1$	−0.041 (−0.063, −0.019)	0.011	−3.687	0.960 (0.939, 0.981)	<0.001
	${\mathsf{\beta }}_2$	0.801 (0.366, 1.236)	0.222	3.606	2.227 (1.441, 3.442)	<0.001
	${\mathsf{\beta }}_3$	0.061 (0.028, 0.094)	0.017	3.588	1.063 (1.028, 1.099)	<0.001
Jinan	${\mathsf{\beta }}_0$	−0.759 (−1.039, −0.479)	0.143	−5.315	0.468 (0.354, 0.619)	<0.001
	${\mathsf{\beta }}_1$	0.003 (−0.017, 0.023)	0.010	0.294	1.003 (0.984, 1.023)	0.769
	${\mathsf{\beta }}_2$	0.391 (0.032, 0.750)	0.183	2.141	1.478 (1.034, 2.114)	0.032
	${\mathsf{\beta }}_3$	0.017 (−0.012, 0.046)	0.015	1.177	1.018 (0.988, 1.048)	0.239
Xuzhou	${\mathsf{\beta }}_0$	−0.461 (−0.667, −0.255)	0.105	−4.386	0.631 (0.513, 0.775)	<0.001
	${\mathsf{\beta }}_1$	−0.007 (−0.023, 0.009)	0.008	−0.926	0.993 (0.978, 1.008)	0.355
	${\mathsf{\beta }}_2$	0.353 (0.065, 0.641)	0.147	2.397	1.423 (1.066, 1.899)	0.017
	${\mathsf{\beta }}_3$	0.028 (0.004, 0.052)	0.012	2.394	1.028 (1.005, 1.052)	0.017
Dongguan	${\mathsf{\beta }}_0$	−1.026 (−1.151, −0.901)	0.064	−16.094	0.359 (0.316, 0.406)	<0.001
	${\mathsf{\beta }}_1$	−0.017 (−0.027, −0.007)	0.005	−3.667	0.983 (0.974, 0.992)	<0.001
	${\mathsf{\beta }}_2$	0.364 (0.174, 0.554)	0.097	3.750	1.439 (1.190, 1.741)	<0.001
	${\mathsf{\beta }}_3$	0.026 (0.010, 0.042)	0.008	3.322	1.026 (1.011, 1.042)	0.001
Guiyang	${\mathsf{\beta }}_0$	−0.386 (−0.596, −0.176)	0.107	−3.613	0.680 (0.551, 0.838)	<0.001
	${\mathsf{\beta }}_1$	0.018 (0.004, 0.032)	0.007	2.530	1.018 (1.004, 1.033)	0.011
	${\mathsf{\beta }}_2$	0.258 (0.017, 0.499)	0.123	2.103	1.294 (1.018, 1.646)	0.035
	${\mathsf{\beta }}_3$	0.031 (0.011, 0.051)	0.010	3.115	1.031 (1.011, 1.051)	0.002
Changzhou	${\mathsf{\beta }}_0$	−0.294 (−0.482, −0.106)	0.096	−3.051	0.745 (0.617, 0.900)	0.002
	${\mathsf{\beta }}_1$	−0.003 (−0.017, 0.011)	0.007	−0.460	0.997 (0.984, 1.010)	0.646
	${\mathsf{\beta }}_2$	0.432 (0.177, 0.687)	0.130	3.311	1.540 (1.193, 1.988)	0.001
	${\mathsf{\beta }}_3$	0.011 (−0.011, 0.033)	0.011	1.056	1.011 (0.990, 1.033)	0.291
Harbin	${\mathsf{\beta }}_0$	−0.67 (−1.025, −0.315)	0.181	−3.699	0.512 (0.359, 0.730)	<0.001
	${\mathsf{\beta }}_1$	0.039 (0.015, 0.063)	0.012	3.395	1.040 (1.017, 1.064)	0.001
	${\mathsf{\beta }}_2$	0.369 (0.014, 0.724)	0.181	2.034	1.446 (1.014, 2.062)	0.042
	${\mathsf{\beta }}_3$	−0.010 (−0.039, 0.019)	0.015	−0.628	0.990 (0.961, 1.021)	0.530
Foshan	${\mathsf{\beta }}_0$	−0.210 (−0.306, −0.114)	0.049	−4.267	0.811 (0.736, 0.893)	<0.001
