Question on safety stats

Please don't regard this as a troll. I have been diving for over 20 years, consider the sport safe when training and common sense are used, and hope to introduce the port to my pre-teen son next summer (okay he's already introduced and I'm waiting to see if he's ready to be certified).

In any case, my question/comment is this: I frequently see on this board posters reference the comparative saftey of scuba to things like flying, driving a car, and using a vending machine (not sure about the vending machine one but we all hear about the number of people killed by vending machines falling on them). I assume that the data given in these references is in absolute numbers, as opposed to % of the participating population. Is that the case and if so, isn't it an invalid comparison? Of couse a lot more people are killed in car accidents on their way to vacation then die diving on that vacation, but a lot more people take vacations than dive, right?

I'd just like to have a little context when looking at those statements.

robh:

Please don't regard this as a troll. I have been diving for over 20 years, consider the sport safe when training and common sense are used, and hope to introduce the port to my pre-teen son next summer (okay he's already introduced and I'm waiting to see if he's ready to be certified).

In any case, my question/comment is this: I frequently see on this board posters reference the comparative saftey of scuba to things like flying, driving a car, and using a vending machine (not sure about the vending machine one but we all hear about the number of people killed by vending machines falling on them). I assume that the data given in these references is in absolute numbers, as opposed to % of the participating population. Is that the case and if so, isn't it an invalid comparison? Of couse a lot more people are killed in car accidents on their way to vacation then die diving on that vacation, but a lot more people take vacations than dive, right?

I'd just like to have a little context when looking at those statements.

Click to expand...

DAN produces annual reports regarding injuries and fatalities. These may be viewed here if you are a DAN member, which you should be under any circumstances:
http://www.diversalertnetwork.org/medical/report/index.asp

There are numerous challenges to both internal and external validity in these studies, as DAN themselves will admit; some accidents are not reported, and some fatalities are misreported - e.g. many dive fatalities look like heart attacks. If these occur on the golf course, they are "heart attacks". If they occur at depth, they are "dive fatalities". On the other hand, to complicate matters, emerging issues suggest that many heart attacks at depth may have causes that involve other factors, such as high concentrations of pollutants in breathing gas, which brings about distress and failures.

Either way, DAN's numbers are about the best you'll find. Enjoy,

Doc

I always find the stats incredibly amusing. I'm not saying that I don't believe the stats or the conclusions they draw. I just get a kick out of them. For example, my big thing getting into the sport was about sharks. Being a northern california resident, the water I would tend to dive in is "habitated" by the big scary shark. When I told my instructor of my "irrational" fears, he mentioned something about being more likely to be killed in a car, in a plane crash or even a falling vending machine than seeing the big scary shark. The thing that came to my mind was, the chances of me getting killed by a vending machine while under water is pretty darned slim.

Like you said, context to the stats would be interesting.

robh:

Please don't regard this as a troll. I have been diving for over 20 years, consider the sport safe when training and common sense are used, and hope to introduce the port to my pre-teen son next summer (okay he's already introduced and I'm waiting to see if he's ready to be certified).

In any case, my question/comment is this: I frequently see on this board posters reference the comparative saftey of scuba to things like flying, driving a car, and using a vending machine (not sure about the vending machine one but we all hear about the number of people killed by vending machines falling on them). I assume that the data given in these references is in absolute numbers, as opposed to % of the participating population. Is that the case and if so, isn't it an invalid comparison? Of couse a lot more people are killed in car accidents on their way to vacation then die diving on that vacation, but a lot more people take vacations than dive, right?

I'd just like to have a little context when looking at those statements.

Click to expand...

What really made me think about this is a quote in another thread where someone talked about doing body recovery and rescue and how small the percentage of divers was versus auto accidents, plane crashes, boating accidents. I'd hazard a guess that alot more people drive cars, fly in planes or ride in boats than dive.

Adobo:

The thing that came to my mind was, the chances of me getting killed by a vending machine while under water are pretty darned slim.

Like you said, context to the stats would be interesting.

Click to expand...

Ya got a point there, Judge!! :cwmddd:

As Mark Twain (or Benjamin Disraeli, depending on where you look) so famously said: "There are lies, damn lies, and statistics." It all depends on the context, which so often is lacking in this type of abstract of numbers. DAN's numbers are better than any other source, but still are greatly lacking, and likely always will be.

Would love to see some per capita numbers, but of course, that would require an accurate number of divers to begin with. ARGH! Took a year of sadistics...er, statistics... in college, and they still drive me bonkers.

robh:

Please don't regard this as a troll. I have been diving for over 20 years, consider the sport safe when training and common sense are used, and hope to introduce the port to my pre-teen son next summer (okay he's already introduced and I'm waiting to see if he's ready to be certified).

