Archive for the ‘NHTSA’ category

Why do we believe what we believe about motorcycle helmets?

April 21, 2010

In the last entry we saw that ordinary people in ordinary circumstances misjudge the actual ability of protective gear to reduce or prevent injury and take on more risk that uses up that safety margin. Motorcyclists are just as likely to fall prey to risk compensation as others. But how do motorcyclists—and non-riders—come to have an exaggerated belief that helmets, specifically, are more effective than they are?


Let’s first take a look at what experts say about helmets. For the sake of conciseness, I’m going to sum up and put longer quotes and links in footnotes:

NHTSA claims that “Motorcycle helmets provide the best protection from head injury for motorcyclists involved in traffic crashes.”[i]

The Michigan State Police claim that “Helmets decrease the severity of injury, the likelihood of death, and the overall cost of medical care…. Just like safety belts in cars, helmets can’t provide total protection against head injury or death, but they do reduce the incidence of both.[ii]

The American College of Emergency Physicians says  “Head injury is the leading cause of death in motorcycle crashes, and helmets provide the best protection from head injuries…”[iii]

Advocates for Highway and Auto Safety—long seen as opposing motorcycling in general—says, “Motorcycle helmets have been shown to save the lives of motorcyclists and prevent serious brain injuries.”[iv]

The Insurance Institute of Highway Safety (IIHS) states the exact same thing in the exact same words as the Michigan State Police website so we’ll use a different part of the quote:  In the event of a crash, unhelmeted motorcyclists are three times more likely than helmeted riders to suffer traumatic brain injuries…”[v]

MSF has a .pdf flyer on helmets that states that “Helmet use is not a “cure-all” for motorcycle safety, but in a crash, a helmet can help protect your brain, your face, and your life.

“Combined with other protective gear, rider-education courses, proper licensing and public awareness, the use of helmets and protective gear is one way to reduce injury.”[vi]

MSF’s Basic RiderCourse handbook states, “Helmets work well in accomplishing their intended function to protect the head and brain from injury…helmet effectiveness has been confirmed by research, not just in the laboratory, but by decades of actual crash analysis as well. So, be safe and always wear a helmet while riding…Since head injuries account for the majority of motorcycle injuries, head protection is vital. The best helmet is no guarantee against injury, but statistics indicate that helmet use reduces the risk of brain injury by 67 percent (and gives the NHTSA 2004 “Traffic Safety Facts” report as the source of the statistic).[vii] However, the NHTSA 2004 Traffic Safety Report

does not contain that statistic.

Media articles on motorcycle safety also repeat the same claims.

Media articles typically include whether a rider was wearing a helmet or not—and do so far more often than whether drivers were wearing seatbelts as in this short news item on the death of a rider from The Geneva County Reaper,

“Motorcyclist killed in wreck” A 60-year-old motorcycle rider died on Easter Sunday in a single vehicle wreck on Walton County Road 181.

Ronnie Denza Hughes was headed west when the bike traveled across the eastbound lane and onto the shoulder, striking a tree, according to the Florida Highway Patrol. The bike rotated and came to rest facing south.

The accident took place around 7 p.m. Hughes was not wearing a helmet.”

WEAU 13 NEWS in Eau Claire, WI published an article on April 13 of this year,  “Motorcycle riders and law enforcement warn about motorcycle safety.” It said, in part, “…“We highly recommend people wear helmets they’re not required by law, unless your under 18 or have an instructional permit, but a helmet’s gonna definitely save you from serious injury in case you are involved in a crash,” Sgt. Jerry Voight with the Wisconsin State Patrol says.”[viii]

The Columbus Dispatch, published an article on April 3, “Caution urged in motorcycle season: Deaths a grim reminder for riders, motorists”.

The latter part of the article focuses on the human interest element. After first detailing how one unhelmeted rider died in a crash it goes on to tell about another fatality: “Computer developer Joseph Matello, 40, of Riverstone Drive in Columbus, died after a crash about 11a.m. Thursday on the Far West Side. Police said he crossed the center line on Feder Road and struck a car head-on.

“His wife, Stephani, said Matello was a strong believer in safety, and a helmet saved his life a few years ago when a car driver didn’t see him and struck him.”[ix]

Iow, even though the crash was—for whatever reason—his fault and though a helmet was worn and did not save his life, the article still stresses how important wearing a helmet is—and that it had saved his life years before.

Reasonable to believe helmets are effective

The above is just a fraction of all the repeated direct and implied claims by those who present themselves as experts. The story told by different groups circle around on themselves by citing each other—and most often NHTSA.

The very official status of the sources gives credibility to their claims. That story then is willingly propagated through the media that repeats those claims and adds testimonials from both dealers and riders—or in the last case, the dead rider’s spouse.

It’s highly likely that a reasonable person, after reading even a portion of the above would believe that helmets were highly effective in preventing death and reducing injuries. In fact, it would be unreasonable to disbelieve such repeated accounts.

As we’ve seen, ordinary people—which fulfills the legal definition of a reasonable person—take more risks in ordinary ways simply because they believe they are safer because they are wearing some kind of protective gear.

Iow, it’s reasonable that a reasonable person would act upon such repeated safety claims and to take on risks he or she wouldn’t if they weren’t wearing a helmet. For example—the risk of riding a motorcycle at all. We

Iow, we believe that helmets are effective because we’ve been told over and over by credible sources that they are. And we don’t just act upon that belief, we stake our lives on it.

But the thing is—we don’t have to take on anything more than the most ordinary risks of riding to outride the protection a helmet can give in the most ordinary circumstances.

Given the strong chorus of approval and recommendations from safety and transportation interests and experts, it’s exceedingly interesting and illuminating and especially surprising—what helmet manufacturers say about their products. Or rather, what they don’t say.

[i] Helmet Use Laws. NHTSA.

