Actually, the US number is less.
From the article: "In a 2003 study the University of Cologne estimated that nearly 1000 suicides in Germany are carried out each year in a similar manner, whereby the person committing suicide intentionally walks or jumps into the path of an oncoming train."
For comparison, go to safetydata.fra.dot.gov/officeofsafety/
This should give you a 161 page pdf document that basically folds, spindles, and mutilates statistics of accidents/incidents on or near any railroad in the US. It gives data by category, state, railroad company, and probably a few other ways. It does not list suicides as such. I have only begun to look at it.
Total fatalities all types for each year since 1993 vary between 1,279 down to 867 per year, with the most recent year in the table being 898 fatalaties in 2004. Again, this is ALL TYPES.
If we get it down to highway-rail and trespasser accidents, this should include all suicides in addition to general stupidity, carelessness, and being in the worng place at the wrong time. Within the same 1993 to 2004 time frame, we have
Traffic accident fatalities
Worst year, 1993, 626 deaths
Best year, 2003, 334 deaths
For 2004, 368 deaths
Tresspasser fatalities - again, not all suicides, but who knows percentage
Worst year, 2002, 540 deaths
Best year, 1996, 471 deaths
For 2004, 482 deaths
On train passenger fatalities - not just Amtrak, but all
Worst year, 1993, 49 deaths
Best year, 1995 0 deaths
For 2004, 3 deaths
That we should be in the same ballpark even though having a much higher population is not really surprising, as Germany has a lot of high frequency predicitable passenger trains. Outside areas with frequent commuter service, you might have to be on the track for hours before getting hit. Our heaviest freight volumes run a lot of their miles in very low population density areas.
"a dozen or more suicides-by-train involving Amtrak each month" seems highly unlikely given that the total all fatalities all causes all railroads averages around 80 per month of which around 30 are traffic fatalities and 40 are tresspassers. In the statistics "by railroad company" I am not sure wither this is by track owner or train operator. All fatalities all causes ascribed to Amtrak are in the 118 to 131 per year range, so that gives you about 10 per month, not all of which are suicides.
I am not trying to say it is not a problem, but let's not exaggerate. When your understanding of the problem is skewed or just plain wrong, thin your solutions will probably also be wrong.
Also, understatd that all road fatalities are on the order of about 40,000 per year, so getting too worked up about the 1,000 per year on the railroad, which is only 2.5% of that is an over reaction.
Doing things that make railroad operations more expensive and passenger service less available under the heading of safety is actually counterproductive to safety, as it drives people to the much more dangerous highways.
George