Pennsylvania, A Tale of One State: The Helmet Effect

Once again and before we start, here’s my position on helmets: I wear, by choice, a full-face helmet and believe should be a matter of personal choice. I also will not write one word to help repeal or reinstitute a helmet law. The following, then, is not in support in one position or another and should not be taken as such.

The repeal of the helmet law would seem to be a likely culprit to the 142.7% increase in motorcycle fatalities—so let’s see what effect it really had:

The bill was signed into law in 2003, and that year, the state motorcycle fatality rate increased by 16.9%. Which seems to prove everything advocates say about helmet laws.

However, in 2003, only 3 more unhelmeted riders were killed in 2003 than in 2002—or a 10% increase. That might surprise you that I use the words “more” and “increase”—but, as you’ll see in the chart below, riders were killed without helmets before the law was passed. So, if there’s a helmet effect, there was already a small degree of that operating in Pennsylvania.

Three additional unhelmeted deaths, though, didn’t really affect the fatality rate in Pennsylvania in 2003. Some would suggest that there were only three because the bill was signed in September and it was almost the end of the riding season. But that’s to minimize that huge jump in the fatality rate—and it was among helmeted riders who had a 19.38% increase—or almost twice the increase than among unhelmeted fatalities.

PA unhelmeted and helmetd deaths

Then came 2004—the first full season where riders could go helmetless as long as they had either two years of experience or passed a training course.

In 2004, the number of unhelmeted deaths increased 145% which is a huge immediate increase. It translates into 48 more riders than the previous year died sans lid. It seems to be a big Aha! Moment for the helmet people.

Logically, some increase would be expected simply because so many more riders were going helmetless. The question is whether helmetless riders were contributing more than their proportion of the riding population. Research claims that helmet use decreased to 58% in Pennsylvania—and in 2004, unhelmeted riders were 49.7% of the fatalities—so they were contributing more than their share of fatalities.

What’s odd though is that overall fatalities only went up 2.5%—only four more riders total were killed in 2004 than in 2003. Hardly the bloodbath on the beltway we were led to believe would happen.

In actuality, 48 more riders died while not wearing a helmet but 35 fewer helmeted riders died.

But that’s not all that was extraordinary about 2004 in Pennsylvania: It was also the year of the single largest increase in motorcycle registrations with over 41,000 additional motorcycles being registered that year. The figure is all the more extraordinary because the prior year saw an increase of just over 1,200 bikes and the following year, 2005, the increase was just over 13,600 motorcycles. So 2004 was remarkable for more than just no more helmet head.

Iow, there was a full year of helmetless riding when tens of thousands of motorcycles  and 18,699 riders took the basic course and were added to the road mix. It should’ve been the year of living dangerously and instead fatalities only went up 4 riders—2.5%—over the previous year.

And while some years experienced a decline in fatalities, 2004 was the smallest increase in the timeframe from 1998-2007.

The helmet law had an effect but it didn’t necessarily mean that more riders died than had the helmet law not been repealed. And it’s also suggestive that the relationship between increased motorcycle registration and fatalities is not so simple or clear-cut as we have been led to believe.

2005 was stranger still: unhelmeted fatalities increased by 12.34% (10) but helmeted fatalities increased 47.5% (39) and overall fatalities increased by 30% while motorcycle registrations only increased by 4.7%.

And an even stranger thing happened the next year: In 2006, motorcycle registrations went up 8.2% but overall fatalities experienced a decrease of 11.3% However, helmeted riders saw a bigger decrease than unhelmeted ones.

Only when the helmet law was five years old in 2007 did the statistics match up to the picture we have been led to believe is the norm: motorcycle registrations went up 7% and 19 more unhelmeted riders than helmeted ones died and the overall fatality rate increased by 23.9%. Iow, only 2007 shows the kind of correlations we have been led to believe is the norm.

PA regs v. unhelmeted fatalities

The relationships between helmet use and fatalities, motorcycle registration and fatalities does not seem to be as simple or straight forward as we’ve been led to believe. None of this is to suggest that riders who may have lived die instead because they weren’t wearing a helmet.

Otoh, it seems to strengthen the belief that repealing universal helmet laws increases motorcycle sales–at least temporarily.

Training to ride helmetless

There appears to be another effect, however—training, which is believed to lower the risk of riding, appears to be correlated with non-helmet use—which is believed to increase that risk.

Training had been generally increasing since 1997 with the biggest jump in 2000 with an 11% increase—but then training actually decreased 5.2% in 2001 and stayed almost exactly the same in 2002.

