Limits...
Causes of accidental childhood deaths in China in 2010: A systematic review and analysis.

Chan KY, Yu XW, Lu JP, Demaio AR, Bowman K, Theodoratou E - J Glob Health (2015)

Bottom Line: Together, these 7 causes explain more than 80% of all accidental deaths when modeling is primarily used, and more than 95% when the analysis is based purely on medians from the 76 available studies.In this paper we provided a detailed breakdown of causes of these deaths in a large middle-income country.We noted that, wherever the share of accidental deaths among all child deaths is increased, drowning is more likely to be the leading cause; asphyxia seems to be equally important in all contexts, while traffic accidents, poisoning and falls are relatively more important in contexts where the overall share of accidents to all child deaths is low.

View Article: PubMed Central - PubMed

Affiliation: The University of Edinburgh Medical School, Edinburgh, Scotland, UK ; Nossal Institute for Global Health, Melbourne School of Population and Global Health, University of Melbourne. Melbourne, Australia.

ABSTRACT

Background: Infectious causes of childhood deaths in the world have decreased substantially in the 21st century. This trend has exposed accidental deaths as an increasingly important future challenge. Presently, little is known about the cause structure of accidental childhood deaths in low- and middle-income country (LMIC) settings. In this paper, we aim to establish cause structure for accidental deaths in children aged 0-4 years in China in the year 2010.

Methods: In this paper, we explored the database of 208 multi-cause child mortality studies in Chinese that formed a basis for the first published estimate of the causes of child deaths in China (for the year 2008). Only five of those studies identified specific causes of accidental deaths. Because of this, we searched the Chinese medical literature databases CNKI and WanFang for single-cause mortality studies that were focused on accidental deaths. We identified 71 further studies that provided specific causes for accidental deaths. We used epidemiological modeling to estimate the number of accidental child deaths in China in 2010 and to assign those deaths to specific causes.

Results: In 2010, we estimated 314 581 deaths in children 0-4 years in China, of which 31 633 (10.1%) were accidental. Accidental deaths contributed 7240 (4.0%) of all deaths in neonatal period, 8838 (10.5%) among all post-neonatal infant deaths, and 15 554 (31.7%) among children with 1-4 years of age. Among four tested models, the most predictive was used to establish the likely cause structure of accidental deaths in China. We estimated that asphyxia caused 9490 (95% confidence interval (CI) 8224-11 072), drowning 5694 (95% CI 5061-6327), traffic accidents 3796 (95% CI 3163-4745), poisoning 3163 (95% CI 2531-3796) and falls 2531 (95% CI 2214-3163) deaths. Based on medians from a few rare studies, we also predict 633 (95% CI 316-1265) deaths to be due to burns and 316 (95% CI 0-633) due to falling objects. Together, these 7 causes explain more than 80% of all accidental deaths when modeling is primarily used, and more than 95% when the analysis is based purely on medians from the 76 available studies.

Conclusions: Reduction in global child mortality is a leading political priority and accidental deaths will soon emerge as one of the main challenges. In this paper we provided a detailed breakdown of causes of these deaths in a large middle-income country. We noted that, wherever the share of accidental deaths among all child deaths is increased, drowning is more likely to be the leading cause; asphyxia seems to be equally important in all contexts, while traffic accidents, poisoning and falls are relatively more important in contexts where the overall share of accidents to all child deaths is low.

No MeSH data available.


Related in: MedlinePlus

Epidemiological modeling of the association between the most significant predictor variable (proportion of accidental deaths in all child deaths in each study; X–axis) and criterion variable (proportion of deaths caused by asphyxia in all accidental deaths; Y–axis). Data points represent studies with available information and the size of the “bubbles” is proportional to the total number of child deaths observed in each study. The regression line with upper and lower limit of 95% confidence interval is shown across the range of data.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4459091&req=5

Figure 3: Epidemiological modeling of the association between the most significant predictor variable (proportion of accidental deaths in all child deaths in each study; X–axis) and criterion variable (proportion of deaths caused by asphyxia in all accidental deaths; Y–axis). Data points represent studies with available information and the size of the “bubbles” is proportional to the total number of child deaths observed in each study. The regression line with upper and lower limit of 95% confidence interval is shown across the range of data.

Mentions: Finally, the fourth aim of our study was to investigate if there are any significant context–related predictors of the proportional contribution of specific causes to the overall number of deaths due to accidents. This aim should inform us whether we can predict the proportional contribution of different specific causes if we have other information about the context. We could specify three different predictors that were available at the provincial level: (i) overall U5MR; (ii) U5MR that is due to all accidental deaths; and (iii) proportional contribution of accidental deaths to all deaths in each province. We used regression analysis and three separate models to explore whether any of these three predictor variables are significantly associated with proportion of any specific cause in all accidental deaths. We also added the fourth model, where we used multivariate design to account for all three predictors at the same time. We used those four models to predict the proportional contribution of the five specific causes with sufficient information available: drowning, asphyxia, traffic accidents, falls and poisoning. In performing these analyses, we followed all procedures as detailed in our previous paper [4]. We presented a summary of the four models applied to five specific causes of accidental deaths in Table 2 and Figures 2 to 6.


