Forecasting global road traffic injury mortality for 2030

Haruhiko Inada, Qingfeng Li, Abdulgafoor M Bachani, Adnan A. Hyder

Research output: Contribution to journalArticle

Abstract

Objective: To forecast the number and rate of deaths from road traffic injuries (RTI) in the world in 2030. Methods: This study was a secondary analysis of annual country-level data of RTI mortality rates for 1990-2017 in the Global Burden of Disease (GBD) 2017 Study, population projection for 2030, gross domestic product (GDP) per capita for 1990-2030 and average years of schooling among people aged 15 years+ for 1990-2030. We developed up to 6884 combinations of forecasting models for each subgroup stratified by country, sex and mode of transport using linear and squared year, GDP per capita and average years of schooling as potential predictors. We conducted a fixed-size, rolling window out-of-sample forecast to choose the best combination for each subgroup. In the validation, we used the data for 1990-2002, 1991-2003 and 1992-2004 (fit periods) to forecast mortality rates in 2015, 2016 and 2017 (test periods), respectively. We applied the selected combination of models to the data for 1990-2017 to forecast the mortality rate in 2030 for each subgroup. To forecast the number of deaths, we multiplied the forecasted mortality rates by the corresponding population projection. Results: During the test periods, the selected combination of models produced the number of deaths that is higher than that estimated in the GBD Study by 5.1% collectively. Our model resulted in 1.225 million deaths and 14.3 deaths per 100 000 population in 2030, which were 1% and 12% less than those for 2017 in the GBD Study, respectively. Conclusions: The world needs to accelerate its efforts towards achieving the Decade of Action for Road Safety goal and the Sustainable Development Goals target.

Original languageEnglish (US)
JournalInjury Prevention
DOIs
StateAccepted/In press - Jan 1 2019

Fingerprint

Mortality
Wounds and Injuries
Gross Domestic Product
Conservation of Natural Resources
Safety
Population
Global Burden of Disease
Population Forecast

Keywords

  • global
  • longitudinal
  • mortality
  • motor vehicle occupant

ASJC Scopus subject areas

  • Public Health, Environmental and Occupational Health

Cite this

Forecasting global road traffic injury mortality for 2030. / Inada, Haruhiko; Li, Qingfeng; Bachani, Abdulgafoor M; Hyder, Adnan A.

In: Injury Prevention, 01.01.2019.

Research output: Contribution to journalArticle

@article{306cc70cda8940879e2b932655fa137b,
title = "Forecasting global road traffic injury mortality for 2030",
abstract = "Objective: To forecast the number and rate of deaths from road traffic injuries (RTI) in the world in 2030. Methods: This study was a secondary analysis of annual country-level data of RTI mortality rates for 1990-2017 in the Global Burden of Disease (GBD) 2017 Study, population projection for 2030, gross domestic product (GDP) per capita for 1990-2030 and average years of schooling among people aged 15 years+ for 1990-2030. We developed up to 6884 combinations of forecasting models for each subgroup stratified by country, sex and mode of transport using linear and squared year, GDP per capita and average years of schooling as potential predictors. We conducted a fixed-size, rolling window out-of-sample forecast to choose the best combination for each subgroup. In the validation, we used the data for 1990-2002, 1991-2003 and 1992-2004 (fit periods) to forecast mortality rates in 2015, 2016 and 2017 (test periods), respectively. We applied the selected combination of models to the data for 1990-2017 to forecast the mortality rate in 2030 for each subgroup. To forecast the number of deaths, we multiplied the forecasted mortality rates by the corresponding population projection. Results: During the test periods, the selected combination of models produced the number of deaths that is higher than that estimated in the GBD Study by 5.1{\%} collectively. Our model resulted in 1.225 million deaths and 14.3 deaths per 100 000 population in 2030, which were 1{\%} and 12{\%} less than those for 2017 in the GBD Study, respectively. Conclusions: The world needs to accelerate its efforts towards achieving the Decade of Action for Road Safety goal and the Sustainable Development Goals target.",
keywords = "global, longitudinal, mortality, motor vehicle occupant",
author = "Haruhiko Inada and Qingfeng Li and Bachani, {Abdulgafoor M} and Hyder, {Adnan A.}",
year = "2019",
month = "1",
day = "1",
doi = "10.1136/injuryprev-2019-043336",
language = "English (US)",
journal = "Injury Prevention",
issn = "1353-8047",
publisher = "BMJ Publishing Group",

