Home - About Us - Riding Safer - Contact Us - Blog - Disclaimer - Links - Sitemap
Risk Hierarchy: Information - Rider Ed - Driver Ed - Conspicuity - Bike Defect - Ultra-Defensive Riding - Crash Avoidance - Injury Mitigation - Crash Scene

bookThe Thailand Motorcycle Crash Study

Motorcycle Accident Causation And Identification Of Countermeasures In Thailand.

This study was performed by Dr. Vira Kasantikul and a team of researchers in the 1998 to 2001 period. Dr Kasantikul is a neurologist and had additional insight into head and neck injuries. The study was financed by several Honda subsidiaries.

Motorcycling in Thailand is a lot different from the USA. Motorcycles are much smaller, a large proportion being mopeds and step-thrus used for basic transportation. In addition, there wais virtually no rider education available in Thailand at the time of the study. Many of the findings have little application outside Thailand.

It is largely of interest because of the methodology used, and we cherrypick some detailed findings on helmet use, the role of alcohol, the effect of rain and the effectiveness of rider observation and evasion skills.

the study was performed and written up in two parts. Volume 1 was performed in Bangkok in a one-year period in 1998 and 1999, with 723 crashes sampled, and volume 2 is based on 359 crashes sampled in five rural locations in the year 2000. Our review focuses on the Bangkok study except where stated.

We were unable to find a freely distributed copy of the reports online and were kindly given a copy by one of the researchers.

OECD methodology

The Thai study is often cited as the first use of the OECD methodology, although this is not mentioned in Dr. Kasantikul's report. The study researchers were closely assisted by the Head Protection Laboratory and the report credits Jim Ouellet, (who worked for free on this study) and Terry Smith. David Thom of HPRL and Collision and Injury Dynamics has the Thai study on his resume, and co-authored the original OECD standards document with the former, and Sandra Browne and Harry Hurt of HPRL. It is clear that the study was used as a testbed for morphing the Hurt methodology into the OECD emerging standard. The OECD methodology was used in Maids and has been adopted by the OSU study in progress.

The Thailand study used two separate population control groups. On-scene exposure data was collected by videoing and counting passing traffic at accident locations later, on the same day and for an hour around the crash time of day, and in the same weather conditions. They counted hundreds of thousands of vehicles and over 65,000 motorcycles. They also collected 2100 detailed exposure samples by interviewing bikers at gas stations at similar times to the accidents. They avoided the mistake of discounting the effect of rain on accidents by noting that riders to not ride in the rain.

Helmet Use

We have summarized data from the report and from this paper in our Helmet Effectiveness page. As helmet use in Thailand is spotty, more than 20 percent of the riders in the Bangkok study were helmetless, and it is clear that those riders died at more than three times the rate of helmeted bikers, and suffered severe brain injuries at more than four times the rate. Helmets are claimed to prevent or reduce head injuries in more than half the crashes.

We have reservations about the non-inclusion of helmets that came off during the crash. Those riders did even worse than helmetless riders. Many helmets came off because of improper fastening or previous damage, but it is clear that several of the helmets that came off were due to helmet failure and should be counted as helmet riders who died or were injured. Many of these crashes were so traumatic that the riders would have died in any event, but we still think that helmeted riders die at about 40 percent the rate of helmeted ones, and the severe brain injuries about one third the helmetless rate.

We also noted that Dr Kasantikul, being a neurologist, personally performed a modified autopsy on more than 90 percent of the fatalities. He paid extra attention to the head and neck. He found that neck injuries are underreported in normal autopsies. The 42 neck spinal injuries, 2 or 3 of them fatal, is a small number compared to the overall 261 head injuries or the 265 riders whose helmets reduced or prevented head injuries.

We see no reason to support the common contention that helmets provide head protection but increase neck injuries. The numbers are just not there.

We would note that differences in helmet standards and the larger bikes prevalent in the US make it difficult to transpose this information to the US, but heads, gravity and pavement work pretty much the same everywhere.


Alcohol is identified as a huge factor in Thailand crashes. Although the types of alcohol and the social situations of its use is a lot different in Thailand, the findings are relevant.

Alcohol-involved accidents differed markedly from non-alcohol accidents. The differences are quite distinct, and alcohol use was so frequent. Alcohol involved accidents were three times as likely to be single vehicle crashes, three times as likely to involve loss of control, seven times as likely to involve running off the road, and twice as likely to involve violation of traffic control signals or signs when compared to non-alcohol accidents. Alcohol accidents also occurred at higher speeds (about 15 - 20 km/hr on average). Alcohol-involved riders were less likely to wear a helmet and more likely to be hospitalized or to die as a result
of the crash. They were twice as likely to be the principal contributing factor in their accidents, and twice as likely to be the only cause of their accidents. Finally, even their time distributions were different: most alcohol-involved accidents occurred between 10 p.m. and 3 a.m., while most non-alcohol accidents happened between 8 a.m. and 7 p.m.

It is clear that alcohol causes increased risk-taking behaviors and debilitates the cognitive processes required for competent motorcycle riding. Nothing unexpected there, and we would think that the findings for alcohol would be pretty universal.


Rain was discounted as a cause of accidents in both Hurt and Maids, but due to the Thailand on-scene exposure control technique, and the seasonal monsoon conditions in Thailand, this study got it right.

Rain was an infrequent cause factor because most riders did not ride in the rain... However, when it was present, adverse weather often contributed to accident causation. In the 18 cases in which the weather was inclement (i.e., raining) it contributed to accident causation in 12 of those cases, usually by limiting the rider’s ability to see.

Other Issues

The Thailand environment is so different to that in the US, that few of the findings except as noted are of major relevance.

Formal motorcycle training was almost non-existent in Thailand, and issues of road condition, poorly marked street furniture and the sheer numbers of motorcycles played a role.

One point of note is that daily riders had fewer crashes, which might be relevant in US conditions where leisure use of the bike at weekends is more usual.

Rider mistakes, such as poor strategies, deficits in situational awareness, choice of evasion tactics and poor execution of the evasion technique were major factors in crashes, which is matched by Hurt and Maids.