[ii] “They’re designed to cushion and protect riders’ heads from the impact of a crash…. Motorcycle crash statistics show that helmets are about 37 percent effective in preventing crash fatalities. The National Highway Traffic Safety Administration estimates an unhelmeted rider is 40 percent more likely to suffer a fatal head injury and 15 percent more likely to incur a nonfatal head injury than a helmeted motorcyclist.”,1607,7-123-1593_3504_22760-13677–,00.html

[iii] “Helmets are estimated to be 37 percent effective in preventing fatal injuries to motorcycle riders. (NHTSA)… Everyone is only one step away from a medical emergency….Helmet use is the single most important factor in people surviving in motorcycle crashes. They reduce the risk of head, brain and facial injury among motorcyclists of all ages and crash severities. Unhelmeted motorists are 40 percent more likely to die from a head injury, according to the National Highway Traffic Safety Administration (NHTSA).”


[v] “Helmets decrease the severity of head injuries, the likelihood of death, and the overall cost of medical care. They are designed to cushion and protect riders’ heads from the impact of a crash. Just like safety belts in cars, helmets cannot provide total protection against head injury or death, but they do reduce the incidence of both. NHTSA estimates that motorcycle helmets reduce the likelihood of crash fatality by 37 percent….Helmets are highly effective in preventing brain injuries, which often require extensive treatment and may result in lifelong disability.” This quote appears verbatim on several other websites.

[vi] “Second, a good helmet makes riding a motorcycle more fun, due to the comfort factor: another truth.

“Third, wearing a helmet shows that motorcyclists are responsible people; we take ourselves and motorcycling seriously. Wearing a helmet, no matter what the law says, is a projection of your attitude toward riding. And that attitude is plain to see by other riders and non-riders alike.”


[viii] “State troopers say just wearing a helmet and the proper gear could help save your life People who drive motorcycles say the feel of the wind on your face is a thrilling experience, Wisconsin doesn’t require helmets, but those who sell motorcycles and those who enforce the law, say safety needs to be of utmost importance.

[ix] “She said she has a message for other motorcyclists: “For riders, wear as much protective gear as possible.

“For cars, watch for them. They’re everywhere, and it only takes a second to take somebody’s life.”


What seat belt usage can teach us about motorcycle safety

February 28, 2010

As we’ve been told again and again, far more drivers wear seat belts than riders wear helmets. The National Occupant Protection Use Survey (NOPUS) estimates seat belt use at 83 percent in 2008 while helmet use at 67 percent in 2009. Statistics like that increase the perception motorcyclists don’t care much about personal safety. But seat belt history offers some insight into helmet use—and a different look helmet use history might change our perception about motorcyclists’ choice:

Manufacturers get the first mandate Seat belts were invented in the mid 1890s just as automobiles hit American streets, but it wasn’t until 1949 that Volvo and Nash first put seat belts in cars.[i] Few other manufacturers followed suit though and few people wore them.

State legislators, convinced of seat belt efficacy first demanded manufacturers put them in cars. By 1964 only half the states had the first seat belt laws—but that’s all it took; a year later all car manufacturers offered seat belts as standard equipment in every state. In 1972 the National Highway Safety Foundation (NHTSA) made it a federal requirement. But usage was extremely low—less than 11 percent.

Education fails Before and during this, though, a huge marketing effort (including the famous Buckle Up For Safety commercials) and an enormous public relations/media campaign to tout seat belt use was flooding the nation. And arguments raged about whether seat belts really were safe or more dangerous, which also happened with helmets.

More regulation In 1974 NHTSA required a buzzer/light reminder system or ignition locks to make it harder not to use seat belts. Ignition locks were more effective than the annoying sound/light that is still with us today. One study with a small number of drivers  found that usage rose to 67 percent but decreased over time as many owners disconnected the system or left them belted to circumvent the light/buzzer or lock.[ii] Studies using rental cars found that there was an insignificant difference in use between cars with or without the warning system.

Legislation not education Seat belts in cars and positive publicity was ineffective: usage was in the low teens through the 1970s. Iow, the public responded to seat belts as we’ve been led to believe riders responded to helmets.

It was only when mandatory seat belt laws were passed that use began to rise by 17-26 percent.[iii] California is a prime example: Before the mandatory seat belt law was passed in 1986 use was 26 percent. After the law it rose to 45 percent and crept up to 73 percent by 1993. After a primary enforcement law (meaning law enforcement could stop a driver solely for seat belt use) was passed in 1993 it rose to 83 percent and to 91 percent by 2002.[iv] Even so, by 2002, national usage was only 75 percent (and has since risen to 83 percent).

Negatives drive seat belt use And even recent studies find it’s only that high because of a combination of factors: use is higher in a primary enforcement states than in secondary enforcement state (where they have to have another reason to stop you). Use is higher among those who have a higher fear of getting a ticket than those who don’t think they at risk of a traffic stop. It’s higher when the ticket has a higher financial penalty. And studies have found that family and friends’ seat belt behavior matter and their pressure to buckle up matters and a general public attitude matter in influencing a driver’s behavior.

Otoh, programs educating drivers as to the risk and nature of injuries, offering incentives or raising fear of injuries weren’t very effective and had high recidivism. Once seat belt use becomes habitual, though, it tends to be self-maintaining.

Iow it’s the negative that drives seat belt usage until habit takes over and the decision is mindless. This attitude is so entrenched that the Committee for the Safety Belt Technology Study for the Transportation Research Board of the National Academies state that those who always wear belts, “… simply follow rules they have developed on the basis of experience, rather than continuously comparing risks against benefits in deciding whether to buckle up.”[v]

Part-time belt users gave these reasons for not wearing a belt included: driving a short distance (59 percent), forgetting to buckle up (53 percent); being in a rush (41 percent); and discomfort from the seat belt (33 percent). These are also reasons that some riders give for not wearing a helmet.

Non-users were by far the smallest percentage of the survey and gave some of the same reasons—laziness, short distances, forgetting, low speeds, short distances but also, “Many hard-core nonusers object to being forced to buckle up, believing that belt use should be a matter of personal choice.” This reason is the same argument anti-helmet law activists give for resisting helmet laws.[vi] Iow, we’re not so different than drivers when it comes to not wearing safety gear.