	${\mathsf{\beta }}_1$	−0.005 (−0.013, 0.003)	0.004	−1.500	0.995 (0.988, 1.022)	0.134
	${\mathsf{\beta }}_2$	0.408 (0.275, 0.541)	0.068	6.039	1.504 (1.318, 1.717)	<0.001
	${\mathsf{\beta }}_3$	0.020 (0.010, 0.030)	0.005	3.725	1.020 (1.010, 1.031)	<0.001
Urumqi	${\mathsf{\beta }}_0$	−0.150 (−0.434, 0.134)	0.145	−1.035	0.861 (0.649, 1.143)	0.301
	${\mathsf{\beta }}_1$	−0.050 (−0.074, −0.026)	0.012	−4.223	0.951 (0.929, 0.974)	<0.001
	${\mathsf{\beta }}_2$	1.014 (0.553, 1.475)	0.235	4.316	2.756 (1.739, 4.368)	<0.001
	${\mathsf{\beta }}_3$	0.063 (0.028, 0.098)	0.018	3.561	1.065 (1.029, 1.102)	<0.001
Lanzhou	${\mathsf{\beta }}_0$	0.585 (0.326, 0.844)	0.132	4.417	1.795 (1.385, 2.327)	<0.001
	${\mathsf{\beta }}_1$	−0.057 (−0.079, −0.035)	0.011	−5.164	0.945 (0.924, 0.965)	<0.001
	${\mathsf{\beta }}_2$	1.152 (0.738, 1.566)	0.211	5.463	3.164 (2.093, 4.783)	<0.001
	${\mathsf{\beta }}_3$	0.099 (0.070, 0.128)	0.015	6.387	1.104 (1.071, 1.138)	<0.001
Taiyuan	${\mathsf{\beta }}_0$	−0.363 (−0.524, −0.202)	0.082	−4.410	0.696 (0.592, 0.817)	<0.001
	${\mathsf{\beta }}_1$	−0.024 (−0.036, −0.012)	0.006	−3.859	0.976 (0.965, 0.988)	<0.001
	${\mathsf{\beta }}_2$	0.442 (0.193, 0.691)	0.127	3.483	1.556 (1.213, 1.995)	<0.001
	${\mathsf{\beta }}_3$	0.038 (0.018, 0.058)	0.010	3.769	1.039 (1.018, 1.059)	<0.001
Huaian	${\mathsf{\beta }}_0$	−1.979 (−2.114, −1.844)	0.069	−28.517	0.138 (0.121, 0.158)	<0.001
	${\mathsf{\beta }}_1$	−0.017 (−0.027, −0.007)	0.005	−3.283	0.983 (0.973, 0.993)	0.001
	${\mathsf{\beta }}_2$	0.246 (0.036, 0.456)	0.107	2.299	1.279 (1.037, 1.577)	0.022
	${\mathsf{\beta }}_3$	0.037 (0.021, 0.053)	0.008	4.443	1.038 (1.021, 1.055)	<0.001
Tianshui	${\mathsf{\beta }}_0$	−3.510 (−3.728, −3.292)	0.111	−31.656	0.030 (0.024, 0.037)	<0.001
	${\mathsf{\beta }}_1$	−0.036 (−0.054, −0.018)	0.009	−4.151	0.964 (0.948, 0.981)	<0.001
	${\mathsf{\beta }}_2$	0.701 (0.354, 1.048)	0.177	3.963	2.015 (1.425, 2.850)	<0.001
	${\mathsf{\beta }}_3$	0.047 (0.020, 0.074)	0.014	3.473	1.049 (1.021, 1.077)	0.001
Sanya	${\mathsf{\beta }}_0$	−2.391 (−2.612, −2.170)	0.113	−21.076	0.091 (0.073, 0.114)	<0.001
	${\mathsf{\beta }}_1$	−0.031 (−0.049, −0.013)	0.009	−3.516	0.969 (0.953, 0.986)	<0.001
	${\mathsf{\beta }}_2$	0.594 (0.245, 0.943)	0.178	3.341	1.811 (1.278, 2.566)	0.001
	${\mathsf{\beta }}_3$	0.061 (0.036, 0.086)	0.013	4.645	1.062 (1.036, 1.090)	<0.001

