Right now I'm worth more dead than alive.
Before I looked at e-bike economy, I hunted for research on bicycle safety. It was interesting. One site which referenced the most research reports (can't remember where exactly) went through a lot of the numbers, but the guy was very dismissive of them. The biggest problem with trying to come up with accurate statistics is figuring out how much time / miles are actually ridden on a bike. There are other complicating factors that make it easy for a researcher to pick and choose reference data. I have yet to see an apples to apples comparison that restricts the data to the typical driving that would be encountered on a bike commute: Daytime driving, city conditions, etc.
I wish I could remember the website so I could post it here for all to see.
Overall, I was rather put off by the tone of the web site, since the guy was blatantly biased towards the bike being very safe (his attitude was if you know what you're doing you'll never get hurt); however, the raw numbers he quoted from government and insurance studies gives you the opportunity to look through the bias and come to your own conclusions. The statistics he quoted showed you have less risk of death PER HOUR on a bike than in a car (by about half). This is the number he was trumpeting about. The overall picture puts you at a higher risk of death PER MILE on a bike, since the average travelled speed of the cars was higher. Most of the sources for numbers were either insurance or governmental studies, but the difficulties listed in the previous paragraph make the risk factors vulnerable to the assumptions used.
As you say, safety is clearly a significant consideration in deciding to e-bike. There are also many other intangibles to deciding to bike or e-bike for a commute.
My company is planning on moving to a new building within the next year, and I have planned out a route (google maps is your friend) on primarily residential roads that are generally parallel to the main traffic routes through the city. The neighborhoods I will be going through mostly follow urban design philosophies popular since the early 70's where residential collectors are fed from housing clusters. Major intersections are light controlled, medium intersections are controlled by 4-way stops, and minor roads entering collector routes have stop signs at the intersection. All this works to my advantage, because most commuters move from the residential collectors on to regional (city-wide) routes fairly quickly, leaving the residential collectors with a lower traffic count. A few pedestrian/bike connects between collectors makes the route good for me and bad for cars.
I have biked into work and home from our current location, and there isn't as good of a route. I do get to avoid the worst traffic area by using an untrafficked levee road for about 1 mile, but the rest of the route is on divided 4-lane and 6 lane city-wide routes with a lot of signal lights. Even with those problems there area only a few areas on the way in that make me a bit uncomfortable. The trip home leaves me a bit more concerned as I have to turn left across traffic at some point. Once the company moves, I think I will probably bike to work and back for a few weeks in the fall to get a sense for how heavy traffic is on my planned route. 5 3/4 miles each way should be pretty easy in the right weather. If it is comfortable, I'll probably go for a kit in the spring. My interest in an e-bike is to keep me from working hard in the 100+ summer temps we can get around here.
One thing the research was pretty clear about: riding at night, without lights, on the wrong side of the road, while drunk is hazardous to one's health.
-Mike