More of the same only tougher However the safety community is convinced that even habit is not enough; the Committee stated, “Strong enforcement is a necessary component of effective seat belt use laws. Motorists must be convinced that violators will be ticketed and nontrivial penalties exacted.”

The Prevention Institute article referred to a report published in 2000, in which  Transportation Department Inspector General Kenneth Mead stated, “Unless additional states enact and enforce primary laws, which are the most effective means of increasing seatbelt use, we see no credible basis to forecast increases in excess of the recent trend,” Mead stated in the report.

Iow, when it comes to helmets and belts traffic safety experts reject education as an effective tool when it comes to wearing safety equipment. Ever-tougher legislation is seen as the only way to force compliance.

Riders, though, don’t behave as drivers However for much of the past 30-some years, helmet use has been higher than seat belt use in states that don’t have helmet laws but do have seat belt ones. And helmet use in universal helmet law states has been higher than seat belt use in those same states before seat belt laws were passed.

Once again, we look at California: According to the Highway Loss Data Institute unit of the Insurance Institute of Highway Safety (IIHS), helmet use before the universal law was passed was 50 percent. Iow, it was already 24 percent higher than seat belt use was before the mandatory seat belt law was passed.

Immediately after California instituted a universal helmet law in 1992, use surged to 99 percent.[vii] In comparison, it took 16 years and a harsh primary enforcement law to achieve slightly less when it came to drivers.

While it’s true that helmet compliance is more obvious than shoulder/lap belt use,[viii] voluntary helmet use was already almost twice as high when the law was passed as voluntary seat belt use was before the seat belt law was passed. And driver compliance only achieved rider compliance after a strict primary enforcement law was instituted.

This is a significant and positive safety difference between drivers and riders that has been unobserved and unstudied.

But it is seat belts we’re talking about and they are provided in every car sold and  require little effort or discomfort to use and have overwhelming social approval attached to their use.

Otoh, even the lightest helmet is a distinct weight on the head, it’s hot to wear at times and the snug fit that’s required for effectiveness is uncomfortable for many. It can catch the wind causing neck strain and some feel that it obstructs their vision. And unlike seatbelts, a helmet must be replaced if it comes in violent contact with a hard surface. To top it off,[ix] even cheap ones are expensive and require additional  effort (compared to seatbelts) to obtain.

Riders’ performance actually better Despite all that, nationally, helmet use is still 67 percent even though only 20 states have universal helmet laws while seatbelt use is finally 83 percent 45 years after seatbelts were standard equipment in cars sold in the USA—even though 49 states have a mandatory seatbelt laws. And that’s a profound safety difference between drivers and riders that has been unobserved, unstudied and unappreciated.

While traffic experts bemoan the low rate of helmet use an equally valid case could be made for the high use of helmets in states without mandatory laws and in states prior to the passage of universal helmet laws. Considering the history of seat belt use, it’s rather extraordinary that so many riders choose on their own to purchase expensive, heavy and uncomfortable helmets and wear them when they aren’t required by law or receive any immediate benefit or incentive for doing so.

In fact, it suggests that riders who choose to wear helmets without a mandate are the opposite of extraordinary risk-takers. Instead it suggests that they are more aware of the risks inherent in motorcycling, believe that their odds of crashing are higher and take steps to mitigate harm.

Iow, it suggests that a significant proportion of motorcyclists take more personal responsibility for their own safety than drivers do.

And that’s a very different view of motorcyclists.

[i] Coincidentally, 1949 was the year Smeed published his “law”.

[ii] Buckling Up: Technologies to Increase Seat Belt Use — Special Report 278. Transportation Research Board (TRB). 2004.

[iii] Curtisa, Kevin M. and Scott W. Rodia and Maria Grau Sepulveda. The lack of an adult seat belt law in New Hampshire: Live free and die? Accident Analysis & Prevention, Volume 39, Issue 2, March 2007, Pages 380-383.

[iv] Gantz, Toni and Gretchen Henkle. Seatbelts: Current Issues. Prevention Institute. October 2002. Highway Loss Data Institute, Insurance Institute of Highway Safety. Q&As: Motorcycle helmet use laws. January 2009.

[v] Committee for the Safety Belt Technology Study. Buckling Up: Technologies to Increase Seat Belt Use, Special Report 278. Transportation Research Board. 2004.

[vi] It would be interesting if someone did a study to find out if those who didn’t wear helmets also didn’t wear seat belts.

[vii] Highway Loss Data Institute, Insurance Institute of Highway Safety. Q&As: Motorcycle helmet use laws. January 2009.

[viii] Though whether the helmet is DOT-certified is not as easy to determine.

[ix] All plays on words in the article are intentional.

The strange case of Louisiana and helmet-no-helmet-helmet…

February 4, 2010

We return now to helmets since they are, beyond a doubt, the most often touted measure to increase motorcycle safety. In the next two entries, we will examine one state’s data over the same ten-year period we looked at in terms of crashing. What we will find illustrates many of the problems with motorcycle safety:

Louisiana is a unique case when it comes to motorcycle safety: Like almost all other states, a universal helmet law was passed in the 1960s (1968, to be exact) and then  modified it in 1976 to only apply to those under 18. Then legislators passed a universal helmet law again in 1982. Seventeen years later, the law was modified once again to require only those under 18 or those without coverage of at least $10,000 in medical insurance to wear a helmet and was in effect in August of 1999. Five years later, in 2004, the universal helmet law was reinstated.

During the same time frame that we examined crashes in terms of three kinds of crashes (fatalities, injuries and property-only crashes), then, Louisiana didn’t have a universal helmet law and then did have one. It is, then, a good state to see the effects of helmet and non-helmet use and the effect of a helmet law in terms of fatalities and injuries.  In this entry we look only at fatalities.