* ${\mathsf{\beta }}_0$: intercept, ${\mathsf{\beta }}_1$: slope before the intervention, ${\mathsf{\beta }}_2$: immediate effect of the intervention, ${\mathsf{\beta }}_3$: amount of change in slope before and after the intervention, SE: standard error, z: normal distribution statistic, RR: relative risk, CI: confidence interval.

).

4. Discussion

Urban rail transit is one of the major modes of public transportation. The crowded and closed urban rail transit environment increases the likelihood of COVID-19 virus transmission compared to other modes of travel [19]. The outbreak not only directly affects its frequency of use but also brings new challenges to future urban travel patterns [20].

In previous studies, Musselwhite C. [21] and Damsara K. [22] demonstrated that government-imposed public health restrictions (e.g., home orders, closure of nonessential sites, and restrictions on assemblies) during the epidemic significantly reduced mobility and had a significant impact on public transportation use. However, in the post epidemic era, Tabassum N. [23], Bi W. [24], Ashour L.A. [25], and Lin Y. [26] discussed the recovery strategies and patterns of public transportation. Their studies proved that a lenient anti-epidemic policy contributes significantly to the growth of urban rail transit passenger volume. This is consistent with the changes in passenger volume brought about by the “Class B infectious disease Class B management” policy.

Similar to the research methodology of this paper, Yu J. [27] used the ITS method to demonstrate that the lifting of epidemic control boosted the passenger volume of Beijing’s flex-route transit. Despite the difference in the study population, the results have the same trend as the recovery of urban rail transit passenger volume studied in this paper.

In related studies in other countries, Goodland F. [28], Christoforou Z [29], Bonera M. [30], and Mashrur SkMd [31] examined the changes in public transportation use under the influence of COVID-19 in the UK, the US, Italy, and Canada, respectively, and all countries studied demonstrated that stringent anti-epidemic policies have a negative impact on public transportation use. These restrictive policies may profoundly affect people’s lifestyles and travel choices and may have lasting effects.

This study employed 43 interrupted time series analyses to evaluate the immediate and lasting effects of the policy, implemented in January 2023, on the passenger volume of urban rail transit in the whole country and 42 different cities.

4.1. Reasons and Variations in Policy Impact

4.1.1. Reasons for the Positive Impact of the Policy in the Whole Country

From the overall results of changes in the passenger volume of urban rail transit in 42 cities, we found that after the implementation of the “Class B infectious disease Class B management” policy, the total passenger volume significantly increased, with positive, immediate, and sustained effects. We attribute these positive effects to two main factors.

Easing of Travel Restrictions

In the early stages of the worldwide outbreak, public transportation systems in many cities adopted varying degrees of restriction, and even some urban rail lines were suspended, with significant reductions in pedestrian volume in all types of venues [32]. However, over time, the mutation trend of the virus gradually showed a lower pathogenicity [33]. With the relaxation of the vaccination policy, the transportation system gradually resumed normal service. Following the introduction of the “Class B infectious disease Class B management” policy, urban rail transit stations no longer require passengers to display health information on their smartphones while traveling. This change has made travel more convenient for older adults and people with disabilities who have limited mobility [34]. Additionally, the government eliminated the requirement for regular COVID-19 testing of frontline public transit workers, reducing staff workloads and lowering operational costs, which in turn improved public transit efficiency. In the context of the resumption of work and production, urban residents are gradually returning to their regular commuting patterns. This implies that the home-office lifestyle during the epidemic is shifting [35]. The use of urban rail transit as a necessary commuting tool has grown significantly [36].

