Temperature, Air Pollution and Fatal Unintentional Injuries in Jiangsu Province China, 2015-2017 CURRENT STATUS: UNDER REVIEW

Background The correlation of unintentional injury mortality to rising temperatures found in several studies could result from changes in behavior that increases exposure to hazards or risk when exposed. An increase of .82 degrees (C) daily maximum temperature in Jiangsu Province, China from 2015 through 2017 provides the opportunity to study further this phenomenon including a potential effect of air pollution. Air pollutants may cause symptoms and distractions that increase risk and avoidance behavior that reduces risk. This study examines data that allows estimates of the effects of temperature, precipitation and five air pollutants on injury mortality risk, each corrected statistically for the potential effects of the others. Methods Daily data on unintentional injury deaths and exposures to temperature, precipitation and pollutants from 9 cities in Jiangsu Province, China during 2015-2017 were analyzed using logistic regression. Application of the regression equations to 2016 and 2017 data substituting 2015 values for each of the predictor variables respectively was used to estimate changes in deaths related to changes in the risk factors. , certain

Results Unintentional injury death risk was unrelated to temperature when temperatures were below 25 degrees (C) but increased substantially in relation to higher temperatures. The correlations mainly occurred for non-transport injuries. Fatal injury risk was higher in relation to higher concentrations of PM 10 , NO 2 , and SO 2 in certain temperature ranges but lower in relation to higher concentrations of CO in the lowest and highest temperature ranges. Lower risk was associated with O 3 at low temperatures but higher risk at high temperatures. Risk was lower on weekends but higher on holidays when temperatures were moderate.
Conclusions The increase in fatal unintentional injuries from 2015 through 2017 was substantially related to the rise in average maximum daily temperature. Negative associations of certain pollutants, particularly O 3, with injury risk suggests the possibility that changes in behavior to avoid those pollutants also results in lower injury risk. Most pollutants, however, are related to increased risk. Background 3 From January, 2015 through 2017 the average maximum temperature measured daily among nine major cities in Jiangsu Province, China rose near 0.82 degrees (C) from 20.24 degrees in 2015 to 20.59 degrees in 2016 and 21.06 degrees in 2017. This event provided the opportunity to explore further the correlation of temperature, precipitation and air pollution to unintentional injury mortality found in previous research and the extent to which temperature increases and pollution control may affect fatal injury rates.
Temperature, precipitation and air pollution may affect risk of injury in two ways: increased or decreased exposure to potentially injurious energy or increased risk if exposed. A review of the literature on the correlation of various types of injuries to ambient temperature noted that most studies focused on the effects of extreme temperatures but those that studied temperature in normal ranges also found increased injury incidence and mortality rates associated with warmer temperatures (Kampe, et al., 2016). Comparisons among U.S. states indicate increased injury mortality risk associated with warmth above normal (Parks, et al., 2020). Research on police-reported road collisions in Spain found a 2.9 percent increase during heat waves (Basagana, 2015). While peoples' attention to tasks at hand may be distracted by extreme temperatures, the more likely explanation for much of the correlation is changes in human activity based on temperature and precipitation that expose them to greater or less environmental hazards. Although certain recreational activities, such as skiing, are more frequent on colder days, many others, such as swimming and boating are warm and dry weather activities. Outdoor construction and other projects are sometimes suspended because of inclement weather. Higher temperatures increase risk for cocaine users (Marzuk, et al., 1998) but not as high as once thought (Bohnert, et al., 2010). The correlation of temperature to drug overdoses could also partly be the result of addicts being more often under the watch of families or other persons who care for them on cold or wet days that inhibit freedom of movement.
A study of road deaths among urban cities and counties in the U.S. found that reversal of the decreasing road death trend during 2015 was mainly associated with increased temperatures in that year (Robertson, 2018a). An alternative hypothesis that economic recovery from the Great Recession of 2008-2009 explained the reversal was not confirmed by comparing data among U.S. states during 2000-2016 (Robertson, 2019b). In both studies, increases in vehicle travel were found in relation to increasing temperature. Use of roads by pedestrians and bicyclists likely increased as well.
Researchers noting the increase in injury deaths in Chinese cities during the latter half of the 20th Century repeated without evidence the conventional wisdom that they are mainly the result of activities associated with rapid economic growth (Zhou, et. al., 2012). From 2010 to 2015, however, injury death rates in Chinese cities were substantially lower than in the 1980s and 1990s while economic growth remained high (Ozanne-Smith and Li, 2018).
Environmental conditions other than temperature and precipitation may increase or decrease injury risk to those exposed and may also alter the probability of exposure (Sager, 2019). Air pollution is irritating to the eyes and may impair vision as well as lead to coughing spasms and breathing problems especially among the asthmatic, all of which would distract from alertness. Sleepiness has been associated with air pollutants (Heyes, A. and Zhu, 2019). Performance on verbal and mathematics tests is reduced among those exposed to higher ranges of air pollutants (Zhang, et al., 2018). Attempts to reduce exposure to pollutants may lead to more sedentary activities that lower risk of injury (Bresnahan, et al., 1997). Studies of avoidance behavior in relation to pollution find that some people stay indoors when the air has higher concentrations of pollutants, particularly when there are broadcast, print and internet warnings of hazard to health. Smog alerts in California were related to less attendance at studied outdoor venues (Neidell, 2008). School absences in Texas were found related to higher concentration of air pollutants (Curry, 2009) which could be due to both increased illness and avoidance behavior.
The purpose of this paper is to report analysis of the extent to which changes in temperature, precipitation and air pollution are related to the increase in injury deaths from 2015 through 2017 in Jiangsu Province, China. While the analysis of injury death rates in correlation with these factors does not specify the extent of avoidance behavior versus risk when exposed, the analysis does quantify the likely net effect of these two factors and, in the instances of negative correlations, suggests that avoidance behavior reduces the risk.