Louisiana in comparison to national picture

As in the rest of the USA, motorcycle fatalities increased from 40 in 1998 to 87 in  2006 or 117.5 percent. In 2008 fatalities were 77 or 92.5 percent and then to 92 in 2009. Overall, then Louisiana’s fatality toll rose 130 percent in eleven years. This is over the national percentage over the same time span.

As in the rest of the USA, motorcycle registrations were increasing in Louisiana during these years—though there are no publicly reliable statistics to document the increase—not even for NHTSA.[i] While the researchers who prepared the report on the effect of the helmet law repeal dismissed the importance of increasing registrations on crash rates, others would strongly disagree.  As we’ve seen in other states we’ve examined, registrations did not increase unilaterally in every state nor from year to year. The missing data, then, leaves a huge gap—if motorcycle safety is a puzzle, then this is information we need to solve it.

Observed helmet use over the ten years

In the years when the universal helmet law was in effect, between 87-100 percent of riders wore them. In the years without mandated helmet use, usage dropped immediately to 52 percent then rose to about 60 percent, dropped to 48 percent rose again to about 60 percent and stayed there until the universal helmet law was reinstated when observed helmet use appears to be 98-99%.

Unsurprisingly, when helmet use dropped unhelmeted fatalities went up[ii] and helmeted fatalities went down. And, unsurprisingly, when the mandate was reinstated, helmeted fatalities went up.[iii]

Studies were done, papers and reports were written to document the effect of the helmet law repeal—and then the reinstatement. Including NHTSA who released a report, “Evaluation of the Reinstatement of the Motorcycle Helmet Law in Louisiana”.

And the stats were astonishing—unhelmeted fatalities increased by 40% the first year and 42% the second year after the repeal and dropped 64% the year after the universal helmet law was repealed. It seems to prove the case—and is religiously held by helmet proponents: helmets save lives.

And, as helmet-proponents would expect, helmeted fatalities were under-represented compared to observed helmet-use:

And unhelmeted injuries were over-represented to non-use:

The helmet story seems to be justified.

However, if we look without preconceptions more closely at the actual numbers, that’s not the full story—and by not telling the full story, motorcycle safety researchers have not served riders well. The data in the following chart Data taken from the Louisiana State University Traffic Safety Research Group traffic safety data reports.

UnHelmet Fatalities Year-to-Year ChangeUnH Estimated Non-Helmet Use Total F Fatality Increase from prior year Observed Helmet Use Year-to-Year Change HEL Helmet Fatalities
1999 15 3% 40 14.2% 97% 25
2000 21 40% 48% 50 25% 52% 16% 29
2001 36 42% 41% 53 6% 59% -41% 17
2002 41 14% 58% 58 9.4% 42% 0% 17
2003 55 34% 40% 77 32.7% 60% 29% 22
2004 36 -34% 40% 68 -11.6% 60% 45% 32
2005 13 -64% 2% 70 -2.9% 98% 78% 57
2006 8 -38% 2% 87 24.2% 98% 38% 79
2007 14 43% 73 -16% -25% 59
2008 19 36% 77 5.4% -1.7% 58
2008 15 -21% 91 18.1% 31% 76
Totals 258 653 395

The NHTSA report doesn’t note that once the law was reinstated,

helmeted fatalities went up 45% then 78% then 38%. Iow, equally scary percentages are found on both sides of the Louisiana Experiment.

In fact, in 2004—the year of reinstatement—unhelmeted fatalities dropped 23—but helmeted fatalities rose 25. It’s almost as if it was simply a trade-off: unhelmeted deaths became helmeted deaths. This is the first indication that even though helmets save lives, their absence doesn’t drive the increase in motorcycle deaths.

While it is true that unhelmeted fatalities increased every year, however helmeted fatalities also increased 8 out of the 10 years—and that included two years when helmet use was depressed. This is the second indication that helmet use is disconnected from the rise or fall of fatalities in essential ways.

Ultimately this is the scariest thing of all: During these years, Louisiana’s motorcyclist death toll soared and helmeted fatalities tripled while unhelmeted fatalities in 2008 returned to exactly what they were in 1999.

The large point is this: the issue isn’t helmet effectiveness. Focusing merely on helmet use or encouraging riders to wear them simply doesn’t address why those fatal crashes are occurring more and more often. In this way, the increases on the helmeted side of the equation are more troubling since the additional protection helmets offer were insufficient in preventing deadly injuries.  We’ll return to this point in the next entry.

In reality, small numbers

But the focus on helmet use disguises something that is also essential to really understanding what’s happening with riders: small numbers make for big—but misleading—percentages on both sides of the equation. And in Louisiana, the numbers are sometimes very small:

2000: Helmet use was 52%. Fatalities increased by 10 riders.

Unhelmeted fatalities up by 6.

Helmeted fatalities up by 4.

2001:Helmet use: 59%. Death toll increased by 3.

Unhelmeted fatalities went up by 15.

Helmeted fatalities down by 12.

2002: Helmet use: 42% fatalities. Fatalities up by 5.

Increase was solely accounted for by unhelmeted deaths.

2003: Helmet use 60%. Fatalities up by 19.

Unhelmeted fatalities up by 14.

Helmeted fatalities up by 5.

2004: Helmet law use 60%. Fatalities down by 9.

Unhelmeted fatalities down by 19.

Helmeted fatalities up by 10.

2005: Helmet use was 98%.[iv] Death toll increased by 2.

Unhelmeted fatalities down by 23.

Helmeted fatalities up by 25.

2006: Helmet use 98%. Fatalities increased by 17.

Unhelmeted fatalities down by 5

Helmeted fatalities up by 22.

2007: Estimated helmet use 98%. Fatalities decreased by 14

Unhelmeted fatalities up by 6.

Helmeted fatalities down by 20.

2008: Estimated helmet use 98%. Fatalities increased by 4.

Unhelmeted fatalities down by 5.

Helmeted fatalities up by 18.

