Impact of smoke-free legislation on cigarette price and affordability in Wuhan, China: a repeated cross-sectional study

Abstract

Background

The smoke-free legislation (SFL) in Wuhan, China was adopted in January 2020. While its health benefits are documented, the impact on cigarette prices and affordability among smokers remains unclear. This study assessed the impact of Wuhan’s SFL on cigarette prices and affordability.

Methods

Three rounds of repeated cross-sectional surveys were conducted in 2019 (pre-legislation), 2022 and 2024 (post-legislation), following the Global Adult Tobacco Survey (GATS) protocol. A multistage cluster sample design selected representative samples of non-institutionalized permanent residents aged ≥ 15 years. Analysis was restricted to current manufactured cigarette smokers. Cigarette prices were adjusted for inflation and purchasing power parity (PPP) to constant 2024 US dollars. Cigarette affordability was measured as the average cost of 100 packs as a percentage of GDP per capita. Complex sampling designs, ordered logit models and marginal effects analyses were employed.

Results

A total of 8,424 participants were interviewed, with response rates of 94.5% (2019), 92.8% (2022) and 100% (2024). The average amount spent on 20 cigarettes did not change significantly from $5.3 (95%CI: 4.7–5.8) in 2019 to $5.2 (95%CI: 4.2–5.9) in 2024. A wide price range persisted, with mean prices in the lowest quintile ranging from $1.88 to $3.73 and in the highest quintile from $10.72 to $13.60. Cigarette affordability increased significantly. The cost of 100 packs as a percentage of GDP per capita decreased from 2.8% (95%CI: 2.5–3.1) in 2019 to 2.2% (95%CI: 1.8–2.5) in 2024. Compared to 2019, the odds of being in the higher-affordability category increased by 18.2% points in 2024 (p < 0.05). The overall odds of choosing different price-tiers did not change significantly. Smokers who were male, younger (15–24 years), highly educated (college and above), employed in government/corporate sectors, or had no intention to quit had higher odds of choosing upper price-tier cigarettes and having higher cigarette affordability.

Conclusions

Following SFL implementation in Wuhan, cigarette prices remained stable while affordability increased significantly. The wide price range enables potential substitution to cheaper brands. SFL alone is insufficient to curb consumption. National tobacco control legislation should integrate robust tax and floor pricing policies.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12889-026-28693-3.

Introduction

Tobacco use, including active smoking and secondhand smoke (SHS) exposure, remains one of the most significant threats to global public health, causing over 7 million deaths annually [1, 2]. In China, the smoking prevalence of adult males was 43.9% recently, and smoking has become the second-leading cause of disability-adjusted life years (DALYs) [2, 3]. The World Health Organization Framework Convention on Tobacco Control (WHO FCTC) identifies comprehensive smoke-free legislation (SFL) as a core intervention of the MPOWER measures [4]. MPOWER is a package of six evidence-based tobacco control strategies introduced by WHO: Monitor tobacco use and prevention policies, Protect people from tobacco smoke, Offer help to quit tobacco use, Warn about the dangers of tobacco, Enforce bans on tobacco advertising, promotion and sponsorship, and Raise taxes on tobacco. Previous studies have demonstrated that the comprehensive SFL leads to rapid reductions in SHS exposure, improvements in respiratory health, and a decrease in hospital admissions for conditions such as acute coronary syndromes [5–7]. While 74 countries worldwide have implemented comprehensive national SFLs, China has no national-level SFL [8, 9]. Currently, only a handful of cities, including Wuhan, have adopted local SFL [10].

The Wuhan SFL, enacted in January 2020, banned smoking in indoor public places, indoor workplaces, and public transportation, established a mechanism of multi-departmental coordinated supervision, and defined enforcement measures, including fines and penalties for individuals or venues. The legislative framework encompasses clear tobacco control responsibilities across various industries and venues, as well as public participation channels, such as the Mayor’s hotline, to address tobacco control-related complaints, and third-party evaluations to achieve a smoke-free environment [11]. One study indicated that the implementation of the SFL had a significant positive impact on reducing nonsmokers’ exposure to SHS in indoor public places [12]. While the health benefits of SFL are well-documented, a critical evidence gap persists regarding its impact on smokers’ consumption behaviors and the economic dimensions of tobacco use, particularly in high-smoking-prevalence environments like China. Specifically, the effects of SFL on cigarette prices and affordability are not fully understood.

As household disposable incomes grow, cigarette affordability becomes a crucial determinant of cigarette consumption. Tax and price policies are the most effective interventions for reducing tobacco use, as emphasized by Article 6 of the WHO FCTC [13, 14]. However, these fiscal measures operate through different mechanisms than smoke-free policies. While the Wuhan SFL established smoke-free environments, it was not accompanied by complementary tax increases, and China’s tobacco tax structure remained unchanged during the study period [11]. Understanding the economic context is nonetheless important, as the range of cigarette prices available within a market is a significant factor related to tobacco consumption [15]. A wider price range may prompt smokers to shift to cheaper tobacco products, thereby undermining the effectiveness of the SFL [16]. Previous studies have noted that China has one of the widest price ranges globally primarily due to its complex tiered cigarette tax structure [15]. More complex cigarette tax structures result in a greater price range, which may increase opportunities for smokers to switch to cheaper cigarettes rather than quit smoking to avoid the full impact of tax increases [17].