Methods
Daily injury deaths during 2015-2017 in each of 11 cities in Jiangsu Province, China were provided by the Jiangsu Provincial Center for Disease Prevention and Control. Two cities were included that did not have data for all three years. These were excluded from the analysis. The deaths were specified as occurring in transport or other circumstances. This classification was based on the International Statistical Classification of Diseases and Related Health Problems 10th Revision (ICD-10) codes V00-V99 for transport cases and W00-W99 for other unintentional cases (ICD10Data.com, 2019). As of 2013, 96.5 percent of transport deaths in China were from road injuries (Zhou, et al., 2016).   Table 1 presents the injury deaths per million inhabitants per year by external cause (transport vs. other) and in total for each city. The data show a remarkable variation among the 9 cities in injury mortality per population. In many instances, Nanjing's rate is less than half that of 8 of the 9 other cities. The three year upward trend in total injury death rates in several cities is mainly due to unintentional injuries other than those experienced in transport. Totals are not always the sum of transport and other deaths because of rounding. Figure 1 shows the transport related deaths per billion person days of exposure to specific maximum temperatures. Each data point is the sum of the deaths on the days that the maximum temperature reached the indicated temperature divided by the sum of the approximate number of people residing in a given city each day the temperature reached the indicated temperature adjusted to billions of person days. A maximum daily temperature at freezing or below in Jiangsu Province is rare. Person days when the maximum temperature was zero (C) or less were 0.23 percent of the total so deaths on those days and the person years are included in the zero category. There does not appear to an association of transport death risk with maximum temperatures other than a dip at the highest temperatures. In contrast, Figure 2 indicates that unintentional injuries other than in transport declined slightly up to about 25 degrees maximum daily temperature, accelerated upward at higher temperatures and did so extraordinarily at temperatures above 34 degrees.

Results
The logistic regression equations for transport deaths and other unintentional deaths separately did not accurately predict actual total deaths for each in a given year so they could not be used for the purpose of estimation of the deaths attributable to changes in the variables. The prediction of the equation for total unintentional injury deaths in 2016 was higher by 3.5 percent (25428 expected vs. 24,567 actual) in 2016 but very similar in 2017 (25737 expected vs. 25666 actual). The coefficients are presented in Table 2. Corrected for the estimated effects of other factors, when temperatures are less than 25 degrees, risk of fatal injury is unrelated to temperature but increases in relation to degrees of moderate and high temperatures. Precipitation has no apparent effect. The effect of pollutants varies depending on temperature range. Higher risk is associated with higher PM 10 at cool but not moderate or hot temperatures. SO 2 is related to increased risk at cool and moderate but not hot temperatures. NO 2 is 9 related to higher risk only at cool temperatures. Risk decreases with higher CO at cool and hot temperatures and O 3 when temperatures are cool but that correlation reverses at hot temperatures.
Risk is lower on non-workday weekends although significantly so only at moderate temperatures.
Holidays are not associated with risk at cool temperatures but risk is higher on holidays when temperatures are moderate and lower when temperatures are hot.   Province. Also, a substantial proportion of the correlation of temperature and road death risk in the U.S. is due to the apparent lack of use of bicycles and motorcycles at subfreezing temperatures (Robertson, 2019b). Such temperatures are extremely rare during daytime in Jiangsu Province.
Unfortunately, we have no data on kilometers travelled per vehicle in China or pedestrian volume and bicycle use in the U.S. and Chinese cities. If there were major road injury prevention efforts during 2015-2017 in Jiangsu Province, they could have offset the effect of temperature. According to one report, China has made little use of "traffic calming" road designs that have reduced fatality rates in other countries but does have engineering standards for vehicle crashworthiness, child safety seats and school bus seats (Fayard, 2017). A seat belt use law was imposed throughout China in May, 2004.
A study if belt use observed in traffic in Nanjing soon after the law indicated about 67 percent use by drivers and 19 percent by front seat passengers (Routley, et al., 2007). During [2005][2006][2007], belt use by drivers in Nanjing had declined to 39 percent and that of front seat passengers to 3 percent leaving substantial room for improvement (Routley, et al., 2008). A study of road deaths in Jiangsu Province during 2012 found that more than half the deaths occurred to pedestrians and that the risk of pedestrian deaths per population increased dramatically with age (Ding, Y. et al., 2017) but the age distribution of the population would not have changed enough in three years to affect the results reported here. A search of Chinese laws and regulations aimed at childhood injury prevention relative to laws and regulations recommended by the United Nations Children's Fund, the World Health Organization or the European Child Safety Alliance found that 10 of 27 were not found in Chinese statutes and regulations (Li, et al., 2015).
The results regarding the possible effects of pollution are complicated but overall suggest that whatever avoidance behavior may reduce injury risk, the net effect of more pollution is increased injury risk. The major exception is increased CO and O 3 which appears to affect avoidance behavior at certain temperatures. More research is needed on observed behavior in relation to concentration of pollutants at different temperatures.

Conclusions
Warming temperatures are associated with increased risk of unintentional injury mortality. Burning fossil fuels that contribute to warming continues to grow, outpacing the adoption of sustainable energy sources (United Nations Environment Programme, 2019). Although the effects of warming on injury mortality vary from country to country depending on the temperature ranges experienced, cultural factors and behavioral responses to temperatures and air pollution, without further efforts at prevention based on data regarding the types of injuries affected, climate warming and possibly some forms of pollution control will likely contribute to an upward trend in injury mortality.

Declarations
Ethics approval and consent to participate: The data are from statistical resources and no contact with individuals occurred.
Consent for publication: The authors agree on submission for publication.
Availability of data and materials: Jiangsu CDC does not permit posting of daily mortality data for public access.
Competing interests: The authors have no financial or other interests that would be altered by publication of this study.
Funding source: The study received no funding support.