Small numbers yield large percentages—but skew the interpretation

There’s several things to observe: If  fatalities can legitimately be expected as a function of the numbers, some kind of increase could be—and maybe should be—expected because motorcycle registrations were increasing. But even as the pool of motorcyclists was growing fatalities increased by 5 or less in four of the ten years.

And 5 or less are surely small numbers—and yet they can loom large and appear to be more important than they perhaps are. For example, a decrease of unhelmeted 5 deaths in 2005—the year after the helmet law reinstatement—resulted in a decrease of 38% and an increase in 5 helmeted deaths results in a 29% increase. Using percentages to make a point without including the numbers can lead to inaccurate analysis and, in this case, exaggerate the supposed effect of the reinstatement when the reality is that people who might have died without helmets died with them on.

A pattern of wild fluctuations

Louisiana’s pattern is one of wild fluctuations where fatalities creep up or down by 5 or less or soar by 22 one year and plunge the next by 20 then rise to 18. This pattern of dramatic year-to-year shifts is not seen in states with large numbers of fatalities where they produce smaller percentages of increase.

Such wild swings raise the possibility that crashes have much more to do with random factors or at least factors that are not considered and therefore not accounted for than any crash causation study to date has considered.

While Louisiana’s motorcycle registrations are shrouded in mystery for some inexplicable reason, the state does provide data on injuries as well as fatalities—and that’s where we follow the helmet story next.

[i] Observed helmet use statistics interpreted from Evaluation of the Reinstatement of the Motorcycle Helmet Law in Louisiana, NHTSA, Number 346 May 2008.

[ii] The helmet law was repealed in August. The death toll went up by 5 but we don’t know if they were helmeted or not.

[iii] Evaluation of the Reinstatement of the Motorcycle Helmet Law in Louisiana.

[iv] Ibid.  According to NHTSA and quoted in other studies.

NHTSA releases study on motorcyclist injuries

August 19, 2009

The National Highway Traffic Safety Administration (NHTSA) recently released DOT HS 811 149, “Traffic Safety Facts Research Note: Motorcyclists Injured in Motor Vehicle Traffic Crashes”. It examines some very basic crash data from 1998-2007 using crashes that resulted in police-reports.

Since far more motorcyclists are injured than killed and injury is a far better outcome than death, it would be most beneficial to determine the key factors that make one crash outcome injury and another one fatal. So something like this has been long-overdue though this study looks at very few elements.

For the most part, the analysis found that while more riders are injured, the proportions in each element the study looked at are almost identical to what they were back in 1998.  What NHTSA didn’t do in all but two cases is compare those injury percentages to fatality percentages—and when we do that, the answer may be disturbing: the biggest difference between survivable and non-survivable crashes may only be the degree of injury and not different situations.

From 1998-2007:

  • Injuries have increased 110 percent while motorcycle registrations increased 84 percent.
  • About half the injuries continue to occur in single-vehicle and half in multi-vehicle crashes
  • About half (55%) of all injury crashes continue to occur during the week and half (45%) occur on the weekend. (However, weekend crashes occur 1.5 more often since the weekend is a much shorter length of time).
  • About 60% of the crashes continue to happen during daylight hours,
  • and 68% occur from April through September.
  • Ninety-percent of injuries happen to riders and 10 percent to passengers.
  • About 85% of all the injured are male and 15% female.
  • Injuries increased among all age groups—but the age group with the largest number of injuries was in the 20-29 year-old age group.
  • And almost the exact same percentage of riders were injured in alcohol-related crashes in 2007 (9%) as in 1998 (10%).

Helmet use The NHTSA report states, “Among the 103,000 motorcyclists injured in motor vehicle crashes in 2007, 65,000 (63%) were helmeted at the time of the crash, 31,000 (30%) were not helmeted, and helmet use was unknown for 7,000 (7%) of the motorcyclists injured. Helmet use from 1998-2007 among motorcyclists injured in crashes has ranged from a low of 55 percent in 2002 and 2003 to a high of 63 percent in 2007. Of the motorcyclists injured in crashes who were not helmeted, the proportion ranged from a high of 40 percent in 2002 to a low of 30 percent in 2006 and 2007.

Comparing injury crashes v. fatal crashes

NHTSA only compared injury analysis results with what’s known about fatalities in two instances: First, the percentage of operators v. passengers injured is almost identical to the fatality data and secondly, the percentage of male v. female injured is almost the same as the percentage of men to women fatalities. Being a woman, I can’t resist pointing out MIC reports women represent 23% of all who rode (front or back) on motorcycles last year. Iow, it appears women are under-represented as both operators and passengers in both injuries and fatalities—and thus may be safer operators and make male operators safer riders when we’re on the back.

NHTSA didn’t mention there’s other aspects that reveal injury crashes are almost identical to their more lethal counterparts: the days of the week; the time of day; the time of year; the percentage of single-vehicle v. multi-vehicle crashes; age.

Intoxicated riding

One very key—and crucial difference is the percentage of alcohol-involved crashes that result in  injured riders (10%) versus dead riders (28% were at or over the legal limit and another 8% had between a trace and 0.07% BAC in 2007). It is unknown why such an enormous difference would exist—and it could be important how the researchers determined alcohol-involvement in injury crashes. But the data raises the question, in my mind at least, whether a crash after drinking is more likely to end in death than injury.


According to MIC, the ≤30 age group increased about 8%. The existence of mandatory training requirements for young riders suggest of all age groups, more trained riders are likely to be found among these young riders. However, since 1998 but using NHTSA data, injuries in this age group increased 105% and fatalities increased almost 83%. In 2007, these young one-third of the riding population had almost 40% of the injury crashes.  The author of the analysis points out that other age groups are catching up to younger rider injury numbers.

While we are well aware that fatalities among the Baby Boomers have dramatically increased in the past decade, they do represent more than 50% of the riding population and in 2007 had just under 43% of the injuries and just over 49% of the fatalities.

