E/escap/cst/inf/9: Improving vital statistics and cause of death statistics: The experience of Thailand

Results 

35. 

The reassessment of causes of death via verbal autopsy on all sample deaths, substituted by 

medical records review where possible, resulted in a substantially different broad cause of death 

distribution to that suggested by the vital registration system (see Figure 2). In particular, verbal 

autopsy methods led to a massive reduction in the percent of deaths assigned to ill-defined causes for 

both males and females, declining from 35-45% of all deaths in vital registration to 4-6%. Most of 

these ill-defined causes were reassigned to cardiovascular diseases, with substantial reassignment also 

to cancers and other non-communicable diseases. Interestingly, deaths from injuries appear to be 

under-diagnosed by about 20-25% for both males and females, whereas proportionate mortality from 

infectious diseases combined appears to be fairly well reported in the vital statistics.  

 

---

E/ESCAP/CST/INF/9 

Page 12 

 

 

Figure 2. Difference in broad causes of death from vital registration (registered) and verbal