แจ้งเอกสารไม่ครบถ้วน, ไม่ตรงกับชื่อเรื่อง หรือมีข้อผิดพลาดเกี่ยวกับเอกสาร ติดต่อที่นี่ ==>
หากไม่มีอีเมลผู้รับให้กรอก thailis-noc@uni.net.th
หากไม่มีอีเมลผู้รับให้กรอก thailis-noc@uni.net.th
Nuntaporn Klinjun. Statistical methods for transport accident mortality Patterns in Thailand. Doctoral Degree(Research Methodology). Prince of Songkla University, Pattani Campus. : Prince of Songkla University, Pattani Campus, 2014.
Statistical methods for transport accident mortality Patterns in Thailand
วิธีการทางสถิติสำหรับรูปแบบการตายจากอุบัติเหตุขนส่งในประเทศไทย
Name: Nuntaporn Klinjun
Organization :
Prince of Songkla University, Pattani Campus. Faculty of Science and Technology
keyword:
accident
;
สถิติ
;
การตาย
Abstract:
This dissertation applies the appropriate statistical methods for improving the low quality of vital registration (VR) data in order to estimate the actual transport accident deaths in Thailand. This study consists of three parts. The first part aimed to create a model for transport accident deaths based on verbal autopsy (VA) study in 2005 for providing more accurate number of transport accident deaths from vital registration (VR) system in Thailand during 1996 to 2009. The second part aimed to estimate the corresponding number of external causes of death (i.e. transport accident, other accident and suicide). And the third part aimed to assess severity levels and trend directions of transport accident mortality rates for each province of Thailand between 2004 and 2009. Three databases were used including (1) the correct causes of death from VA data in 2005, (2) VR data from 1996 to 2009 obtained from the Thai Ministry of Public Health, and (3) the population denominators in years 2004 to 2009 obtained from the National Statistical Office of Thailand.
In part 1 and 2, logistic regression was used to create three models for transport accident deaths, other accident and suicide for nine provinces, 16 gender-age groups and 14 VR cause-location groups. A receiver operating characteristic (ROC) curve was used to validate the logistic regression models. After that, the models were used
to estimate probabilities of deaths from transport accident, other accident and suicide for the VR data in years 1996 to 2009. In part 3, the estimated number of transport accident deaths from the results of part 1 from 2004 to 2009 was used for further analysis. A Poisson regression model was used to explore patterns of transport accident mortality rates in Thailand from 2004 to 2009 by 16 gender-age groups and 456 province-year groups. A simple linear regression model was used for investigating the direction of association of transport accident mortality rates.
The results of this study show that the number of transport accidents, other accidents and suicides can be predicted correctly by logistic regression models with sensitivities 73.8%, 46.3% and 62.0% and false positive rates 1.6%, 2.0% and 0.6%, respectively. Under-reporting of three causes of death increased over the 14-year period. Transport accident deaths from 1996 to 2009 had higher under-reporting biases over time and had lower gender differences than other accident and suicide deaths. In 2004 to 2009, the mean estimated transport accident mortality rate was 34.5 per 100,000 populations and was highest among males aged 20-29 years. The high mortality rate groups occurred in the Central regions, lower-North region and Nakhon Ratchasima province in the Northeast region. The patterns of the transport accident mortality rates were separated into nine groups: (1) high mortality rate and fast decreasing trend (2) high mortality rate and slow decreasing (3) high mortality rate and slow increasing (4) medium mortality rate and fast decreasing (5) medium mortality rate and slow decreasing (6) low mortality rate and flat (7) low mortality rate and fast decreasing (8) low mortality rate and fast increasing and (9) low mortality rate and slow decreasing. Increasing trends were found in three provinces in the North region, four provinces in the Central and Eastern region and five provinces in the South region.
Prince of Songkla University, Pattani Campus
Address:
PATTANI
Email:
oar.psu@gmail.com
Name: Don McNeil
Role:
Advisor
Created:
2014
Modified:
2015-06-09
Issued:
2015-06-04
วิทยานิพนธ์/Thesis
application/pdf
eng
DegreeName: Doctor of Philosophy
Level: Doctoral Degree
Descipline: Research Methodology
©copyrights Prince of Songkla University, Pattani Campus
RightsAccess:
ใช้เวลา
0.197766 วินาที
Nuntaporn Klinjun
| Title | Contributor | Type |
|---|---|---|
| Statistical methods for transport accident mortality Patterns in Thailand
มหาวิทยาลัยสงขลานครินทร์ วิทยาเขตปัตตานี Nuntaporn Klinjun | Don McNeil | วิทยานิพนธ์/Thesis |
| Estimating external causes of death in Thailand 1996-2009 based on the 2005 Verbal Autopsy study
มหาวิทยาลัยสงขลานครินทร์. สำนักทรัพยากรการเรียนรู้คุณหญิงหลง อรรถกระวีสุนทร Nuntaporn Klinjun;Apiradee Lim;Kanitta Bundhamcharoen | บทความ/Article |
Don McNeil