Comparison between injury and death when it comes to helmet use deserves its own entry because of the implications.

There but for the grace…

The bottom line, though, is that NHTSA’s limited injury data analysis reminds of what death by motor vehicle really is—one or more injuries that are so severe that life cannot be sustained. So it’s not surprising that so many of the characteristics are exactly the same between injury and fatal crashes. both occur at the same times of days and in the same seasons to basically the same people.

If we look at only these few characteristics NHTSA chose to highlight, it suggests a “there but for the grace of God” scenario as if it random whether the rider was merely injured or injured so severely they died. And this would be very difficult for riders to accept. A further, more comprehensive analysis—and comparison must be done. Here’s an idea—how about doing the long-delayed accident causation study that the motorcycle industry was forced to finance but doesn’t want to be done?

Otoh, it suggests that most of these issues such as day and hour (with the possible exception of alcohol blood level) aren’t what causes one crash to be fatal and another to be survivable. The answer may lie elsewhere—and now we know where not to look.

More injurious today than a decade ago

This new document is what it is—a research note. It does raise some issues though: While percentages of crash-involved riders stay the same in specific areas the analysis reveals a significant gap in a critical area: registration went up 84% while injuries went up 110%—a gap of 26% and that’s a very significant change. Iow, it can’t be just a matter of more motorcycles=more crashes=more injuries.  Something else is going on—something that this analysis didn’t pop out.

More deadly today than a decade ago

Crashes aren’t just more injurious, they’re more deadly than a decade ago.  While injuries increased 110%, fatalities increased almost 125%. That’s a difference of almost 15%. (It also means that fatalities increased 41% over registrations compared to that 26% disparity between registrations and injuries).

Ten years of data reveals that crashing out of proportion to the increase in motorcycles on the road—and more of those crashes are ending in injury and more of those injuries are deadly.  That should concern us—and the more so after we look at the helmet issue in the next entry.

Distracted/cellphone-using driver risk isn’t the only thing NHTSA has covered up

July 23, 2009

According to an article in yesterday’s New York Times, “Driven to Distraction:  U.S. Withheld Data Showing Risks of Distracted Driving” NHTSA researchers in 2003 estimated that 6% of all daylight hours on USA roads were spent talking on the phone. Today, the Transportation Department estimates that figure has nearly doubled.

Cellphone use has been found by extensive research to be equivalent to driving drunk and have a 4x greater chance of crashing. Iow, up to 12% of all daylight hours on US roads are filled with drunk drivers.

The researchers estimated that 955 fatalities and 240,000 accidents in 2002 were cellphone-involved, and the  talking points memo said that NHTSA estimates 25% of crashes are caused by distracted driving.

The article quotes, Clarence Ditlow, director of the Center for Auto Safety, “We’re looking at a problem that could be as bad as drunk driving, and the government has covered it up.”

Otoh, how likely is it unlikely that 12 percent of all daylight drivers are over the legal blood alcohol limit? The problem may be far worse then than drunk driving.

Since more motorcyclists are killed in multi-vehicle crashes and distracted driving means drivers aren’t paying attention and visibility is one of the main reasons drivers cite for causing crashes with motorcyclists, this should be—but isn’t—a major issue with motorcycle rights activists with the notable exception of Bruce Arnold.

But it’s what else the article claims—that NHTSA deliberately withheld how dangerous cellphone use was from the American public—that led me to see a correlation with the languishing motorcycle accident causation study:

The  NHTSA researchers who investigated and reported on distracted driving prepared that talking points memo at the end of  a 266 page report that laid out all the research and evidence of the growing and lethal problem. That report and the memo was not released until six years later—and only because of the Freedom of Information Act request by the Center for Auto Safety.

As most if not all my readers know, the federal government set aside money for a new comprehensive motorcycle accident causation study to update the famous Hurt Study—and yet years later it’s dogged with delays and only recently has the tiny pilot study been launched that will look at less than 100 accidents.

So it was ironic, in a way, to discover that the NHTSA researchers “proposed a long-term study of 10,000 drivers to assess the safety risk posed by cellphone use behind the wheel. They sought the study based on evidence that such multitasking was a serious and growing threat on America’s roadways.”

Instead, NHTSA, under Dr. Jeffrey Runge, “rather than commissioning a study with 10,000 drivers, handled one involving 100 cars,” It’s starting to sound awfully familiar, isn’t it?

NHTSA did the same to motorcyclists

Actually, NHTSA went farther when it came to a danger to motorcyclists and other road users. In 1997, NHTSA produced a report, DOT HS 808 570 “Relationships between vehicle Size and Fatality Risk in Model Year 1985-93 Passenger Cars and Light Trucks”.
”[A] draft of the report was peer-reviewed by a panel of experts under the auspices of the Transportation Research Boardof the National Academy of Sciences” and then “revised in response to the panel’s recommendations.” Iow, the researchers knew the results were controversial and were making sure readers knew it had been vetted by the best of the best.  If you’re looking for the hot link to that report–keep on reading as it’s exactly the point.

NHTSA researchers studied the effects on just a 100 lb. decrease in weight for SUVs, pickups and other light trucks and found that in 1993—long before SUV sales took off and so did the motorcyclist death toll—that out of the 2,217 motorcyclists, pedestrians and bicyclists hit by light trucks in 1993, the fatality rate would’ve dropped by 2.03% or a net fatality change of -45 percent. In contrast, a similar reduction in the weight of passenger cars would’ve resulted in a change of – 0.46 percent or a net change of -19 percent.

They stated, “…downsizing of light trucks would significantly reduce harm to pedestrians, motorcyclists and, above all, passenger car occupants,” with a minimal effect on increasing rollovers. It went on to say, “The benefits of truck downsizing for pedestrians and car occupants could more than offset the fatality increase for light truck occupants.” And concluded, “Continued growth in the number and weight of light trucks is likely to increase the hazard in collisions between the trucks and smaller road users (cars, motorcyclists, bicyclists and pedestrians), while a reduction in the weight of the trucks is likely to reduce harm in such collisions.”

