However, like anything, the safety-measure pendulum can swing too far in the other direction, and when it does it can become a safety hazard in of itself.
When I am doing fieldwork, I am required to have a minimum amount of safety gear. At minimum, I wear long sleeved shirts and long pants, as well as boots and usually an orange safety vest (good for being seen by vehicle and equipment operators in urban and construction environments, and in rural environments it also allows crew members to keep track of each other and notifies hunters that we are not strange, clothed, flesh-colored deer). When working around heavy equipment, this is usually supplemented by safety glasses, a hard hat, steel-toed boots (oh so comfortable), and hearing protection. Usually the requirements for the higher level of PPE (personal protective equipment) makes perfect sense, and we're happy to have it.
Sometimes, however, the higher PPE requirements reflect the draconian application of a policy to a situation to which no sensible person would try to apply it.
For example, a couple of years ago I worked in the oil fields of Kern County. We were required to have the full deal - steel-toed boots, hard hat, safety glasses, etc. When we were in active oil production fields, this was sensible - there's enough heavy equipment and vehicle traffic to make the safety equipment a wise investment in such areas. For much of the project, though, we were on land that was technically part of the oil fields, but which had no pumps, no vehicles, no equipment, nothing at all that could blow up, pop out, fly away, or drop on our heads. Nothing that necessitated the additional safety equipment. Now, this would have been annoying, but nothing more, if you didn't take into account the fact that we were working there during the summer, when temperatures were normally over 110 degrees Fahrenheit, and we were being required to lug around extra equipment that not only weighed us down, but also restricted air movement in and through our clothing and prevented us from wearing other safety equipment made for hot weather (such as some special-made vests and hats that stored and released water) that would have helped us lower our body temperatures. The safety equipment had become a safety hazard.
Another case comes to mind, one where I was not in the field, but did work for the company that had the problem. My previous employer, a large multi-national company that provided a wide range of environmental and engineering services, had a contract with a petroleum company to provide environmental services in their fields in southern San Luis Obispo County. During fieldwork, one of the wildlife biologists was bit by a soft-bodied tick. Soft-bodied ticks are particularly nasty, and severe adverse reactions to their bites are not uncommon, and the biologist ended up in the hospital (they fully recovered, by the way).
The client was outraged. They demanded to know how this could have happened, and were unwilling to accept that fieldwork, by its very nature, has risks associated with it. One of these risks is the possibility of being injured by wildlife. I left the company before the matter was resolved, but last that I had heard, the client was demanding that all field workers be dressed in head-to-toe protective garments that would prevent ticks from being able to enter people's clothing. The problem, of course, is that the garment in question would also diminish the wearer's body's ability to cool off via sweat, which would greatly increase the chance of heat injury, including the very deadly condition of heat stroke. So, the oil company executives were so obsessed with preventing a low-probability event that had occurred once (the tick bite) that they were willing to require measures that created an even greater danger.
Attempts to explain this to the petroleum company executives fell on deaf ears, with one of them even going so far as to say:
"It's possible that deadly snake could crawl into a crate of merchandise packaged in Africa, and then get out in the U.S., slide off of a train, and find its way to the field and bite one of your biologists. You have to protect against even the very small possibility of that happening! These garments will keep out ticks, and will even stand up to the snakes! We make our workers in Africa wear them for that very reason!"
The executive then ignored everyone who pointed out that in order to prevent the highly improbable (in fact, I'd go so far as to say practically impossible) snake, he was greatly increasing the chances of the very likely death of someone due to heat stroke. It was a classic case of the unlikely even that happened eclipsing the very likely event that was avoided - a classic error that leads to extremely poor risk management*.
On more example comes from the same employer. The upper management decided that they wished to improve our company's safety record, and to that end, they assigned a team to develop safety standards that could be followed by all company personnel. Now, keep in mind that this was a big, multinational company that provided a wide variety of services (yes, I worked for Veridian Dynamics), and so a wise set of safety standards would allow flexibility for the people performing different types of tasks under different conditions.
Needless to say, we did not receive a wise set of safety standards. The problem with the new standards can best be summed up by the recommended use of a ladder for crossing over all walls, fences, and other obstructions 3' tall or higher. Recommending the use of a ladder for such purposes is fine, if perhaps a bit overzealous, when one is working within a small area where the ladders can be put in place at the beginning of the work day and left in place throughout the day. When one is performing archaeological or biological survey, though, it becomes wildly impractical. The usual method for dealing with fences and walls when performing survey is to crawl under them or climb through them (in the case of barbed wire) or climb over them (in the case of sturdy wood, metal, concrete, or stone walls and fences). If done correctly, this is just as safe as climbing up and down a ladder (remember, with a ladder, you're increasing the maximum height a person can fall, therefore potentially actually triggering yet another set of OSHA requirements). When there's an unlocked or open gate nearby, we obviously just go through that, but it's not uncommon for the nearest gate to be a mile or more away from where you are. Add to this that carrying a ladder means lugging another heavy object along with your other field equipment, and that ladders tend to get caught on plants, making them hazards to try to carry through dense vegetation - which we routinely move through. The simple fact of the matter is that carrying a ladder on survey actually creates a few new risks, while doing little to eliminate existing ones.
When the use of the ladder during survey was put to a room of archaeologists and wildlife biologists, we all looked at each other, and then at the presenter. We explained the basic problem, and had to spend the next hour describing out work in detail so that she could understand why this was a lousy way to improve safety. At least she listened, but had we actually been consulted about our work before the new safety standards were developed in the first place, more practical and useful standards might have been created.
The simple fact of the matter is that field work does have risks. These risks can be minimized, however, through the appropriate use of PPE, and through the application of both training and some basic common sense. There is always going to be the possibility of injuries, though, no matter how hard we try to prevent them. The problem is that very often we are required to work with rules written by people who don't understand the nature of our work, and therefore don't understand that what makes someone safer in one environment may actually put them in danger in another.
*A more common example: how many times have you heard someone say that they won't use a seat belt because they know of someone who was thrown free from a collision and therefore survived something that smashed the vehicle? You may have noticed that showing someone that the statistical likelihood of surviving a horrible collision because you wore your seat belt is orders of magnitude higher than the odds of you surviving by being thrown free from an accident never seems to sway them. The weird exception will often trump the quieter reality in people's minds. Which is, incidentally, the primary reason why we as a species are doomed.