Training numbers increased as the law moved through the legislature and to the Governor’s desk and continued to grow. Training grew 28% in 2003, another 11% in 2004, stayed at that level in 2005 and increased another 16% in 2006 even as motorcycle registration growth slowed down. Training decreased in 2007.

This suggests that the significant increase in training appears to be correlated with riders who went out and bought a bike because they now could ride without a helmet if they got training. But did training act as the antidote to riding without a helmet?

PA training v. unhelmetd fatalities

Now that we’re more than two years past the helmet law repeal the only way someone can begin to ride without a helmet is by taking training. So any new (≤2 years) unhelmeted rider fatality has gone through training—if they are licensed, that is. It’s unknown how many of the fatalities–trained or untrained–were trained or untrained. The relationship between training and fatalities and training and unhelmeted fatalities and what it might say about the quality and effectiveness of training should be explored.

56% (95,618) of all the students have been trained and 56% (955) of the fatalities have all occurred since the helmet law was passed. It’s unlikely that all riders who died were trained. But it’s equally unlikely that they were all untrained.

The increase in motorcycle registrations since the helmet law has been repealed is 61% (103,983).

It also should be noted that the motorcycle manufacturers not only control the course curriculum and licensing standards but the administration of the training program in Pennsylvania and benefit financially from the increased sale of motorcycles and accessories.

The question is whether other states show similiar evocative patterns. And we’ll get to that soon.

Explore posts in the same categories: Motorcycle helmet use, Motorcycle Industry, Motorcycle legislation, Motorcycle Rights, Motorcycle Safety, Motorcycle Safety Foundation, Motorcycle Training, State Motorcycle Safety Programs, Uncategorized

9 Comments on “Pennsylvania, A Tale of One State: The Helmet Effect”

  1. aidanspa Says:

    Nice work. Just as total annual motorcycle fatality numbers don’t mean much except as a percentage of the total number of (all) motor vehicle fatalities, I think there are some numbers worth mentioning here. Your chart representing “PA Unhelmeted and Helmeted Fatalities” gives us the following number of unhelmeted fatalities as a percentage of all PA m/c fatalities:

    Helmet Law:
    2001 – 21%
    2002 – 23%
    2003 – 22%

    No Helmet Law:
    2004 – 49%
    2005 – 42%
    2006 – 47%
    2007 – 54%

    These are the figures that matter, IMHO, regarding PA unhelmeted fatalities.

  2. Bob Says:

    first question… in the years before the helmet law was changed, were the non helmet wearing fatalities on or off road? Do you have a break down of age over this period for the non helmeted fatalities?

  3. wmoon Says:

    Well, I think you’re missing the point–putting all the deaths into context, it suggests that it’s what the late Karen Bolin said–it’s not what’s on your head that saves you, it’s what’s under the helmet. And the point is: training does not seem to be saving these riders.

  4. wmoon Says:

    Off-road fatalities, to the best of my knowledge, are never included in the fatalities.

  5. Bob Says:

    I took a look at the FARS search tool…. no ability to select for off road scenarios.

    From other postings I’ve read, it seems to me that a helmet is going to prevent a fatality in some instances that having no helmet would not. But it’s like every other piece of safety equitpment…. nowhere near infallible.

    What would be interesting in this evaluation would be to make that calculation…. how many ‘extra’ deaths occured due to the lack of helmet? Many of the fatalities would have occured regardless of use…..

    When the law was in effect, by the calculations above, there were still 20% of the fatalities sans helmet…….

    Is it reasonable to assume that you can’t ‘cure’ that percentage?

    The population you can effect then becomes 80% of the fatalities in a given year. Out of that, how many could be projected to still be alive, if they had chosen to wear a lid? I’ve read elsewhere that a helmet gives you a higher likelihood of surviving a crash…. 150% of not, sound right?

    What’s the ‘true’ delta? Year on year, through this span, the number of fatalities increases (01-02 call that a steady state)

  6. wmoon Says:

    Bob, as I told you, FARS doesn’t include off-road fatalities. See: NHTSA’s latest study finds that helmets are 37% effective in reducing fatalities and research presented at the NTSB forum found that helmets are 25% effective at reducing injuries.