Iow, 12 years ago NHTSA found light truck vehicles (LTVs) are extremely dangerous to other road users and were so sure of it that they didn’t recommend a larger study but a reduction in weight of a mere 100 lbs. NHTSA did nothing about this.

Further research confirmed this finding. A 2002 Dynamic Research, Inc. study, “An Assessment Of The Effects Of Vehicle Weight On Fatality Risk In Model Year 1985-98 Passenger Cars And 1985-97 Light Trucks Volume I: Executive Summary DRI-TR-02-02” examined 1999 fatality statistics and confirmed the 1997 NHTSA document.

A year later, in 2003, Dr. Michelle J. White, professor of economics at the University of San Diego, published the paper, “The Arms Race” on American Roads: The Effect of SUV’s and Pickup Trucks on Traffic Safety”. In it she concluded, “For each one million light trucks that replace cars, between 34 and 93 additional car occupants, pedestrians, bicyclists or motorcyclists are killed per year and the value of the lives lost is between $242 and 652 million per year.”  She went on to say that for each fatal crash the occupants of light truck vehicles themselves avoid, “at least 4.3 additional fatal crashes involving other occur. Iow, the drivers of SUVs make others pay for their selfish self-interest at an unconsciously high price.

Her study found that if a light truck hits a motorcyclist, specifically, the probability of dying rose by 56 percent but the probability of only being seriously injured rose to 26%. While pedestrians/bicyclists’ probabilities also rose (45% fatality, 11% serious injury), motorcyclists, then, are particularly at risk from LTVs.

But just as NHTSA had ignored its own study, it continued to ignore further studies. And, rather than weight decreasing, the weight of  SUVs, at least, increased: For example, in 1993 Ford Explorer’s curb weight was 3, 679 lbs. In 2009, the Expedition weighs 5,578 lbs—a weight increase of 52%.

And NHTSA ignored easily accessible information such as the survey by Roy Morgan Research one of over 24,000 SUV drivers. While “[L]arge 4WD” owners were determined survey to be such things that don’t necessarily affect driving such as male SUV drivers are  more likely to be overweight and  more likely to prefer beer and femial SUV drivers are more materialistic and more likely to say, “I was born to shop.”

It also found that they were: more aggressive; less tolerant; more likely to suffer road rage; less charitable; more likely to use force to get their way—and much more importantly—more likely to be involved in accidents that kill or maim people in other vehicles.

It also ignored research, such as this study that found that “Evidence suggests that because [SUV owners] sit higher, drivers of SUVs (and vans and pickups) are less able to judge speed accurately.”

While SUVs bloated like a fat lady with PMS—with a huge financial boon to the American auto industry, the motorcyclist death toll soared—and particularly in terms of LTV collisions.

In 1994 (the earliest date available), FARS reports 376 fatal LTV/motorcycle crashes. In 2006, FARS reported 1,083 fatal LTV/motorcycle crashes—a 188% increase. Meanwhile, passenger car/motorcycle fatalities went from 595 to 943—a 58% increase.

And, in another parallel to the Distracted Driver research, the original NHTSA document disappeared from the Internet: In 2004 (when I found it on line and printed it out) this document could be easily found on the Internet, today, a Google search leads one to Summaries of Published Evaluation Reports–and it’s one of a handful of NHTSA documents that are not accessible through a hot link. Curiously, it’s the only one without a hot link that does not include the information it has been superceded by a later report. The url that worked in 2004 no longer works. It is, though, cited in numerous other papers. However, you can find the summary of the peer review—which criticizes the report before­ it was changed in response to the review—here

Van and SUV drivers more likely to be on the cell phone than other drivers

Not only that—and to tie it back into the NYT’s articles, NHTSA’s 2001 DOT HS 809 293 reveals it also knew that Van and SUV drivers were more likely (4.8) to use cell phones while driving than passenger car drivers (2.6). Interestingly, it also found that particularly female rather than male and rural rather than urban Van and SUV drivers were more likely to use their cellphones while driving than female or urban passenger car drivers. Equally interesting–it found pickup drivers were less likely to use cellphones (1.9).

Iow, NHTSA hasn’t just ignored the cellphone issue that literally impacts so many motorcyclists’ lives—it’s the LTV issue as well–and the LTV owner talking on his or her cellphone issue. So why would NHTSA ignore so much research in various ways that have lethal consequences to the most vulnerable of road users—motorcyclists, pedestrians and bicyclists?

Fear of fiscal retaliation

According to the NYT, NHTSA deep-sixed the report because of fears that stakeholders would be upset at the findings—and any subsequent laws against cellphone use that may result—and that would result in loss of appropriations.

“Those stakeholders, Dr. Runge said, were the House Appropriations Committee and groups that might influence it, notably voters who multitask while driving and, to a much smaller degree, the cellphone industry.”

The article went on to say, “Mr. Monk and Mike Goodman, a division head at the safety agency who led the research project, theorize that the agency might have felt pressure from the cellphone industry. Mr. Goodman said the industry frequently checked in with him about the project and his progress. (He said the industry knew about the research because he had worked with it to gather some data).”

“Can you hear me now?” Money talks–but we get the dead zone

That wouldn’t surprise us; powerful interests are powerful and there is nowhere in America that money speaks louder than in Washington. And the telecommunications sure knows how to talk Washingtonian: AT&T tops’s All-Time Donors List. It’s spent over $43 million in lobbying from 1989-2008. The Communications Workers of America is no. 12 and Verizon comes in at 32—just behind the AFL-CIO and beating out such heavy-hitters as FedEx, Lockheed Martin, General Electric and the NRA. The Cellular Telecom & Internet Association—which is the wireless/Internet industry’s MIC—spent $1,790,000 million last year alone on lobbying and spent an additional $395,000 on several other lobbying firms.