In any case, my question/comment is this: I frequently see on this board posters reference the comparative saftey of scuba to things like flying, driving a car, and using a vending machine (not sure about the vending machine one but we all hear about the number of people killed by vending machines falling on them). I assume that the data given in these references is in absolute numbers, as opposed to % of the participating population. Is that the case and if so, isn't it an invalid comparison? Of couse a lot more people are killed in car accidents on their way to vacation then die diving on that vacation, but a lot more people take vacations than dive, right?

I'd just like to have a little context when looking at those statements.

Click to expand...

As others have posted there really isn't any fully accurate database of diving accidents/incidents. DAN takes as good a shot at producing one as anyone. But, lacking mandatory accident/incident reporting and analysis there just can't be an accurate accounting.

On the other hand there are many folks who, for whatever reason, greatly exagerate the dangers of scuba diving. There seems to be a like number who exagerate the safety of scuba diving.

So, all a person can do is take DAN data, a bit of personal experience and make a personal decision whether the perceived risks are worth it.

As for context: Since the database is not complete scuba accident/incident numbers are just not comparable to much of anything else. Accident/incident reports do provide some food for thought about personal practices; but that is about it.

.

Getting good statistics on diving accidents has three major issues.

First not all accidents are treated, let alone reported so it's not clear how many accidents occur in the diving population.

Second, there is no clear data on that population as no one knows for sure how many active divers there are, how many of these actually dive on a regular basis, how many dives they make per year, or how long those dives are on average. So it's hard to get a picture of the exposure to risk that is involved.

Finally, there are other variables that need to be considered before you could apply those general statistics to a more specific case including such things as experience level, equipment configuration, the type of diving done (deep, deco, rec, wreck, cave, etc) specific diving conditions that may increase or decrease risk, etc.

All three issues can be addressed to a great extent with properly designed and conducted survey research. To do that you have to get a suitably random sample of all divers which is very difficult to do given that even training agencies do not have current addresses for many of the currently certified divers. Another option is to collect survey data from a representative sample of dive sites or dive shops, but the problem here is getting a truly representative sample of sites/shops which is further compounded by getting suitably cooperative, unbiased and trained shops to properly survey a random number of divers coming through the shops. In either case, if the sample is skewed or the divers are not totally honest in completing the survey, the data will be in error.

Mathmatically, statistics is really just probability calculations and the camparing of different distributions to each other. The trick the sampling of the data and the way the math is applied.

I think comparisons like the chances of getting killed in a car to that of getting killed diving are of little or no value. My favorite is the famous PADI chart that shows that diving is about as dangerous as bowling using injury rates (I guess). When was the last time you heard of a fatal bowling accident? And I can tell you that I've had to clear the way for ambulances many more times at quarries than I have in bowling alleys.

Besides, aren't there good drivers and bad...good divers and bad? Is the likelyhood of injury the same for each?

Most of what we have for diving is more so raw data than it is statistics. Even if we can count the number of accidents we don't know how many active divers there are or how many dives are conducted.

the best use I've found for documents like the DAN report is it gives a little bit of a picture of "who" the divers are who are getting hurt. A large percentage are divers with little training and/or little recent experience. Buoyancy control problems are reported in a large percentage of dives that result in injury. Those points and a few others tell me that the less skilled divers are in a higher risk group. There's a surprise huh that there may be a correlation between a divers skill and the likelyhood that they will be injured? We still don't know how likely though because we don't know how much time they're spending in the water.

Another thing that I think should be useful, although the industry apparantly disagrees is that, I at least, see a correlation between the skill related issues reported by DAN and the way training is done. how many divers comming out of entry level training complain of still have buoyancy control problems? How often are those same issues associated with injury in the DAN report? Now, when you shop for training are there any questions you might want to ask? If nothing else the DAN report tells us a little about what an injured diver looks like and I'll bet you're doing yourself a favor if you go out of your way not to look like them.

Simplistically:
numerator (the top number in a fraction)
denominator (the number on the bottom)

DAN representatives I’ve talked with readily agree that having a denominator is key. For most of their data gathering history, DAN was limited to collecting the numerator (numbers of DCI “hits”, deaths, etc.). Early attempts to get a denominator, like number of tank fills, were not very successful. Under PDE, DAN looks at a known slice of diving, and now has a denominator (probably number of dives). This provides a measure of the risk. It’s somewhat skewed because the PDE participants usually know they are being scrutinized. Still, it’s much better than anything we’ve had in the past.

When comparing the safety of (say) air travel and automobile travel, you need a common denominator. Miles traveled is what the airline industry often uses.
To compare scuba with anything else, you also need a common denominator. But what should you use? Number of dives is unique to diving. Hours spent seems more reasonable, but would be difficult to gather for many other activities (swimming, hiking in lightning country, etc.)

So, I doubt we’ll ever completely discard our pet statements like:
“There are more people killed annually by pigs in Iowa than in scuba worldwide.”
(I heard that somewhere, so it must be true.)