While the health benefits of SFL are well-documented, a critical evidence gap persists regarding its impact on smokers’ consumption behaviors and the economic dimensions of tobacco use, particularly in high-smoking-prevalence environments like China. Specifically, the effects of SFL on cigarette prices and affordability are not fully understood. Therefore, the study aims to assess the impact of Wuhan’s SFL on prices and affordability of cigarettes among current smokers. Data were from three waves of the Wuhan Adult Tobacco Survey (WATS), following the protocol of the Global Adult Tobacco Survey (GATS), conducted in 2019, 2022, and 2024. We analyze the level, range, and affordability of cigarettes [16, 18], and explore potential factors related to cigarette price-tier choices and affordability.

Methods

Sampling design

The WATS employed a multistage cluster sample design to generate representative data [22]. First, three in 2019, four in 2022, and four in 2024 administrative districts were selected from all 15 administrative districts in Wuhan, China, with the simple random sampling (SRS) method. Second, the SRS method was employed to select four streets in 2019, three streets in 2022, and four streets in 2024. Third, three in 2019, two in 2022, and two in 2024 neighbourhood committees were selected using the SRS method. Fourth, certain family households, 90 in 2019, 145 in 2022, and 120 in 2024, were selected within each neighbourhood committee with the SRS method. Fifth, one family member was chosen from each household using the Kish table method [23]. Once 75 in 2019, 120 in 2022, and 100 in 2024 family members in each neighbourhood committee were interviewed, the survey was finished in that neighbourhood committee. The fourth and fifth stage contributes to a high response rate. While the numbers of primary sampling units (districts, streets, neighbourhood committees, and households) varied across survey waves to accommodate logistical considerations, the core multistage cluster sampling methodology and random selection procedures remained consistent with GATS protocols. Sampling weights were calculated for each wave to account for the probability of selection at each stage, and post-stratification calibration was applied to ensure representativeness of the Wuhan population, enabling valid comparisons across survey years.

Questionnaire

The GATS core questionnaire consists of nine sections, including background characteristics, tobacco smoking, electronic cigarettes (ever use, current frequency, duration, and reasons such as quitting or perceived reduced harm), smokeless tobacco (current and past use of snuff, chewing tobacco, and betel quid, frequency, and time to first use), cessation (past-year quit attempts, duration of last attempt, methods including pharmacotherapy and e-cigarettes, and healthcare advice received, SHS exposure (home and workplace smoking rules and exposure in public venues including government buildings, healthcare facilities, restaurants, bars, and public transport), economics (last purchase of manufactured cigarettes: unit, quantity, cost, brand, and place), media (exposure to anti‑tobacco information, health warnings and their impact on thoughts of quitting, and tobacco advertising and promotions), knowledge, attitudes and perceptions (beliefs that smoking and smokeless tobacco cause serious illness and specific diseases, relative harm perceptions of newer products, and support for policies such as smoking bans and tax increases) [19]. The WATS Questionnaire was employed, which added some additional questions to the GATS core questionnaire to account for China’s specific characteristics. As Chinese is a common and official language, the simplified Chinese version of the WATS questionnaire was used.

Data collection

Survey management staff were from the municipal and district Centres for Disease Control and Prevention (CDC), while field interviewers and field supervisors were from Community Health Centre (CHC). Before the fieldwork, two levels training, municipal and district, were held, and all personnel participated and passed the on-site exams. Field investigation conducted from August to November in 2019, 2022 and 2024, 2–4 weeks after the training. In the field investigation, face-to-face interviews were conducted at participant’s home or in other public places at their convenience, using an electronic data collection device. Should participants be unable to complete the questionnaire due to low educational levels or poor eyesight, the field interviewer shall neutrally interview them and complete the questionnaire on their behalf. In addition, participants were sent small gifts as an incentive after they successfully completed the interview. If the individuals were already participants, they could withdraw at any time without penalty or loss of benefits. Strict quality control was applied to the whole investigative process. The investigator complied with the GATS protocol during all processes to ensure high-quality data. Informed consent was obtained before the questionnaire, and individuals aged 15–17 years were required to provide their parents’ or guardians’ consent to participate. This study, in accordance with the Declaration of Helsinki, was approved by the Ethics Committee of the Wuhan Centre for Disease Control and Prevention (CDC), China (WHCDCIRB- K- 2019012, WHCDCIRB- K- 2022019).

Measurements

Cigarette price, the average cost of a pack of cigarettes (20 sticks) is calculated by first determining the price paid per cigarette and then multiplying by 20. A “manufactured cigarette weight” is created by multiplying the individual sampling weight by the number of manufactured cigarettes smoked per day. Finally, the weighted average amount paid per pack of manufactured cigarettes across all individuals is calculated using the new “manufactured cigarette weight.” The sampling weight is calculated in accordance with the GATS statistical analysis guidelines [24, 37]. Cigarette affordability was assessed using the relative income price (RIP) concept, defined here as the ratio of the average cost of 100 packs (2,000 sticks) of manufactured cigarettes to per capita GDP, expressed as a percentage [18]. A smaller percentage indicates that cigarettes are more affordable relative to income [25]. While per capita GDP does not capture within-country income distribution, it serves as a standardized metric for cross-country and over-time comparisons of affordability [15]. Cigarette prices, initially collected in Chinese Yuan (CNY), were converted to Purchasing Power Parity adjusted United States Dollars (PPP USD) using year-specific PPP conversion factors [26], which account for differences in the cost of living across settings [15]. Cigarette prices were further adjusted for inflation relative to 2024 levels to ensure comparability across the three survey waves [16, 27]. PPP conversion factors, inflation adjustment data, and per capita GDP figures were obtained from the International Comparison Program (ICP), the World Bank (WB), and the International Monetary Fund (IMF) World Economic Outlook Database [27–29].

