Global, Regional, and National Burden of Accidental Carbon Monoxide Poisoning, 1990–2019: A Systematic Analysis for the Global Burden of Disease Study 2019

Health system planning requires careful assessment of accidental carbon monoxide poisoning (ACOP) epidemiology, but data of this disease are scarce or non-existent in many countries. This article investigates the global burden of ACOP based on the Global Burden of Disease Study 2019 (GBD 2019) and the World Bank database. Numbers and age-standardized rates (ASR) of ACOP incidence, prevalence, deaths, Disability-Adjusted Life Years (DALYs), Years Lived with Disability (YLDs), and Years of Life Lost (YLLs) were analysed at global, regional, and national level. The estimated annual percentage change (EAPC) of age-standardized rates (ASR) was calculated by generalizing the linear model with a Gaussian distribution. Age, sex, and economics parameters are included to access their internal relevance. (95%UI 32957 to lower article also nds that the relevant parameters of ACOP are closely related to the This correlation enables us to evaluate the level and status of public health services in various countries. Lao Lebanon, Malaysia, Mauritius, Moldova, Oman, Poland, Saudi Arabia, Sri Lanka, St. Timor-Leste, United Kingdom, Yemen, Rep., Bermuda, Niue, San Marino, Slovakia, Venezuela, RB). The data of ASDR are serious too. 100 countries have the growing trend of ASDR from 2010 to 2019, and 34 countries are signicantly related to the year (Barbados,

All the data analysis was performed by R 4.0.2. The missing data of CPI were modelled by the 'mice' package (method: Predictive mean matching). CPI-adjusted was calculated by cumulative multiplication of completed CPI to represent the relative level of price.
All null values in GDP per capita (PPP) and GNI per capita (PPP) were removed before calculation to ensure the accuracy and authenticity of the data.
Andean latin america Age and sex patterns Compared to 1990 by age and SDI regions, in 2019, most of ACOP parameters were lower (Fig. 4). In 2019, whether it's by age or SDI regions, ASPR, ASIR, ASDR, YLDs, and YLLs were higher in males ( Fig. 2 & Fig. 5). ASPR, ASDR, and YLDs increased with age.
However, ASIR showed a trend of increasing rst and then decreasing, and its peak was about 15-19 years old, which was the same trend in males and females. For females, YLLs and DALYs also indicated a trend of increasing with age, while the males group peaked after 35-39 years old and maintained a high level, which was very different.

Risk factors
There are two risk factors listed in the database: alcohol use and occupational injury. In the past 30 years, occupational injury is the main risk of ACOP. By sexes, the risk of occupation the risk of males is higher than that of females (Fig. 6). The 30-year-trend of global deaths, DALYs, YLDs, and YLLs are listed in Fig. 7. The gure of different SDI regions (e.g., Low SDI, Low-middle SDI, Middle SDI, Middle-high SDI, High SDI) are listed in the Additional le 2 Fig. 11-15. Most of them are in line with the law.
As shown in Fig. 7, the two risks are similar in countries with different SDIs. By sexes, the pattern is similar to before. Males patients were more than female patients, and occupation injury accounted for a larger proportion than alcohol use.