    You are basing way, way too much on just one state and a limited amount of years and you’re assuming way too much of the helmet’s ability to protect a rider–and assuming that lowering head trauma will automatically raise the survival rate. That is another false assumption you’re making. In a great many cases there’s chest trauma or abdominal trauma and amputation (and the consequent bleeding out) or severing of the femoral artery (and therefore bleeding out) is also more common than you might expect and all of those injuries can be equal to or greater than the head trauma.

    As for the delta–I thought that’s what you were going to figure out with your 18 state comparison.

  7. Bob Says:

    This is how PA plots out against the National fatality data over a much longer period than what you are looking at in this particular discussion. almost 30 years, vice 7.

    Seems to me, given the rough parallel between the national number year on year and the individual state numbers, you are working with the ‘best fit’ data for the dissertation.

    Why not include an equal number of years before and after the change in helmet law? That bump in 2000 seems at least as significant as ‘wow’ in ’05-06…..what happened, there? If you are looking for trend information, seems to me the interval needs to be much longer…..

    I (for the record) do not claim that helmets make a large difference in survivability. I understand quite clearly that all the other factors that cause death are not addressed by this law change. Cycle World had a really good dissertation on this a year of so ago. Age of the rider has a lot to do with it, as well. Older folk don’t survive trauma as readily as youngsters. they had that all the way down to a points system, as I recall.

    There’s very little question in my mind that the 37% you quote may be alive, had they worn helmets. My question is, what would that number be, of the fatalties year on year in PA? There’s no question that head trauma alone is rarely the cause of death in a MC accident of any type. But how do you calculate the difference from what has happened, and what might have happened?

    As knowledgable as you are on all of this, I’m sure you are aware of the articles I am mentioning.

    Your discussion seems to indicate that there is significant delta in fatalities in PA, as a result of the change in law. I can’t say I can ‘see’ that as clearly as you seem to be indicating.

    The ’93-94 ‘bump’ and the ‘bump’ in 2000 would appear to be at least as significant as the ‘dip’ in ’06.

  8. wmoon Says:

    Like usual, you’re confusing me. I’m not sure what you mean half the time. As you should be aware, I was looking at 97-07 because that’s the length of time that the SMSA data covered–therefore it’s the span of time for mc regs, unhelmeted fatalities, etc. Sure it would be great to have more years of data–but it’s not there unless states make it available to me.

    If you’re interested in how many unhelmeted fatalities back five years from 2003, you can, as you well know, go to the FARS data. But here you are: 2003: 33 2002:30 2001 27 2000: 29 1999: 23 1998: 24. As to the fatality numbers, they’re already in other graphs–and you have them in your data. I have no idea what happened in 2000. And if you looked at the stats you collected, you’d see sharp increases and decreases in fatality rates in a given year, which is why you have to look at trends. I read absolutely no significance into a dip or rise in any one year, myself.

    I have no intent to calculate the difference from what has happened or what might have happened. I’m not even drawing any conclusions (except the effect of the helmet law repeal on a temporary increase in motorcycle regs which is something the anti-helmet law people have always said is true). I have absolutely no idea if any of this has any meaning at all. I pointed out that the change after the helmet law was repealed is not what we’ve been led to expect in terms of all the states that repealed laws. It’s always presented as a rise in fatalities (which it is) but the assumption (since the numbers/percentages of helmeted and unhelmeted) is that all the increase was in unhelmeted whereas it appears that is not necessarily true. I do think it’s interesting that there’s a relationship between a law that requires training so a rider doesn’t have to wear a helmet passing and the growth in training numbers–and the growth in both helmeted and unhelmeted fatalities.

    However, as I said over an over, we have no idea how many of the dead had been trained.

  9. aidanspa Says:

    W. –

    I am a little confused by something you said in your last post to Bob.

    “I pointed out that the change after the helmet law was repealed is not what we’ve been led to expect in terms of all the states that repealed laws. It’s always presented as a rise in fatalities (which it is) but the assumption (since the numbers/percentages of helmeted and unhelmeted) is that all the increase was in unhelmeted whereas it appears that is not necessarily true.”

    Looking at your first graph, “PA Unhelmeted and Helmeted Fatalities”, it appears to me that for the years 2004-2007 the increase in total fatalities was due entirely to unhelmeted riders. Before the repeal, unhelmeted deaths accounted for roughly 20% of the fatalities, and following the repeal, nearly 50%. The number of helmeted fatalities appears to have remained somewhat constant. Helmeted fatalities in 2001 & 2002 = ~100, again in 2006 & 2007 = ~100.

    It is very possible that I am misinterpreting the data, so I would appreciate your thoughts.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: