The middle-aged and elderly urban residents’ individual perceptions of PM2.5 pollution and its influencing factors in Wuxi, China


 Background PM 2.5 pollution has become a major public health concern in urban China. Understanding the residents’ individual perceptions toward haze pollution is critical for policymaking and risk communication. However, the perceptions of middle-aged and elderly residents, who particularly vulnerable to haze pollution, are poorly understood. Methods A cross-sectional study of 400 randomly sampled individuals (aged 40 to 90 years) was conducted in Wuxi, a typical PM 2.5 -polluted city. Each participant’s demographic and health information, individual perception and pulmonary function outcomes were collected to explore the relationship between personal characteristics and pulmonary function parameters and perception factors. Results We found that the mean values for controllability (4.99 ± 2.78) and dread of self-risk (6.90 ± 2.45) were the lowest and the highest values, respectively, in our study. Education and average family income were positively related with all individual perception factors, while age was negatively associated. A history of respiratory disease was positively associated with all individual perception factors except controllability . Significant positive associations were observed between PEF (coefficients ranged from 0.18 to 0.22) and FEF75% (coefficients ranged from 0.18 to 0.29) with a variety of individual perception factors. Conclusions There was a lack of concern and knowledge, weak self-protection consciousness and a strong dread of PM 2.5 pollution among the middle-aged and elderly residents in Wuxi. Their individual perceptions were associated with age, education levels, average family income, history of respiratory disease and pulmonary function outcomes. Our findings may help policymakers develop effective policies and communication strategies to mitigate the hazards of haze among older residents.


Introduction
With rapid economic growth, energy consumption and air pollutant emissions have substantially increased in China [1]. Fine particulate matter with a diameter of <2.5 μm (PM 2.5 ) pollution has become one of the most severe environmental problems, especially in highly industrialized urban areas of China [2]. Numerous studies have established that PM 2.5 pollution threatens human health in many ways; it increases the morbidity and mortality of respiratory and cardiovascular diseases, impairs humans' pulmonary and cognitive function, and has adverse effects on mental health and well-being [3][4][5]. Although PM 2.5 is hazardous to the whole population, elderly people are especially vulnerable [6].
In addition to health impacts, large-scale PM 2.5 pollution in urban China has also led to some social problems. In the winter of 2018, many areas of China had haze pollution alerts and closed important expressway stations due to the serious smog [7]. In areas with high concentrations of PM 2.5 , PM 2.5 pollution-related diseases cause additional medical expenses, work time loss and GDP loss [8,9].
Because frequent haze pollution has led to a rise in Chinese public concern and has caused the potential risk of social unrest [10]. The Chinese government has thus far established a series of environmental regulations to control industrial and vehicular emissions, encourage clean energy and set up air pollution early warning systems and response plans for haze episodes [11][12][13]. "Risk perception" is how people judge the magnitude and degree of risk with their intuition [14]. It is a very important and effective indicator of public concern about air pollution. Risk perception can guide individuals' self-protective behaviors and help them respond to government work [15]. On the other hand, understanding the population's risk perception can help the government engage in efficient risk communication to bridge the gap of risk perception between the experts and the public and to create effective policies to protect public health and mitigate potential adverse socioeconomic impacts.
Given that a proper understanding of individual perceptions of air pollution is critical for policymaking and risk communication, many studies have been conducted to explore the public's perception of air pollution in recent years. Previous studies have reported that a lower level of education and income might be associated with more dissatisfaction with respect to air pollution [16,17]. Qian and Kim et al. found that women and younger people are more sensitive to air pollution risks [18,19]. However, other studies have indicated that middle-aged and elderly people perceived greater risk and health threats associated with air pollution [20][21][22], and people with higher education and income levels tend to be more concerned about air pollution [23][24][25]. Moreover, individual perception could be influenced by health status, thermal sensations, and personal experiences with air pollution [15,26,27].
The majority of current studies have focused on the adult population; many of these studies included respondents who were younger and more educated than the average individual in the target population, potentially because older, uneducated people have fewer chances to participate in such studies and have greater difficulty in understanding questions [15,27,28]. Middle-aged and elderly residents are particularly vulnerable to air pollution [29,30]; therefore, it is crucial to understand their risk perceptions and develop targeted policy strategies to protect them. However, it might be inappropriate to generalize the conclusions of previous studies directly to middle-aged and elderly individuals because they are older, and many of them are less educated than younger individuals.
Moreover, compared to young people, middle-aged and elderly individuals are more likely to have cardiovascular a n d /o r respiratory diseases and worse pulmonary function [31,32]. Those characteristics may largely affect their risk perceptions.
In light of these findings, our study included 400 middle-aged and elderly residents (age between 40 to 90 years old) from Wuxi, an important economic and industry center in the Yangtze River Delta region. The aims of this study are 1) to explore middle-aged and elderly urban residents' risk perception of air pollution, 2) to identify the determining factors influencing the risk perception of haze among middle-aged and elderly residents, and 3) to determine the relationship between health status and pulmonary function parameters and risk perception.

Sample selection
This study was conducted in Wuxi, an important industry and economy center in the Yangtze River Delta region. According to the reported data, the mean concentration of PM 2.5 in Wuxi was 52.73 μg/m 3 between 2015 and 2017, which was 5.27 times higher compare to the WHO air quality guideline's stipulation (PM 2.5 not exceed 10 μg/m 3 annual mean) [33].
We used a two-step sampling method in this study. First, one district and one county were randomly selected from 5 districts and 2 counties, respectively, in the city. Then, all communities in the district/county with over 10,000 residents served as the primary sample units based on probabilities proportional to population size (PPS). In the second step, one such community was randomly selected from both the chosen district and county, and 200 middle-aged and elderly residents (age between 40 to 90 years old, living at their present residence for more than three year before this study) from each community were in turn selected using a simple random sampling method (SRS). A table containing the IDs of all the residents who met the age criteria was built using SRS, and then a series of random numbers were generated to select one resident; the other 199 residents were selected by using equal interval numbers based on the starting resident ID.
All participants were interviewed face-to-face by a member of our research team, and the pulmonary function test was performed by trained specialists. A total of 400 residents from the two communities completed the questionnaire, 398 of which successfully underwent pulmonary function testing.

Questionnaire
The questionnaire was composed of two parts. The first part of the questionnaire surveyed demographic information and health status. For health status, we collected information about history o f cardiovascular disease and history of respiratory disease. The participants with diagnosed hypertension, arrhythmia, coronary heart disease, myocardial infarction or any other cardiovascular disease that had been reported to be related to PM 2.5 pollution were defined as "have a history of cardiovascular disease". Those with diagnosed asthma, chronic obstructive pulmonary disease, lung cancer, chronic respiratory inflammation or acute respiratory inflammation (occurring within one year) were defined as "have a history of respiratory disease".
The second part consisted of eight questions that addressed individual perceptions of PM 2.5 pollution and its related health effects. It was designed based on the psychometric paradigm method [14], which uses scaled questions to measure individual preferences with respect to different risks [14,15]. Each question measured perception levels by asking participants to provide a score ranging from 1 to 10 for each question (see Supplemental Table 1).
Generalized linear models (GLMs) were used to explore the relationship between pulmonary function parameters and individual perception factors of PM 2.5 . Age in years, education, average family income and history of respiratory disease were included in all initial models, and stepwise regression was then used to add or remove variables. All statistical analyses were conducted using R software (version 3.3.1, R Foundation for Statistical Computing, http://cran.r-project.org/), and the GLM was fitted using the splines package. Pvalues <0.05 were considered significant.

Results
A total of 400 volunteer residents surveyed effectively, and 398 of these successfully underwent pulmonary function tests. The demographic characteristics and health status of the participants are summarized in Table 1

Pulmonary function outcomes
Descriptive statistics of the participants' pulmonary function outcomes are provided in Table 2.
According to previous studies, pulmonary function (spirometric measures FEV1, FVC., etc.) declines with increasing age, and the loss in function will especially accelerate after 65 years of age [34,35].
Therefore, we calculated the pulmonary function outcomes of the participants aged 41-65 (middleaged group) and of those aged 66-90 (elderly group) separately. The analysis showed that the pulmonary function outcomes of the middle-aged group were all significantly higher than those of the elderly group.

Individual perceptions
As seen in Table 3, the mean values of individual perception factors in the total sample of participants ranged from 4.99 to 6.90. Dread of self-risk scored highest, and controllability scored lowest in our study. In addition, we compared the differences in individual perception between the middle-aged group and the elderly group through an independent-sample t-test analysis. We found that the values of the individual perception factors of the middle-aged group were all significantly higher than those of the elderly group (indicated in Figure 1).

Associations between pulmonary function outcomes and individual perception of PM 2.5
We analyzed the associations between pulmonary function outcomes and individual perceptions, with separate analyses conducted for the middle-aged group (aged 41-65, N=195) and the elderly group (aged 66-90, N=203). Table 5 Similar associations were observed among the elderly residents ( Table 6). We observed significant positive associations of FEF 75% with knowledge (coefficient = 0.25, P < 0.05), familiarity (coefficient = 0.28, P < 0.05), dread of self-risk (coefficient = 0.29, P < 0.05) and dread of risk to others (coefficient = 0.23, P < 0.05). Furthermore, we also found that FEV 1 and FVC were positively associated with familiarity.

Discussion
In recent years, haze pollution has raised great social concern in China. Understanding the public's risk perception can help government managers make targeted polices and provide efficient risk communication to relieve social anxiety and protect vulnerable groups.
Our study focused on middle-aged and elderly Chinese urban residents, who are generally vulnerable to air pollution. We found that the mean values for self-reported controllability (4.99 ± 2.78) and concern (5.65 ± 2.82) were the lowest in our study, and the mean values for dread (dread of self-risk: 6.90 ± 2.45; dread of risk to others: 6.89 ± 2.43) were the highest in our study (  [36,37]. Meanwhile, 78.8% of the respondents in Ningbo felt dread toward the possible aggravation of the haze, and 83% of respondents in Nanchang worry about the potential adverse impact on their respiratory system caused by a high level of air pollution [19,37]. Despite the dread of haze pollution, only 19 (4.8%) participants in our study had ever used air purifiers to improve indoor air quality ( Table 1); this ratio is much lower than that reported among younger residents in Nanjing (15.2% air purifier use) [15]. This may be due to the low levels of self-perceived controllability in our study; in other words, the residents do not believe that they can effectively reduce the risks associated with haze by engaging in self-protective behaviors.
Currently, the internet has become one of the most important ways to deliver health information, and it was ranked as the most favorable way to obtain knowledge and protective measures of haze in a previous study conducted among the comparatively younger population [19].
However, it could be difficult for older adults to regularly obtain haze-related information from the internet because their lifestyle habits differ from those of young people, and some older adults even lack the skills of surfing the internet. Therefore, we suggest that government managers combine traditional methods (such as newspaper, broadcast and community/hospital lectures) with new media to make haze-related information more accessible t o older residents. Moreover, the importance of self-protective behaviors should be described and emphasized in environmental education to help older adults mitigate the health risks of haze pollution.
Previous studies have indicated that individual perceptions of air pollution could be influenced by many factors, such as age [38,39], gender [40], education level [41], family income [42], individual experiences [15], and health symptoms [27]. Supporting the previous findings, we found that education and average family income were positively associated and age was negatively associated with all individual perception factors. History of respiratory disease was positively associated with all individual perception factors except controllability in our study ( Table 4). However, although cardiovascular injury is one of the most important health hazards of air pollution, no significant correlation was found between history of cardiovascular disease and any individual perception factor of PM 2.5 . This result indicates that the residents with cardiovascular disease may not identify haze pollution as a health threat to their disease and did not pay more attention to it than the healthy group did. A similar situation was found by Liu et al., in which only 21.2% of the respondents considered heart problems a health consequence of air pollution [36]. Nevertheless, as indicated by the self-reported perceived knowledge level (mean score: 6.55 ± 2.47), Wuxi's middleaged and elderly residents believe that they have adequate haze-related knowledge. Taken together, these results may reveal a potential obstacle in current air pollution-related health education: there is a gap between residents' self-perceived knowledge level and their actual level, which may cause insufficient self-protective behavior among vulnerable groups and incorrect risk perceptions. Therefore, we suggest that government managers develop targeted health education strategies and risk communication messages for vulnerable groups, especially for residents with cardiovascular diseases.
Considering t h a t both pulmonary function outcomes and individual perceptions could be influenced by age, we found that they were significantly different between the middle-aged and elderly groups (shown in Table 2 and Figure 1). We analyzed the associations between pulmonary function outcomes and individual perceptions separately among the middle-aged group (aged 41-65, N=195) and the elderly group (aged 66-90, N=203). It is very interesting that we observed that better pulmonary function outcomes were related to higher self-perceived levels of severity of health effects, knowledge, familiarity, dread of self-risk and dread of risk to others in both middle-aged and elderly groups. These results indicated that residents with worse pulmonary function did not pay enough attention to the hazards of PM 2.5 pollution. As reported in a previous study, lower knowledge and dread levels may result in less self-protection behaviors during haze pollution [15], and less selfprotection may further worsen pulmonary function. In our study, we found that FEF 75% was associated w i t h familiarity, dread of self-risk and dread of risk to others in both the middle-aged and elderly groups. This result suggested that FEF 75% could be considered a sensitive indicator for hospital-based health educators to use to identify those more likely to need to be educated to improve their knowledge and those who need to relieve their anxiety toward the potential health risk caused by PM 2.5 . The above findings suggest that policymakers should consider residents' health status when making health education and risk communication strategies for the targeted groups.
To the best of our knowledge, this is the first study to specifically explore risk perception among Chinese middle-aged and elderly residents, who were generally defined as the vulnerable group to PM 2.5 pollution. We first identified the relationships between the residents' disease history and pulmonary function outcomes with their individual risk perceptions. We discussed several policy implications in the sections of discussion in this article, and our results may help government managers make target policy strategies.
Our study also has some limitations. First, as a cross-sectional study, the pulmonary function tests were performed on the same day as the questionnaires, and we observed that the residents' pulmonary function outcomes was associated with their individual perceptions. Although a previous study has reported that higher levels of risk perceptions may help residents take additional protective actions [15]. However, in this study, we are not sure about whether higher levels of risk perceptions helped with older residents' pulmonary function outcomes. Second, a larger scale of older residents from different cities should be involved in future studies.
Despite these limitations, our study provides a reference for government managers about the associations between disease history and pulmonary function with risk perceptions among Chinese middle-aged and elderly urban residents for the first time and lays the foundation for subsequent researchers.

Conclusions
In summary, our study suggests that middle-aged and elderly residents may lack concern and knowledge about air pollution, have no confidence in mitigating PM 2.5 -related health risks by self-

Declarations
Ethics approval and consent to participate The study protocol was approved by the Ethics Commission of Jiangsu Provincial Center for Disease Control and Prevention; all study participants provided written informed consent.

Consent for publication
Not applicable.

Availability of data and materials
The data that support the findings of this study are available from Jiangsu Provincial Commission of Health and Family Planning; however, restrictions apply regarding the availability of these data, which were used under a license for the current study, and the data are not publicly available. However, data are available from the authors upon reasonable request and with permission of t h e Jiangsu Authors' contributions QC contributed to the study design, data procurement, interpretation of the results and drafting of the article. JZ was involved in the collection of the data. ZD, HS and YX took part in the study design, supervision of the research, data procurement and results interpretation. All authors read and gave their approval for publication. suptables.docx