ACOP and economic parameters
From Fig. 8A, Eastern Europe is the highest ACOP ASIR, ASDR, ASPR, DALYs, YLDs, and YLLs region. Central Europe, Central Asia, and High-income North America are highly in ASIR, ASDR, and YLDs.
While investigating at the national and territorial level, the results are clearer. On the whole, ASIR rises with the rise of SDI, while ASDR is the opposite, and falls with the rise of SDI (Fig. 9A&B). The ASR of prevalence and YLDs have the same trend as ASIR (Additional le 2 Fig. 20, 22). The ASR of DALYs and YLLs have the same trend as ASDR (Additional le 2 Fig. 21, 23). On a whole, in 2019, the countries with lower SDI have a lower ACOP morbidity, but higher mortality.
As shown in Fig. 10, the global level of ACOP parameters was nonlinear tting with SDI from 1990 to 2019. We nd that the nonlinear ttings between parameters are very signi cant. The gures of different SDI regions' nonlinear ttings are list in Additional le 2   [20,21] were published in top medical journals in the early 21st century that the application of hyperbaric oxygen can effectively cure ACOP. Based on the above, we considered that hyperbaric oxygen is one of the main reasons for the decline in ASR of deaths, DALYs, and YLLs of ACOP.
By sex, on the whole, females have a higher ASR of incidence than males. However, the ASDR of females is lower than that of males. This pattern exists in different years, regions, and ages. Previous studies [22][23][24] have shown that estrogen and estrogen receptors can effectively enhance the tolerance of cardiomyocytes and neurons to respond to hypoxia injury. This article appointed that the females' ACOP mortality rate rises rapidly after the age of 60, and the increased rate exceeded that of males. However, the ACOP mortality rate of females up to the age of 80 did not match that of males. Therefore, it is still uncertain whether estrogen and estrogen receptors are the main reasons for females' lower mortality of ACOP. For different poisoning factors, the severity of COP is often different. For different occupations, their sexes are often uneven. For example, there are more male employees than women in heavy industries such as steelmaking. What's more, some technologies can lower the probability of ACOP in certain scenarios. For instance, many countries have seen a substantial reduction in lethal car exhaust poisoning with the rise of catalytic converters. [25,26] This may be the reason for the different severity of COP in different sexes.
The GBD 2019 study only counts two risks of ACOP -occupational injuries and alcohol use. In terms of mortality, occupational injuries are the primary cause of deaths and disability from COP, while alcohol use is the secondary cause in most regions with different SDI level. By sexes, males are more than females. On the whole, COP is more like an occupational disease, and most COP patients are caused by occupational injuries.
Just like other reports based on the GBD database [27,28], the correlation between disease-related parameters and SDI was investigated too. In 2019, the countries with lower SDI have lower ACOP morbidity and higher mortality. Besides, it has a strong correlation with SDI in the past 30 years in global and different SDI regions. Then we included some economic parameters, including Australia coal price, CPI-adjusted, GDP per capita (PPP), GNI per capita (PPP), and SDI. The ACOP-related parameters of most countries in the world are not related to Australia's coal price, while there is a signi cant linear correlation between more than 60% of countries' parameters and GDP per capita (PPP), GNI per capita (PPP), and SDI. Within these parameters, SDI has the best correlation. Some studies [29,30] have reported that economic development or the pollution caused by it is related to the incidence of certain diseases.
The application of fuel is closely related to economic development, and the former is also closely related to the incidence of ACOP.
Therefore, we try to use SDI as a confounding factor to evaluate the trend of ACOP parameters under the non-economic in uence. The in uence of non-economic factors mainly stems from the promotion of medical technology, culture, domestic, and international political environment. The results show that when the economic impact is excluded, the ASIR of most countries in the world and the ASDR of nearly half of the countries are on the rise. Considering that severe CO poisoning can cause serious sequelae to patients, for these countries, the prevention, and control of carbon monoxide poisoning is still very important.
It is worth noting that some traditional developed countries such as the United Kingdom and Denmark have also appeared on the list, which seems very puzzling. On the one hand, it may be caused by errors in some data. On the other hand, taking the United Kingdom as an example, Roca-Barceló et al. [6] investigated the epidemiology of carbon monoxide poisoning in the United Kingdom. They found that the morbidity of ACOP has obvious area-level characteristics (i.e., deprivation, rurality, and ethnicity) from 2002 to 2016. The increase in morbidity may be due to the increase in the poor population, and the increase in the poor population may come from the polarization of the rich and the poor. We have tried to download the GINI coe cient from the database of the WORLD BANK, but the GINI coe cient has too many nulls for effective and reasonable analysis. This is very regrettable.
The research of this article also has certain defects. First of all, exposure to carbon monoxide by self-poisoning or undetermined intent are not included, which contrasts with most countries where suicidal poisoning (car exhaust or charcoal) is one of the leading causes of diagnosed COP. In Australia, for example, accidental poisoning accounts for less than 10% of calls about COP, and the deaths often outnumber admissions.
[31] Secondly, the CPI data was predicted and lled by predictive mean matching, which will cause data distortion. For GDP and GNI data, we directly discard the null value. Although this can ensure the accuracy and authenticity of the data, it may still cause errors between the data and the real situation. Thirdly, our consideration of ACOP parameters and economic parameters is largely based on the data from the last ten years. Mainly because of the social instability in some countries in the early 1990s, and there was a strong transition in ASDRs, DALYs, and YLLs around 2005. After overall consideration, most countries around the world are in a stage of stable development from 2010 to 2019, so the data is more convincing. However, this will result in low values for the 2010-2019 group in Fig. 11.

Conclusion
In conclusion, except for a few countries and regions, the incidence of accidental carbon monoxide poisoning has shown a slow upward trend in the past 30 years, and this trend is highly correlated with economic development in most countries, which may be due to the occupational disease attribute by ACOP. We also try to evaluate the development of medical systems in different countries by multiple linear regression. Some countries must pay more attention to the development of national medical service.

Declarations
Ethics approval and consent to participate Not applicable.

Consent for publication
Yes, all authors consent for publication.

Availability of data and materials
Yes, all the data and materials can be asked from the author.