Causes of accidental childhood deaths in China in 2010: A systematic review and analysis.

Chan KY, Yu XW, Lu JP, Demaio AR, Bowman K, Theodoratou E - J Glob Health (2015)

Epidemiological modeling of the association between the most significant predictor variable (proportion of accidental deaths in all child deaths in each study; X–axis) and criterion variable (proportion of deaths caused by asphyxia in all accidental deaths; Y–axis). Data points represent studies with available information and the size of the “bubbles” is proportional to the total number of child deaths observed in each study. The regression line with upper and lower limit of 95% confidence interval is shown across the range of data.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4459091&req=5

Figure 3: Epidemiological modeling of the association between the most significant predictor variable (proportion of accidental deaths in all child deaths in each study; X–axis) and criterion variable (proportion of deaths caused by asphyxia in all accidental deaths; Y–axis). Data points represent studies with available information and the size of the “bubbles” is proportional to the total number of child deaths observed in each study. The regression line with upper and lower limit of 95% confidence interval is shown across the range of data.
Mentions: Finally, the fourth aim of our study was to investigate if there are any significant context–related predictors of the proportional contribution of specific causes to the overall number of deaths due to accidents. This aim should inform us whether we can predict the proportional contribution of different specific causes if we have other information about the context. We could specify three different predictors that were available at the provincial level: (i) overall U5MR; (ii) U5MR that is due to all accidental deaths; and (iii) proportional contribution of accidental deaths to all deaths in each province. We used regression analysis and three separate models to explore whether any of these three predictor variables are significantly associated with proportion of any specific cause in all accidental deaths. We also added the fourth model, where we used multivariate design to account for all three predictors at the same time. We used those four models to predict the proportional contribution of the five specific causes with sufficient information available: drowning, asphyxia, traffic accidents, falls and poisoning. In performing these analyses, we followed all procedures as detailed in our previous paper [4]. We presented a summary of the four models applied to five specific causes of accidental deaths in Table 2 and Figures 2 to 6.

Bottom Line: Together, these 7 causes explain more than 80% of all accidental deaths when modeling is primarily used, and more than 95% when the analysis is based purely on medians from the 76 available studies.In this paper we provided a detailed breakdown of causes of these deaths in a large middle-income country.We noted that, wherever the share of accidental deaths among all child deaths is increased, drowning is more likely to be the leading cause; asphyxia seems to be equally important in all contexts, while traffic accidents, poisoning and falls are relatively more important in contexts where the overall share of accidents to all child deaths is low.

View Article: PubMed Central - PubMed

Affiliation: The University of Edinburgh Medical School, Edinburgh, Scotland, UK ; Nossal Institute for Global Health, Melbourne School of Population and Global Health, University of Melbourne. Melbourne, Australia.

ABSTRACT

Background: Infectious causes of childhood deaths in the world have decreased substantially in the 21st century. This trend has exposed accidental deaths as an increasingly important future challenge. Presently, little is known about the cause structure of accidental childhood deaths in low- and middle-income country (LMIC) settings. In this paper, we aim to establish cause structure for accidental deaths in children aged 0-4 years in China in the year 2010.

Methods: In this paper, we explored the database of 208 multi-cause child mortality studies in Chinese that formed a basis for the first published estimate of the causes of child deaths in China (for the year 2008). Only five of those studies identified specific causes of accidental deaths. Because of this, we searched the Chinese medical literature databases CNKI and WanFang for single-cause mortality studies that were focused on accidental deaths. We identified 71 further studies that provided specific causes for accidental deaths. We used epidemiological modeling to estimate the number of accidental child deaths in China in 2010 and to assign those deaths to specific causes.

Results: In 2010, we estimated 314 581 deaths in children 0-4 years in China, of which 31 633 (10.1%) were accidental. Accidental deaths contributed 7240 (4.0%) of all deaths in neonatal period, 8838 (10.5%) among all post-neonatal infant deaths, and 15 554 (31.7%) among children with 1-4 years of age. Among four tested models, the most predictive was used to establish the likely cause structure of accidental deaths in China. We estimated that asphyxia caused 9490 (95% confidence interval (CI) 8224-11 072), drowning 5694 (95% CI 5061-6327), traffic accidents 3796 (95% CI 3163-4745), poisoning 3163 (95% CI 2531-3796) and falls 2531 (95% CI 2214-3163) deaths. Based on medians from a few rare studies, we also predict 633 (95% CI 316-1265) deaths to be due to burns and 316 (95% CI 0-633) due to falling objects. Together, these 7 causes explain more than 80% of all accidental deaths when modeling is primarily used, and more than 95% when the analysis is based purely on medians from the 76 available studies.

Conclusions: Reduction in global child mortality is a leading political priority and accidental deaths will soon emerge as one of the main challenges. In this paper we provided a detailed breakdown of causes of these deaths in a large middle-income country. We noted that, wherever the share of accidental deaths among all child deaths is increased, drowning is more likely to be the leading cause; asphyxia seems to be equally important in all contexts, while traffic accidents, poisoning and falls are relatively more important in contexts where the overall share of accidents to all child deaths is low.

No MeSH data available.


Related in: MedlinePlus