The primary dependent variables in this study are cigarette price and cigarette affordability. For the regression analysis, two ordinal dependent variables were constructed. The first, price-tier choice, was derived from the weighted average cost of a pack of cigarettes and categorized into low, middle, and upper-tier brands by price tertiles, with tertile cut-off points calculated from the pooled data of all three survey waves. The second, cigarette affordability, was classified into low, middle, and higher affordability categories using the same pooled-data approach, based on the distribution of the affordability index across the combined sample.

The main independent variable of interest is the implementation of the smoke-free legislation (SFL), represented by the survey year (2019 as pre-legislation reference, compared to 2022 and 2024 as post-legislation years). Based on the research purpose and previous studies [15–17, 26, 30–32], a range of independent and control variables were included: demographic characteristics (sex, age, education level, occupation), smoking-related behaviors (average number of cigarettes smoked per day, attempt to quit smoking in the past 12 months, interest in quitting smoking), and exposure to secondhand smoke (SHS) in public places in the past 30 days.

Attempt to quit smoking past 12 months was measured by answering “Yes, in the past 12 months” to the question “During the past 12 months, have you tried to stop smoking?” [33]. Interest in quitting smoking past 12 months was defined as answering either “Quit within the next month” or “Thinking within the next 12 months” to the question “Which of the following best describes your thinking about quitting smoking?” [34]. Exposed to SHS in public places in the past 30 days was obtained from the question: “Did anyone smoke inside of any government buildings or government offices, health care facilities, restaurants, public transportation, university buildings, or school buildings that you visited in the past 30 days?”. The indicator is calculated as individuals who visited any of the public places (government buildings/ healthcare facilities/ restaurants/ public transportation/ university/or school) in the past 30 days and were exposed to SHS inside it [35].

All variables were defined and measured following GATS standard protocol [19, 24], ensuring consistency with international tobacco surveillance systems and enabling comparability with studies from other countries.

Statistical analysis

The statistical analysis, including the construction of sampling weights and the use of complex survey designs, follows GATS guideline [36, 37] to ensure that estimates are representative of the Wuhan adult population and comparable across survey waves. Data cleansing rules were created to ensure accuracy and eliminate internal inconsistencies [38]. Since the survey employed a multi-stage sampling procedure, the three waves of data were weighted: calculation of the sampling weight, non-response adjustment, and post-stratification calibration adjustment of the sample totals to the known population totals [37]. All analyses were based on complex survey designs [36]. Descriptive statistics were used to describe the 50th quantile (median) for the average cost of a pack of cigarettes and cigarette affordability. The correlates of two ordinal dependent variables were examined using ordered logit models [39]. For these ordinal dependent variables, the tertile cut-off points were derived from the pooled sample across all three survey years to ensure a fixed, comparable definition of ‘low,’ ‘middle,’ and ‘upper/higher’ categories over time. The first variable, price-tier choice, was constructed from the weighted average cost of a pack of cigarettes and categorized into low, middle, and upper-tier brands by price tertiles. The second, cigarette affordability, was classified into low, middle, and higher affordability. Results are presented as regression coefficients, and to aid interpretation, average marginal effects are reported to illustrate the change [16]. Statistical inferences were assessed by a two-sided 5% significance level. Data cleaning, weighting, and analysis were conducted using SAS software version 9.4 (SAS Institute Inc., Cary, NC) and Stata version 18 (StataCorp LLC, Texas, USA), with SAS primarily used for data management and weight calculation following GATS guidelines, and Stata used for ordered logit models and marginal effects estimation; key results were cross-validated using both programs to ensure accuracy.

Study population

The WATS is a representative household survey conducted in accordance with the standardized methodology of the Global Adult Tobacco Survey (GATS) to maintain consistency and international comparability [19]. Three waves of repeated cross-sectional WATS in 2019, 2022, and 2024 were conducted following GATS protocols. The target population includes all non-institutionalized permanent residents aged 15 years or above who consider their current address their usual residence. Military and other institutionalized populations were excluded [20]. Given that the prevalence of manufactured cigarettes among Chinese residents stands at 94.5% [21], we restricted our analysis to current manufactured smokers who smoked at least once per week (abbreviated as current smokers).

Results

Participant characteristics

A total of 8,424 participants were interviewed across three survey waves. In the pre-legislation wave, 2,551 out of 2,700 (94.5%) targeted subjects were interviewed in 2019. In the post-legislation waves, 2,673 out of 2,880 (92.8%) targeted subjects were interviewed in 2022, and 3200 out of 3200 (100%) targeted subjects were interviewed in 2024. The sample characteristics of all participants and current smokers are for all independent and control variables across the three survey waves are presented in Supplementary Table 1. The average ages in 2019, 2022, and 2024 were 49.35 ± 0.29, 49.32 ± 0.31, and 52.51 ± 0.28, respectively; the male-to-female ratios were 0.99:1 in 2019, 0.90:1 in 2022, and 0.92:1 in 2024, respectively. The differences in weighted proportions of demographic characteristics among current smokers, such as sex, age, and education, across survey years were not statistically significant across survey years. In relation to smoking patterns, the change in the number of cigarettes smoked per day among current smokers was not significant.

The average cost of a pack of cigarettes and cigarette affordability pre- and post-legislation

As shown in Table 1, the range for the average cost of a pack of cigarettes among current smokers by arranging individual prices paid into quintiles and estimating the means of each quintile for the years 2019, 2022, and 2024 (Table 1). The range of cigarette prices, as indicated by the difference between the highest and lowest quintile means, varied over time. In 2019, the mean price in the lowest quintile was 1.88 PPP USD, while that in the highest quintile was 10.72 PPP USD, resulting in a range of 8.84 PPP USD. In 2022, the range widened to 11.92 PPP USD (Q1: 2.84, Q5: 14.76). By 2024, the range narrowed to 9.87 PPP USD (Q1: 3.73, Q5: 13.6).

Table 1.

Range of average amount spent on 20 manufactured cigarettes, in Purchasing Power Parity adjusted constant 2024 dollars (PPP USD)

Year	n	Mean of Quintile 1 of prices paid	Mean of Quintile 2 of prices paid	Mean of Quintile 3 of prices paid	Mean of Quintile 4 of prices paid	Mean of Quintile 5 of prices paid	Range (Q5 − Q1)
2019	825	1.88	2.91	4.84	5.97	10.72	8.84
2022	796	2.84	4.38	5.66	6.76	14.76	11.92
2024	938	3.73	5.35	5.63	7.06	13.60	9.87

n represents the unweighted number of current smokers in each survey year

Table 2 presents cigarette prices and affordability among current smokers by selected demographic characteristics. The median for the average cost of a pack of cigarettes was $5.3 (4.7, 5.8) in 2019, $5.2 (3.8, 6.6) in 2022, and $5.2 (4.2, 5.9) in 2024. In the ordered logit model (Table 3), the coefficients for the 2022 and 2024 year dummies were not statistically significant. The average cost of 100 packs of manufactured cigarettes as a percentage of GDP per capita was 2.8% (2.5, 3.1) in 2019, 2.4% (1.7, 3.1) in 2022, and 2.2% (1.8, 2.5) in 2024. However, the ordered logit model for affordability (Table 4) showed that the odds of being in a higher affordability category increased significantly in 2024 compared to 2019 (β = 1.003, p < 0.05).

Table 2.

The average cost of a pack of cigarettes, and cigarette affordability among current smokers in the pre-legislation (2019) and post-legislation (2022, 2024) periods, in Purchasing Power Parity adjusted constant 2024 dollars (PPP USD) in Wuhan, China

	Average Cost of a Pack of Cigarettes (PPP USD) (Median 95%CI)			Cigarette Affordability* (% of GDP per capita) (Median 95%CI)
	Year = 2019	Year = 2022	Year = 2024	Year = 2019	Year = 2022	Year = 2024
Sex
Male	5.3(4.8,5.8)	5.2(3.6,6.8)	5.2(4.1,6.3)	2.8(2.4,3.1)	2.4(1.7,3.1)	2.2(1.7,2.6)
female	4.2(3.7,4.8)	5.1(3.8,6.3)	4.4(3.2,5.7)	2.2(1.9,2.6)	2.3(1.8,2.9)	1.8(1.3,2.4)
Age (years)
15–24	6.0(5.4,6.5)	6.7(6.2,7.2)	5.2(3.5,6.9)	3.2(2.8,3.5)	3.1(2.7,3.4)	2.2(1.4,2.9)
25–44	5.4(4.9,6.0)	5.2(3.6,6.8)	5.2(4.9,5.5)	2.9(2.5,3.2)	2.4(1.7,3.1)	2.2(2.1,2.3)
45–64	3.9(3.4,4.5)	5.2(3.5,6.9)	5.1(4.3,6.0)	2.1(1.7,2.4)	2.4(1.6,3.2)	2.1(1.8,2.5)
≥ 65	2.7(2.2,3.3)	4.6(3.7,5.4)	3.7(3.4,3.9)	1.4(1.1,1.8)	2.1(1.7,2.5)	1.5(1.4,1.6)
Education level
Junior high school or below	2.8(2.2,3.3)	5.1(3.9,6.2)	5.1(4.5,5.8)	1.5(1.1,1.8)	2.3(1.8,2.9)	2.1(1.9,2.4)
Senior high school	5.4(4.1,6.8)	5.2(3.9,6.5)	5.2(4.6,5.8)	2.9(2.1,3.6)	2.4(1.8,3.0)	2.2(1.9,2.4)
College and above	5.5(5.0,6.1)	5.7(2.5,8.8)	6.4(4.1,8.8)	2.9(2.6,3.3)	2.6(1.2,4.0)	2.7(1.7,3.7)
Occupation
Government/public institute employees	6.7(3.4,10.0)	5.3(2.6,7.9)	6.7(4.5,8.8)	3.5(1.8,5.3)	2.4(1.2,3.7)	2.8(1.9,3.7)
Corporate/business/service employees	5.4(3.4,7.5)	5.2(4.7,5.8)	5.2(4.3,6.1)	2.9(1.8,3.9)	2.4(2.1,2.8)	2.2(1.8,2.5)
Others	5.4(3.9,6.8)	5.2(3.5,6.9)	5.2(4.0,6.3)	2.8(2.1,3.6)	2.4(1.6,3.2)	2.2(1.7,2.6)
Farmers	2.2(2.1,2.3)	3.9(3.4,4.5)	2.0(1.4,2.5)	1.2(1.1,1.2)	1.8(1.5,2.1)	0.8(0.5,1.2)
Average number of cigarettes smoked per day
1–4	5.1(4.6,5.7)	5.3(4.7,5.8)	5.2(4.2,6.2)	2.7(2.4,3.1)	2.4(2.1,2.8)	2.2(1.8,2.6)
5–9	5.3(4.8,5.8)	5.2(4.6,5.7)	5.1(4.6,5.6)	2.8(2.5,3.0)	2.4(2.0,2.7)	2.1(1.9,2.3)
10–19	5.4(3.4,7.4)	5.3(3.4,7.1)	5.2(4.1,6.3)	2.8(1.8,3.9)	2.4(1.6,3.3)	2.2(1.7,2.6)
≥ 20	3.9(3.4,4.5)	5.1(3.5,6.6)	5.2(3.8,6.5)	2.1(1.7,2.4)	2.3(1.6,3.1)	2.2(1.6,2.7)
Attempt to quit smoking past 12 months
Yes	5.1(4.6,5.6)	5.2(4.6,5.7)	5.2(3.6,6.9)	2.7(2.3,3.0)	2.4(2.0,2.7)	2.2(1.5,2.9)
No	5.0(4.5,5.5)	5.2(4.4,6.0)	5.2(4.1,6.3)	2.6(2.3,3.0)	2.4(2.0,2.8)	2.2(1.7,2.6)
Interest in quitting smoking past 12 months
No	5.3(4.8,5.8)	5.2(3.9,6.6)	5.2(4.5,5.9)	2.8(2.5,3.1)	2.4(1.8,3.0)	2.2(1.9,2.5)
Yes	2.8(2.3,3.3)	3.9(2.7,5.0)	5.0(4.5,5.5)	1.5(1.1,1.8)	1.8(1.3,2.3)	2.1(1.7,2.4)
Exposed to SHS in public place past 30 days
Yes	5.4(4.9,5.9)	5.2(3.6,6.9)	5.2(4.1,6.3)	2.8(2.5,3.2)	2.4(1.7,3.2)	2.2(1.7,2.6)
No	4.0(3.5,4.5)	5.1(3.8,6.5)	5.2(4.2,6.1)	2.1(1.8,2.5)	2.4(1.7,3.0)	2.2(1.7,2.6)

Table 3.

Correlates of average cost of a pack of cigarettes in Purchasing Power Parity adjusted constant 2024 dollars (PPP USD): ordered logit model coefficients and marginal effects

	Coefficients (β, 95% CI)	Marginal effects (β, 95% CI)
		Upper price-tier brands	Middle price-tier brands	Low price-tier brands
Year(ref = 2019)
2022	0.022 (-0.950, 0.994)	0.003 (-0.136, 0.142)	-0.001 (-0.012, 0.013)	-0.003 (-0.155, 0.149)
2024	0.499 (-0.397, 1.395)	0.079 (-0.058, 0.216)	-0.007 (-0.032, 0.018)	-0.072 (-0.217, 0.073)
Sex (ref=Female)
Male	0.757 (0.469, 1.045)***	0.101 (0.071, 0.131)***	-0.019 (-0.027, 0.066)	-0.120 (-0.179, -0.062)***
Age (years) (ref = ≥ 65)
15–24	1.670 (0.459, 2.882)**	0.289 (0.018, 0.561)*	-0.081 (-0.267, 0.105)	-0.208 (-0.327, -0.089)**
25–44	0.533 (-0.223, 1.290)	0.077 (-0.032, 0.186)	-0.005 (-0.025, 0.036)	-0.082 (-0.197, 0.033)
45–64	0.350 (-0.084, 0.783)	0.048 (-0.009, 0.105)	-0.007 (-0.017, 0.032)	-0.056 (-0.128, 0.017)
Education level (ref=Junior high school or below)
Senior high school	0.610 (0.303, 0.916)***	0.090 (0.050, 0.129)***	0.006 (-0.027, 0.038)	-0.096 (-0.144, -0.047)***
College and above	1.169 (0.433, 1.904)**	0.193 (0.057, 0.330)**	-0.029 (-0.118, 0.060)	-0.164 (-0.248, -0.080)***
Occupation (ref=Farmers)
Government/public institute employees	2.529 (0.724, 4.333)**	0.319 (0.160, 0.478)***	0.127 (-0.109, 0.363)	-0.446 (-0.763, -0.129)**
Corporate/business/service employees	1.943 (0.464, 3.421)*	0.208 (0.137, 0.279)***	0.168 (-0.078, 0.414)	-0.375 (-0.673, -0.078)*
Others	1.792 (0.042, 3.543)*	0.183 (0.065, 0.301)**	0.171 (-0.072, 0.413)	-0.353 (-0.693, -0.013)*
Average number of cigarettes smoked per day(ref = 1–4)
5–9	-0.248 (-0.856, 0.361)	-0.039 (-0.134, 0.057)	0.003 (-0.012, 0.019)	0.035 (-0.049, 0.119)
10–19	-0.082 (-0.459, 0.623)	0.014 (-0.075, 0.102)	-0.003 (-0.018, 0.013)	-0.011 (-0.085, 0.063)
≥ 20	-0.219 (-0.782, 0.343)	-0.034 (-0.123, 0.055)	0.003 (-0.011, 0.018)	0.031 (-0.051, 0.113)
Attempt to quit smoking past 12 months(ref = No)
Yes	-0.218 (-0.731, 0.295)	0.035 (-0.050, 0.120)	-0.005 (-0.029, 0.019)	-0.030 (-0.094, 0.034)
Interest in quitting smoking past 12 months(ref = yes)
No	1.238 (0.434, 2.038)**	0.151 (0.069, 0.234)***	0.057 (-0.033, 0.147)	-0.208 (-0.359, -0.058)**
Exposed to SHS in public place past 30 days(ref = No)
Yes	0.286 (-0.001, 0.572)	0.045 (0.001, 0.089)*	-0.005 (-0.022, 0.012)	-0.040 (-0.078, -0.002)*

* p < 0.05, ** p < 0.01, *** p < 0.001

Table 4.