And then there’s the campaign contributions—AT&T donated almost $4.5 million, Verizon $2.5 million,  the National Cable & Telecommunications Assn $1.5 million, Qwest just over $1 million—and that’s just some of them and that’s just the contributions to federal Congressional candidates and doesn’t count Presidential elections (overall, the industry gave Obama almost a million, Clinton over half a million and poor John McCain just over a third of a million—but then Democratic candidates for Congress get, by far, much more money than Republican ones).

In a similar way, LTVs were already becoming a juggernaut in the economy back in 1997: LTV registration had doubled from 1985-1997, passenger car population had remained relatively stable but LTV registration had doubled. For example, in 1994 passenger cars were 67% and LTVs were 18%. of all registered vehicles.

In 2006, passenger car registration had dropped to 55% (-27%) and LTVs had become 31% of all vehicles registered (+73%). SUV registration, alone, had gone up 400% since 1994.

Meanwhile, the automobile industry was spending up to $71 million a year in lobbying alone and giving up to $21 million to candidates (in ’04—in ’08 it dropped to about $18 million).

Even if MRF and AMA were lobbying on these issues, there’s no way they could even come close in spending for political influence.

Vox populi are the ones doing the talking—and voting

But, as Runge said, there’s an awful lot of voters with cellphones in their hands. According to CTIA, 82.4% of Americans have some kind of cellphone plan. And it’s the voice of the people who are talking on the phones while driving.

And, of course, there’s those 74,797,241 LTV owners.  And that alone is a powerful voting block.  Put the two together–the chatty Van and SUV owners–who also tend to be married, adults 25-54 48% male and 52% female, college educated, professional/managerial and affluent (HHI $40k+). Iow, the kind of people who tend to give campaign donations.

It’s very believable, then, that pressure was exerted and fears were created and NHTSA succumbed.

The motorcycle industry has acted similarly when it comes to NHTSA research

That the motorcycle industry, in particular, influences NHTSA in a similar way is no stretch at all:

As we know, Tim Buche said that the new accident causation study would be done “over his dead body”. We also know the motorcycle manufacturers have put up most of the money for the new accident causation study—but that money came with stipulations and it’s unknown if we now what all of them are. But one of those stipulations we do know is that that no conclusions nor recommendations be drawn at the end of the study by those that do it.

Plausible deniability

The NYT’s article points out, “…[Goodman] could offer no proof of the industry’s influence. Mr. Flaherty said he was not contacted or influenced by the industry.”

Nor can anyone prove that the automobile industry pressured NHTSA nor can riders  prove that the motorcycle industry is preventing the accident causation study from moving forward and influencing the design of the study to protect its self-interest over our well-being.

But the facts remain: NHTSA has long known that more motorcyclists and other vulnerable road users would die if LTVs even remained at their 1993 weights let alone got heavier. And it knew that more people would die if cellphones were used while driving. And it knew that SUV and van drivers talking on the phone weren’t just equivalent to the ordinary drunk–they were extremely care-less and careless drivers driving every day at all hours as if they were four sheets to the wind drunk.  Iow, NHTSA has time and time again valued the lives of LTV owners over the lives of riders.

And why? So their appropriations are safe, which means their jobs are safe while riders suffer and die.

And for what? So they can do a little good? Well, it seems like they’re doing precious little good. But there are always those who don’t care what damage they do as long as their self-interest is served. That shouldn’t surprise us either. Then again, it doesn’t mean we should continue to let those more concerned with their jobs than our lives continue to ignore what raises our risk.

NHTSA responded by dedicating itself to safety—not the safety of vulnerable road users—and specifically motorcyclists—but the safety of those who spend the most lobbying and give the most to campaigns.

MSF licensing tests screen out only the bottom 20%–and the BRC doesn’t even do that

April 28, 2009

Jimmy Jon sent me a comment today that read in part, “Have you ridden the North Carolina “skills test” or “road test”? If you haven’t, take notice that it is administered in a parking lot. Take further notice that this test does not involve either moving or parked traffic while it is administered in a parking lot. Further notice that the test is successfully completed by negotiating the through a cone pattern. Please also note that the testee provides the motorcycle, and this testee is not limited to this motorcycle in the future.”

I responded that it sounded like the Alt-MOST. I thought that perhaps people could use a little background on MSF’s licensing products:

In 1978 MSF published the first in a series of licensing tests, the Motorcycle Operator’s Skill Test (MOST). Shortly thereafter, it came out with an on-road/in-traffic one, the Motorcyclist In Traffic (MIT) test and the Motorcycle Licensing Skills Test (MILT). MSF also trained license-examiners across the country.

It was anticipated by NHTSA—and state DMVs—that the test would be taken on the motorcycle the applicant would be riding on the street. The MOST and MILT tests were so difficult that crashes—some even with broken bones occurred during the testing.

Even so, according to an update published in Safe Cycling, MSF’s then licensing director, Carl Spurgeon, wrote in the Spring 1988 edition, “When testing is administered by a state licensing examiner…only basic skills and abilities are evaluated. To be blunt, these tests screen out the bottom 20 percent or so and send the rest on their way with a license.”

Only the bottom 20% is very discouraging—and it’s even more discouraging that MSF is now paying PIRE to show that the end-of-course-evaluations measure up to the tests at the DMV.

Despite what Jimmy Jon wrote, most of those who go to the DMV take the test on their own bike, it’s more difficult to pass than when taken on a small training bike on a range that the students have been practicing on for two days. Iow, the course may be allowing many of that 20% who would’ve flunked at the DMV to pass in the course.

Jimmy Jon went on to write, “A successful test should be conducted upon a public road, with traffic, and administered by a certified motorcycle rider/tester. The rider must utilise for the test the motorcycle that he will be licensed to drive on the road.”

While I agree only a road test is sufficient to show whether a rider is traffic-ready, I’m not sure that instructors/examiners who have only had MSF-training are any more adequate to the task than the end-of-course evaluations or the current set of MSF tests now used in DMVs across the country.