Reduction in Travel Fear

The pandemic significantly influenced individuals’ travel choices [37], with public transportation usage largely dependent on perceived comfort and safety [38]. Previous studies showed that people’s fear of crowded environments increased significantly during the epidemic, and some groups preferred to avoid crowded public transport due to the high mortality rate of COVID-19 [39], which led to a significant decline in urban rail transit demand. After the implementation of the “Class B infectious disease Class B management” policy, urban rail transit stations intensified their care and assistance services for passengers with fever or physical discomfort, ensuring that they can quickly access treatment and recovery support. These measures alleviated public fear of virus transmission, strengthened public trust in the transportation system, and created favorable conditions for the recovery of urban rail transit passenger volume.

4.1.2. Reasons for the Different Impacts of the Policy in Certain Cities

From the perspective of specific cities, we observed that while the “Class B infectious disease Class B management” policy led to a significant increase in urban rail transit passenger volume in most cities, the magnitude of this change varied considerably. Prior studies have highlighted the spatial heterogeneity in urban rail usage across different cities [40]; based on the context of the pandemic and policy implementation, we summarize the factors contributing to these differences in passenger volume changes.

Variations in Policy Implementation and Enforcement

There are differences in the intensity of policy implementation among local governments. Some cities took a cautious approach, retaining certain preventive measures. For example, Guiyang lifted the requirement for a valid nucleic acid test and health code checks in its urban rail transit on 8 December 2022, while Harbin delayed similar adjustments until 6 January 2023. These discrepancies in enforcement may have influenced residents’ confidence in urban rail travel, affecting passenger volume. Additionally, cities differ in the efficiency of promoting policy information. Cities with rapid information dissemination enable residents to stay informed regarding policy changes in a timely manner, directly affecting their travel decisions [41].

Regional Differences in Public Travel Needs

The travel needs of citizens are influenced by various factors in different cities, including city size, population structure, and lifestyle [42]. For example, in large cities and densely populated areas, residents have a high dependence on urban rail transit and a relatively strong demand for travel [43]. With the adjustment of epidemic policies, the increase in migrant workers and tourists from other places has further promoted the increase in the passenger volume of urban rail transit. In addition, there are differences in medical resources and epidemic prevention and control capabilities among different cities, which directly affects the confidence and willingness of citizens to travel [44]. Cities with sufficient medical resources and effective prevention and control measures can quickly handle confirmed cases, alleviate citizens’ fear, and enhance their travel confidence. After the implementation of the “Class B infectious disease Class B management” policy, the passenger volume of urban rail transit in these cities recovered faster.

4.1.3. Reasons for the Lack of Impact of the Policy in Certain Cities

From the perspective of immediate effects, the “Class B Infectious Disease Class B Management” policy has had a significant positive impact on the recovery of urban rail transit passenger volume in most cities. Starting from the second month of policy implementation, 41 out of the 42 cities studied showed a significant increase in rail transit passenger volume. The only city that did not show significant immediate effects was Xiamen. On the one hand, Xiamen’s economy is mainly driven by tourism and service industries, which have been greatly impacted during the pandemic. Due to the long time required for the recovery of the tourism and service industries to adapt to policy adjustments, the travel needs of Xiamen residents have not been addressed as quickly as other cities that mainly rely on industrial, commercial, or large-scale commuting, resulting in an unclear recovery of rail transit passenger flow. On the other hand, Xiamen has a subtropical maritime climate that is suitable for outdoor activities. After the relaxation of prevention and control policies, residents may choose more open modes of transportation (such as walking and cycling) rather than relying on closed and densely populated rail transit systems. In addition, due to Xiamen’s coastal location and island features, many areas still rely on other public transportation methods such as buses and ferries. This allows citizens to avoid excessive reliance on urban rail transit after relaxing epidemic prevention and control policies, further weakening the direct impact of policies on the recovery of urban rail transit passenger volume.

From the perspective of lasting effects, the “Class B Infectious Disease Class B Management” policy has had a significant positive and lasting impact on the recovery of urban rail transit passenger volume in most cities, but the effect is not obvious in Qingdao, Ningbo, Changzhou, Nanning, Harbin, Jinan, and other places. For coastal ports or tourist cities such as Qingdao and Ningbo, they have more diverse modes of transportation, such as buses and ferries. The diversified transportation modes have diverted the passenger volume of rail transit, resulting in a weakened lasting effect. For cities with strong regional characteristics such as Nanning and Harbin, they are influenced by special climate conditions (Nanning’s tropical climate and Harbin’s cold climate). Resident travel patterns have seasonal or regional characteristics and are less affected by policies. The situation in Jinan is the most unique, as the main urban area of Jinan is filled with groundwater, and the rail transit lines are planned and laid out around the city. Most citizens do not choose urban rail transit as their main mode of commuting, which makes it difficult for policy changes to affect changes in passenger volume.

4.2. Recommendations and Suggestions

The “Class B infectious disease Class B management” policy provides robust policy support for the recovery of urban public transportation, especially urban rail transit. As we enter the post-epidemic era, urban rail transit serves as a vital tool for daily commuting [45], alleviating urban traffic pressure and promoting green travel. To enhance commuting efficiency and improve residents’ quality of life, governments and transportation management departments should fully leverage the flexibility afforded by “Class B Management” to enhance the safety and convenience of urban rail transit services, rebuild public confidence, and actively promote the comprehensive recovery and sustainable development of cities. To this end, we propose the following specific policy recommendations aimed at providing operational and forward-looking guidance for urban rail transit planning and management.

4.2.1. Integration of Multi-Modal Transportation

To restore the passenger volume of urban rail transit, it is necessary to strengthen the connectivity of transportation hubs and create convenient and efficient transfer experiences [46]. For coastal and tourist cities such as Qingdao, Ningbo, and Xiamen, it is recommended to build a public transportation network that seamlessly connects rail transit with buses, ferries, and other modes of transportation. Through a unified ticketing platform and a multi-mode transfer system, information interconnection and fare linkage between various modes of transportation can be achieved. In addition, in important transfer stations and transportation hub areas, it is recommended to strengthen the construction of pedestrian environments and transfer facilities, optimize travel routes, and enhance the overall travel experience.

4.2.2. Adjustments in Operational Planning

Since the passenger volume of urban rail transit formed an obvious double-peak pattern during the epidemic, the passenger volume was more dispersed in other periods [1]. Therefore, it is recommended that operators flexibly adjust the frequency of train departures and operating schedules based on the distribution of peaks and valleys of passenger volume. In addition, passenger volume management can be strengthened at the station level. For example, Beijing has experimented with the practice of “subway reservation”, which implements a booking mode and provides passengers with a 30 min time slot to enter the station in order to reduce congestion at the station entrances during peak hours.

4.2.3. Design of Station Environments

Rich social activities and better public service facilities can promote the recovery of rail transit passenger volume [47]. Therefore, commercial complexes, office buildings, residential areas, and other functional areas near urban rail transit stations can be reasonably planned to enrich the stations’ functions; green landscapes and open spaces can be added to create a comfortable and pleasant public environmental atmosphere. In addition, the government should also vigorously promote the development of social activities and cooperate with enterprises to organize various community activities, cultural performances, commercial district promotions, etc., so as to attract residents to actively participate [48]. Event venues are prioritized near urban rail stations, thus enhancing travel demand.

4.2.4. Health and Safety Protection Measures

To enhance passengers’ perceived travel safety, the cleaning and disinfection management of urban rail transit stations should be strengthened. The health monitoring system should be promoted, gradually establishing a standardized health management network [38]. Additionally, health and safety awareness campaigns should be reinforced through advertising screens, public announcements, and other channels to improve passengers’ self-protection awareness and encourage healthy travel habits [49]. Furthermore, contactless payment should be encouraged, and the scanning entry process should be optimized to minimize contact risks during ticketing and entry.

For vulnerable groups, such as persons with disabilities and the elderly who rely on public transportation [50], additional accessibility features, such as priority seating and designated green channels, should be installed in urban rail transit stations. Specialized health consultation and emergency assistance service windows should be established to ensure their safety in the post-pandemic era.

4.2.5. Data-Driven Management Approaches

We should develop a passenger volume monitoring system based on IoT and big data analytics to collect and analyze real-time passenger volume data, providing robust data support for operational strategy adjustments. By integrating data and leveraging intelligent analysis, the system supports operational decision-making and enables timely adjustments to scheduling, pricing, and service strategies. Existing studies indicate that pandemic-induced declines in passenger volume have exacerbated financial pressures on urban rail transit operators [9]. By conducting in-depth analyses of historical passenger volume data, future trends can be predicted, allowing for optimized frequency scheduling and departure intervals. This ensures sufficient capacity during peak hours while preventing resource waste during off-peak periods. Additionally, a regular evaluation mechanism should be established to continuously optimize service levels and contingency plans by integrating passenger satisfaction surveys, operational data, and social feedback, ensuring that policy adjustments are promptly reflected in operational management.

These recommendations are grounded in this study’s findings on the effects of the “Class B infectious disease Class B management” policy while also considering the regional characteristics of each city. By integrating multi-modal transportation, adjusting operational planning, designing station environments, implementing health and safety protection measures, and adopting data-driven management approaches, the service quality and operational efficiency of urban rail transit systems can be further enhanced. This approach ensures a safer, more convenient, and more comfortable travel experience for urban residents while also providing a replicable and sustainable model for future public transportation planning and management in similar policy contexts.

4.3. Study Limitations

This study has certain limitations. First, as an approach within a quasi-experimental design, interrupted time series analysis has a lower capacity for causal inference than randomized controlled trials, making it challenging to establish a direct causal link between the “Class B infectious disease Class B management” policy and variations in urban rail transit travel behavior across different conditions. Additionally, due to the limited availability of monthly urban rail transit passenger volume data before 2021 across cities in China, this study only analyzed data from the post-epidemic period (January 2021 to June 2024). To deepen the understanding of the impact of pandemic policies on urban rail transit passenger volume, future studies should incorporate data from the initial stages of the pandemic to capture passenger volume trends across pre-, mid-, and post-epidemic phases. Lastly, urban rail transit passenger volume is influenced by a range of factors, including residents’ socioeconomic attributes and travel behavior patterns. Future research could integrate questionnaire surveys with large-scale data to gain more comprehensive results. A more nuanced sample of influencing factors would further enhance the explanatory power and credibility of the model if sufficient data are available.

Although these limitations may have some impact on the interpretation of the results, we introduced time variables and intervention indicator variables in our methodological design. This approach is effective in controlling for lasting trends and seasonal fluctuations inherent to each city, ensuring the robustness of the main findings. We will further extend the data coverage and incorporate more economic and social variables in subsequent studies to enhance the broad applicability and explanatory power of this study.

5. Conclusions

Since the beginning of the 21st century, human society has encountered several respiratory infectious diseases, such as SARS, avian influenza, MERS, and COVID-19. Massively spreading viruses have seriously affected people’s work and travel. Especially in the current high-density urban space, crowded urban rail transit inevitably accelerates the spread of viruses. In this context, existing compact urban planning strategies should be reevaluated, and the adoption of resilient urban management strategies should be expedited to foster safer and more sustainable urban travel environments.

To the best of our knowledge, this study is the first to employ an interrupted time series (ITS) design, integrating quasi-Poisson regression modeling with counterfactual scenario analysis. The findings confirm that following the implementation of the “Class B Management of Class B Infectious Diseases” policy, urban rail transit passenger volume increased significantly in 42 Chinese cities. Specifically, the national average monthly passenger volume rose by approximately 101.34% after policy implementation relative to the projected value in the absence of the policy. The results of this study show that the policy of “Class B Management of Class B Infectious Diseases” has a significant positive immediate effect (significant in 41 cities) and a significant positive lasting effect (significant in 33 cities). This finding highlights the benefits of reduced travel restrictions and the restoration of public confidence resulting from the relaxation of prevention and control measures. Moreover, although the policy has generally facilitated a nationwide rebound in urban rail transit passenger volume, notable disparities exist among cities.

Finally, based on this study’s findings, we propose recommendations for government policymakers and transportation planners in five key areas: the integration of multi-modal transportation, adjustments in operational planning, station environment design, health and safety protection measures, and data-driven management approaches. These measures aim to enhance the safety and convenience of urban public transportation services in the face of potential future infectious disease threats.