}

TY - JOUR

T1 - Forecasting global road traffic injury mortality for 2030

AU - Inada, Haruhiko

AU - Li, Qingfeng

AU - Bachani, Abdulgafoor M

AU - Hyder, Adnan A.

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Objective: To forecast the number and rate of deaths from road traffic injuries (RTI) in the world in 2030. Methods: This study was a secondary analysis of annual country-level data of RTI mortality rates for 1990-2017 in the Global Burden of Disease (GBD) 2017 Study, population projection for 2030, gross domestic product (GDP) per capita for 1990-2030 and average years of schooling among people aged 15 years+ for 1990-2030. We developed up to 6884 combinations of forecasting models for each subgroup stratified by country, sex and mode of transport using linear and squared year, GDP per capita and average years of schooling as potential predictors. We conducted a fixed-size, rolling window out-of-sample forecast to choose the best combination for each subgroup. In the validation, we used the data for 1990-2002, 1991-2003 and 1992-2004 (fit periods) to forecast mortality rates in 2015, 2016 and 2017 (test periods), respectively. We applied the selected combination of models to the data for 1990-2017 to forecast the mortality rate in 2030 for each subgroup. To forecast the number of deaths, we multiplied the forecasted mortality rates by the corresponding population projection. Results: During the test periods, the selected combination of models produced the number of deaths that is higher than that estimated in the GBD Study by 5.1% collectively. Our model resulted in 1.225 million deaths and 14.3 deaths per 100 000 population in 2030, which were 1% and 12% less than those for 2017 in the GBD Study, respectively. Conclusions: The world needs to accelerate its efforts towards achieving the Decade of Action for Road Safety goal and the Sustainable Development Goals target.

AB - Objective: To forecast the number and rate of deaths from road traffic injuries (RTI) in the world in 2030. Methods: This study was a secondary analysis of annual country-level data of RTI mortality rates for 1990-2017 in the Global Burden of Disease (GBD) 2017 Study, population projection for 2030, gross domestic product (GDP) per capita for 1990-2030 and average years of schooling among people aged 15 years+ for 1990-2030. We developed up to 6884 combinations of forecasting models for each subgroup stratified by country, sex and mode of transport using linear and squared year, GDP per capita and average years of schooling as potential predictors. We conducted a fixed-size, rolling window out-of-sample forecast to choose the best combination for each subgroup. In the validation, we used the data for 1990-2002, 1991-2003 and 1992-2004 (fit periods) to forecast mortality rates in 2015, 2016 and 2017 (test periods), respectively. We applied the selected combination of models to the data for 1990-2017 to forecast the mortality rate in 2030 for each subgroup. To forecast the number of deaths, we multiplied the forecasted mortality rates by the corresponding population projection. Results: During the test periods, the selected combination of models produced the number of deaths that is higher than that estimated in the GBD Study by 5.1% collectively. Our model resulted in 1.225 million deaths and 14.3 deaths per 100 000 population in 2030, which were 1% and 12% less than those for 2017 in the GBD Study, respectively. Conclusions: The world needs to accelerate its efforts towards achieving the Decade of Action for Road Safety goal and the Sustainable Development Goals target.

KW - global

KW - longitudinal

KW - mortality

KW - motor vehicle occupant

UR - http://www.scopus.com/inward/record.url?scp=85071036579&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85071036579&partnerID=8YFLogxK

U2 - 10.1136/injuryprev-2019-043336

DO - 10.1136/injuryprev-2019-043336

M3 - Article

JO - Injury Prevention

JF - Injury Prevention

SN - 1353-8047

ER -