Road safety in India is a matter of great concern, given that a person dies every 4 minutes on our roads.

In 2015, the National Crime Records Bureau (NCRB) put road accident numbers at over 4,50,000, resulting in 1,41,526 deaths and 4,77,731 injuries. With the government’s proposed Road Transport and Safety Bill 2015, the Transport Ministry has set an ambitious target of saving two lakh lives in the next five years, raising questions on it being just a tall claim.

Road safety experts state that imposing high penalties and hoping that enforcement will bring down accidents and deaths is not enough. Instead, a solid and holistic plan of action comprising the five Es of road safety (Engineering, Enforcement, Emergency Care, Education and Environment) is critical if we are to make any dent in road accidents.

Data-backed intelligence is also a crucial part of any plan. Without data, we will be shooting in the dark and designing interventions that have little impact on improving road safety. We’ve compiled a list of the six most important data sets that would help to improve the road safety act in 2015.

1. Overall road safety data

Foremost, we need to know who is dying on our roads, what age group they are in, what category of vehicle they were driving and what the causes were. Data is typically available with individual police stations and accident thanas in the form of First Information Reports (FIRs). From each station, data is gathered and collated to provide broad classifications on vehicle types, causes, age group etc. However, this does not provide much intelligence. The data collection and analysis methods are not just rudimentary and time consuming, but often flawed due to human intervention. A classic instance is “drunk driving” as a cause of accident, which is underreported in FIRs since the parties involved find it difficult to claim insurance. More accurate technology-based data gathering and analysis is needed to frame stronger short and long-term programs.

We’re written a whole blog about about issues around drunk driving data here. Check it out.

2. Road and vehicle engineering data

An incident shared by a road safety expert goes like this: in a hilly area at a particular steep turn, accidents were taking place at an alarming rate. Despite plastering warning signages, accidents continued, even involving experienced drivers. On thorough investigation, it was found that at that particular turn, a huge rock formation was jutting out of the hill on the right side. To the vehicles climbing from the left, it appeared that an oncoming vehicle would need a larger turning radius and, to accommodate them, climbing vehicles would swerve further left and slip off the road. India does not conduct scientific crash investigation into accidents. As a result, we lack concrete data on road design flaws, vehicle design issues, or even what percentage of accidents are caused by engineering issues. Proper road safety interventions cannot be designed until we have scientific crash investigation data.

3. Road enforcement data

Road safety cannot be achieved without proper enforcement by agencies such as the Police, Traffic Police, and Regional Transport Offices. Herein lies the biggest gap in road safety data — human discretionary element, corruption and lack of technology cause data to be misreported. In this age of technology, it is surprising that we don’t have real-time data on traffic rule violations (including the violator and rules violated), or compliance data on driving schools or candidates seeking licenses. The proposed road safety bill recommends imposing high penalties on violators. However, given the socio-economic conditions of both offenders and enforcers, road offences are going to be further underreported and high penalties will just fuel more corruption. Using tools like SocialCops Collect, real-time violation and enforcement data from the ground can be gathered. Analysis of such data can feed into designing better solution to strengthen enforcement than just increasing penalties.

4. Emergency care data

Fatalities often happen because victims don’t receive proper emergency medical care during the golden hour. In the road accident data published annually by the National Crime Record Bureau, there is hardly any detailed information on deaths caused by lack of timely medical attention. Technology can play a major role here in providing data to help save lives. For instance, a road safety organization called Save Life Foundation is working on an app for the general public to alert the nearest first responder (trained and signed up on the app), who will then rush to provide emergency care. Real-time data from such applications can be used to develop stronger emergency care measures and reduce fatalities.

5. Fatality and injury data

Fatality data collection is not uniform across states. Police stations follow varying practices in counting deaths that happen from accidents. Some include deaths that occur within a specified duration, whereas some others include deaths that occur throughout the time of the investigation. Most records don’t include deaths that take place during prolonged treatment periods. All of these issues are the same for injury related data. Typically, it is estimated that the number of injured are four to five times the number of deaths, but that is an unreliable figure. Besides, there is little information on disability-related data. Programs backed by poor data are often ineffective in bringing down accidents, fatalities and injuries.

6. Impact assessment data

Every year the Ministry of Road Transport and Highways conducts road safety week across the country. Volunteer organizations, state transport departments, public works departments and others conduct regular drives and drills. However, there is no authentic data available on the impact of these interventions in reducing accidents. Road safety audits should be conducted before new construction, then safety should be measured once the roads are in use. Similarly, audits should be conducted before and after programs are implemented on accident-prone zones, and findings should be fed into funding decisions and designing programs.