Correlates of Cigarette Affordability in Purchasing Power Parity adjusted constant 2024 dollars (PPP USD): ordered logit model coefficients and marginal effects

	Coefficients (β, 95% CI)	Marginal effects (β, 95% CI)
		Higher affordability	Middle affordability	Low affordability
Year(ref = 2019)
2022	-0.502 (-1.995, 0.990)	-0.095 (-0.379, 0.188)	0.044 (-0.107, 0.196)	0.051 (-0.082, 0.184)
2024	1.003 (0.426, 2.433)*	0.182 (0.087, 0.451)*	-0.069 (-0.222, -0.017)*	-0.113 (-0.234, -0.008)*
Sex (ref=Female)
Male	0.414 (-0.179, 1.007)	0.070 (-0.021, 0.160)	-0.015 (-0.031, 0.001)	-0.055 (-0.137, 0.028)
Age (years) (ref = ≥ 65)
15–24	1.062 (-0.132, 2.255)	0.192 (-0.052, 0.436)	-0.064 (-0.196, 0.068)	-0.128 (-0.245, -0.011)*
25–44	0.587 (0.001, 1.173)*	0.101 (-0.007, 0.209)	-0.023 (-0.058, 0.011)	-0.078 (-0.157, 0.001)
45–64	0.408 (0.077, 0.738)*	0.069 (0.012, 0.125)*	-0.013 (-0.031, 0.005)	-0.056 (-0.103, -0.009)*
Education level (ref=Junior high school or below)
Senior high school	0.568 (0.339, 0.797)***	0.101 (0.066, 0.135)***	-0.024 (-0.038, -0.010)***	-0.077 (-0.116, -0.038)***
College and above	1.212 (0.497, 1.927)**	0.230 (0.094, 0.366)**	-0.088 (-0.174, -0.001)*	-0.143 (-0.210, -0.075)***
Occupation (ref=Farmers)
Government/public institute employees	3.160 (1.201, 5.138)**	0.430 (0.296, 0.564)***	0.107 (-0.203, 0.418)	-0.538 (-0.924, -0.151)**
Corporate/business/service employees	2.699 (0.937, 4.461)**	0.330 (0.225, 0.435)***	0.164 (-0.149, 0.477)	-0.494 (-0.861, -0.128)**
Others	2.548 (0.471, 4.625)*	0.300 (0.130, 0.470)**	0.177 (-0.112, 0.466)	-0.477 (-0.877, -0.078)*
Average number of cigarettes smoked per day(ref = 1–4)
5–9	0.026 (-0.539, 0.591)	0.005 (-0.096, 0.106)	-0.001 (-0.033, 0.030)	-0.003 (-0.073, 0.067)
10–19	0.181 (-0.334, 0.695)	0.033 (-0.060, 0.125)	-0.011 (-0.041, 0.018)	-0.022 (-0.085, 0.042)
≥ 20	-0.080 (-0.560, 0.400)	-0.014 (-0.099, 0.070)	0.004 (-0.021, 0.029)	0.010 (-0.050, 0.070)
Attempt to quit smoking past 12 months(ref = No)
Yes	-0.073 (-0.560, 0.705)	0.013 (-0.100, 0.126)	-0.004 (-0.042, 0.034)	-0.009 (-0.084, 0.067)
Interest in quitting smoking past 12 months(ref = yes)
No	1.135 (0.318, 1.952)**	0.175 (0.071, 0.280)**	-0.010 (-0.050, 0.030)	-0.166 (-0.277, -0.055)**
Exposed to SHS in public place past 30 days(ref = No)
Yes	0.524 (0.233, 0.815)***	0.095 (0.042, 0.147)***	-0.031 (-0.051, -0.011)**	-0.063 (-0.103, -0.024)**

* p < 0.05, ** p < 0.01, *** p < 0.001

Correlates of price-tier choice and cigarette affordability

Tables 3 and 4 show the adjusted ordered logit model and the marginal effects of cigarette prices and affordability among current smokers. The direction of the logit coefficients indicates the cigarette price or affordability category: positive coefficients are associated with higher odds of choosing the upper price-tier brand or being in the higher affordability category, while negative coefficients are associated with higher odds of choosing the low price-tier brand or being in the low affordability category [16].

Regarding the choice of cigarette price tier, as shown in Table 3, the coefficients for the year dummy variables were positive but not statistically significant, indicating no change in price-tier choice between 2019 and 2024. Although this structure remained stable over time, it is significantly related to specific demographic and smoking-related characteristics. Current smokers who were male (β = 0.757, 95%CI: 0.469, 1.045, p < 0.001), who were employed as government/public institute employees (β = 2.529, 95%CI: 0.724, 4.333, p < 0.01), corporate/business/service employees (β = 1.943, 95%CI: 0.464, 3.421, p < 0.05), who were not interested in quitting smoking (β = 1.238, 95%CI: 0.434, 2.038, p < 0.01), whose education level was college and above (β = 1.169, 95%CI: 0.433, 1.904, p < 0.01) or senior high school (β = 0.610, 95%CI: 0.303, 0.916, p < 0.001), and those who were aged 15–24 years (β = 1.670, 95%CI: 0.459, 2.882, p < 0.01), were more likely to purchasing upper price-tier brands. The marginal effects for each category of brand choice showed a similar, consistent, and complementary pattern to the coefficients.

For cigarette affordability (Table 4), the coefficient for the 2024 year dummy variable was positive and statistically significant (β = 1.003, 95%CI: 0.426, 2.433, p < 0.05). The marginal effects show that, relative to 2019, the odds of current smokers in the higher affordability category increased by 18.2% points in 2024, while the odds in the low affordability category decreased by 11.3% points. Current smokers aged 25–44 years (β = 0.587, 95%CI: 0.001, 1.173, p < 0.05) or 45–64 years (β = 0.408, 95%CI: 0.077, 0.738, p < 0.05), whose education level was college and above (β = 1.212, 95%CI: 0.497, 1.927, p < 0.01) or senior high school (β = 0.568, 95%CI: 0.339, 0.797, p < 0.001), who were employed as government/public institute employees (β = 3.160, 95%CI: 1.201, 5.138, p < 0.01), corporate/business/service employees (β = 2.699, 95%CI: 0.937, 4.461, p < 0.01), those who were not interested in quitting smoking (β = 1.135, 95%CI: 0.318, 1.952, p < 0.01), and those who were exposed to SHS in public places (β = 0.524, 95%CI: 0.233, 0.815, p < 0.001) had higher odds of cigarette affordability. The marginal effects for each affordability category showed a similar, consistent, and complementary pattern to the coefficients.

Disscusion

This is the first study to assess the impact of the local SFL on tobacco consumption and affordability in China. The Wuhan SFL was promulgated and implemented in January 2020, establishing comprehensive smoke-free environments in indoor public and workplace settings. This study aimed to explore the economic effects of the local SFL by comparing changes in cigarette prices and affordability among current smokers before and after the implementation of the SFL. Our analysis reveals three key economic outcomes following the legislation: stable cigarette prices, a wide price range, and significantly increased affordability of cigarettes.

The average cost of a pack of cigarettes among current smokers did not change significantly over time in Wuhan, China, inconsistent with previous studies [16, 30, 40]. There may be several reasons contributing to this phenomenon. On the one hand, despite growth in per capita GDP, the lack of increased tobacco taxes under smoke-free legislation at both the local and national levels during the study period removed a core driver for price increases. On the other hand, the tobacco industry’s marketing strategies, which may have prioritized product portfolio optimization over broad price hikes, could have further contributed to market-level price stability.

The range of cigarette prices in Wuhan remained wide throughout the study period, consistent with previous studies [15]. It reflects a national complex tax structure characterized by tiered excise taxes, where cheaper cigarettes are taxed at lower rates than higher-priced brand cigarettes [21]. This wide range exacerbated price disparities and widened the range of cigarette price choices, which poses a potential risk to tobacco control efforts by enabling substitution behaviors. Although our study did not measure brand switching, the stable choice of low-price tiers of cigarettes suggests that current smokers facing economic pressures may respond by switching to cheaper brands rather than quitting. To address this, a complex, tiered tax system should be replaced by a uniform, specific excise tax, and the price range should be narrowed by raising the floor price of the cheapest products, reducing the availability of more affordable alternatives, and minimizing the opportunity for substitution.

This stability in overall consumption patterns is further reflected in consumer behavior. The marginal effects indicate that between 2019 and 2024, the probability of current smokers choosing cigarettes across different price tiers (upper, middle, and low) did not change significantly. This stability in price choice may reflect ingrained consumption habits, which are reinforced by high nicotine addiction, low price elasticity, and targeted marketing from the tobacco industry. Long-term smokers often develop deep brand loyalty, creating fixed comfort zones that are resistant to change.

However, this stability in consumer behaviour stands in stark contrast to broader economic trends. Our findings highlight a divergence phenomenon: the stability in cigarette prices, the continuation of a wide price range, and a significant increase in cigarette affordability. The average cost of 100 packs of cigarettes as a percentage of Gross Domestic Product (GDP) per capita declined significantly from 2.8% (2.5, 3.1) in 2019 to 2.2% (1.8, 2.5) in 2024, a level lower than that in Bangladesh, Romania, and Thailand [15]. It indicates that, compared to growth in household disposable income, tobacco products have become more affordable for current smokers over time, revealing that cigarette prices, particularly tobacco tax, failed to keep pace with income growth [32]. From a public health perspective, it is concerning, especially in a country with a large population of smokers like China. As a country’s economy develops, increases in household disposable income can result in further increases in cigarette consumption. To defend this, cigarette prices need to grow faster than family income. Raising tobacco taxes will increase the unit price, which will prevent initiation, promote quitting [41], and lead to reduced consumption among current smokers [42, 43]. This approach is strongly supported by global evidence and frameworks. Article 6 of the WHO Framework Convention on Tobacco Control (FCTC) states that price and tax measures are an effective and necessary measure to decrease tobacco consumption [14]. A 2024 study estimated that raising excise taxes enough to generate a one-time price increase of 50% would lead to 102 million people globally quitting smoking, and at the same time, tobacco taxes can provide significant fiscal revenue for governments [44]. Furthermore, tobacco taxation is considered a highly cost–effective ‘best-practice’ intervention, yielding returns and economic benefits that far exceed the costs incurred [2, 45]. Therefore, although tobacco control legislation has successfully reduced exposure to SHS [12], its failure to align with fiscal policy prevented it from effectively curbing the economic drivers behind tobacco consumption.

The ordered logit model analysis further revealed that the choice of cigarette price tier among current smokers is related to socioeconomic characteristics and smoking behavior, consistent with previous studies [16, 17, 26, 30]. In terms of marginal effects, socioeconomic factors ranked as the most significant determinants: government and public institution employees had a significantly increased probability of choosing upper price-tier cigarettes, with an increase of 31.9% points (β = 2.529), probably due to their higher stable income and occupational social demands; individuals with college degrees or above educational level is 19.3% points more likely to choose upper price-tier cigarettes (β = 1.169), demonstrating that education influences consumption decisions by both enhancing purchasing power and shaping consumption preferences. Males were more likely to choose upper price-tier cigarettes, a pattern that a combination of factors, such as higher incomes, gender-specific cigarette marketing, and social norms, may influence. Current smokers with no intention to quit showed a significantly increased odds of choosing upper price-tier cigarettes by 15.1% points (β = 1.238), suggesting that high nicotine dependence or deep smoking habits may reduce price sensitivity; the 15–24 age group was 28.9% points more likely to choose upper price-tier cigarettes than those aged 65 years and older (β = 1.670), potentially attributable to younger demographics being more susceptible to brand marketing, coupled with the psychological tendency to construct identity through consumption.

These findings highlight the necessity for more effective tobacco control policy interventions. Our analysis reveals a difference in the choice of cigarette price tier across population subgroups: employed individuals with a relatively stable occupation and higher education demonstrate a clear preference for upper-tier price brands, while those with no intention of quitting exhibit significantly higher odds of choosing expensive cigarettes. These patterns underscore the need to implement targeted measures that address the specific drivers of tobacco consumption in each group. Although China introduced cigarette tax adjustments in 2015, evidence suggests that these changes did not transmit to retail prices [15]. Moreover, the local SFL adopted in 2020 in Wuhan, China, did not involve tobacco tax changes [11]. Therefore, to enhance the effectiveness of tobacco control legislation, future strategies should integrate robust fiscal measures alongside smoke-free policies. Specifically, priority should be given to transforming the current complex tiered tax system into a unified high consumption tax. This measure will supplement existing legislative measures by raising the minimum price of tobacco products, thereby reducing affordability and minimising opportunities for substitution [26, 46]. Additionally, for price-sensitive subgroups such as younger and less-educated smokers, substantial tax increases coupled with a minimum unit price policy are crucial to break their economic comfort zones [31]. Simultaneously, integrating comprehensive smoking cessation services into the national healthcare system and providing targeted support tailored to different socioeconomic groups will create positive pathways for quitting. Particular attention should be paid to individuals with no intention to quit smoking, who may require more intensive behavioural interventions.

This study has limitations. First, the prices reported in the study reflect the prices of brands purchased by smokers included in the survey and thus failing to capture the full price range of all cigarettes available. Second, since these prices are derived from self-reported survey data, they are subject to limitations related to the survey, such as reporting bias. Third, we are limited to single cross-sections of data. We are therefore unable to claim any causal relationships. Fourth, we acknowledge that no direct mechanism or body of evidence explains how smoke-free legislation, in the absence of tax increases, would be expected to influence cigarette prices or affordability. This study does not test a causal pathway from SFL to economic outcomes; rather, it uses the SFL implementation as a policy event to assess the economic context in which smokers make consumption decisions. Finally, the Coronavirus Disease 2019 (COVID-19) pandemic, which occurred during the study period, may have influenced smoking behaviors and thus confounded our results. Pandemic-related factors such as increased health awareness, stress, social restrictions, and changes in disposable income could have affected cigarette consumption, quitting intentions, and purchasing patterns in complex ways. While our survey years (2019, 2022, 2024) capture pre- and post-legislation trends, the absence of systematic control for the pandemic’s effects means we cannot fully disentangle its impact from that of the legislation itself, potentially introducing confounding bias.

Conclusion

This repeated cross-sectional study in Wuhan, China, assessed the economic dimensions of tobacco consumption following the implementation of smoke-free legislation. We found that after the SFL took effect in 2020, cigarette prices remained stable with a wide price range, while cigarette affordability increased significantly among current smokers. The absence of complementary tax policies allowed tobacco products to become more affordable relative to rising incomes, potentially undermining cessation efforts. These findings underscore the necessity for China’s national tobacco control strategy to integrate robust fiscal measures—particularly uniform excise tax increases and minimum floor pricing policies—to complement existing smoke-free legislation and effectively curb tobacco consumption.

Abbreviations

CDC

Centre for Disease Control and Prevention

CI

Confidence Interval

FCTC

Framework Convention on Tobacco Control

GATS

Global Adult Tobacco Survey

GDP

Gross Domestic Product

PPP

Purchasing Power Parity

SFL

Smoke-free Legislation

SHS

Secondhand Smoke

WATS

Wuhan Adult Tobacco Survey

WHO

World Health Organization

Authors' contributions

XM, GC, and TZ: conceptualization, formal analysis, investigation, data curation, and writing, reviewing and editing. XM, GC, and TZ: formal analysis, investigation, and writing, reviewing and editing. XM, GC, XH, JZ and TZ: conceptualization, methodology, writing, reviewing and editing, supervision, and project administration. LZ: conceptualization, methodology, investigation, writing, reviewing and editing, supervision, and project administration. All authors contributed to the article and approved the submitted version.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Data availability

The data analyzed in this study is subject to the following licenses/restrictions: The datasets generated during and/or analyzed during the current study are not publicly available due to restrictions applied to the availability of these data but are available from the corresponding author on reasonable request. Requests to access these datasets should be directed to zhangling@whcdc.org.

Declarations

Ethics approval and consent to participate

This study, in accordance with the Declaration of Helsinki, was approved by the Ethics Committee of the Wuhan Centre for Disease Control and Prevention (CDC), China (WHCDCIRB- K- 2019012, WHCDCIRB- K- 2022019). Informed consent was obtained from all participants before the survey. For respondents aged 15–17 years, informed consent was obtained from their parent or guardian, in addition to the assent of the minor participant.

Consent for publication

Not applicable. The manuscript does not contain any individual person’s data in any form.

Competing interests

The authors declare no competing interests.