Prostitute Mortality & Homicide Rates

According to the US Bureau of Labor Statistics, the industry with the highest rate of job related fatalities is a miscellaneous grouping called Agriculture, Forestry, Fishing, & Hunting, with 29.6 fatalities per 100,000 workers in 2006. A very close second is Mining, with 27.8 fatalities per 100,000 workers in 2006. Taxi driving, a sub-industry not broken out in the BLS's statistics, has been estimated by other researchers to have a homicide rate of 29 per 100,000 taxi drivers per year. Taxi drivers carry large sums of cash in their cars and are easy to rob, which makes them the frequent target of armed robbers and makes their job very dangerous.

The recent observance of the International Day to End Violence Against Sex Workers got me wondering about how dangerous it really is to be a prostitute. It's impossible to accurately calculate job related fatalities among prostitutes for the US, for obvious reasons. As a proxy for an accurate estimate, I used a study done using police and health department records for Colorado Springs, Colorado, published in the American Journal of Epidemiology in 2004. The records cover the period from 1967 to 1999. The rate of death among active prostitutes due to homicide was estimated to be 229 per 100,000 prostitutes per year. Job related mortality due to all causes for prostitutes was estimated to be 459 per 100,000 prostitutes per year. However, the authors of the paper said the following about their research: "It is likely that we underestimated mortality in our current analysis."

The authors cited a study done in Nairobi, Kenya, which estimated the mortality rate for prostitutes at 310 per 100,000, and another study done in London, England, with 401 deaths per 100,000 prostitutes. Three different Canadian data sets yield homicide mortality rates of 181, 112 - 225, and 127 per 100,000. A study of prostitute mortality in London, published in Sexually Transmitted Infections in 2006, estimates the overall mortality rate at 480 per 100,000 prostitutes per year. So the estimates for the homicide rates and the overall mortality rates in the US are in line with estimates from elsewhere in the world.

A complication is that the mortality estimates for prostitution is based only on data for visible prostitution: streetwalkers and other prostitutes who are known the the police and health authorities. The majority of prostitutes operate out of sight of public officials. These are prostitutes who work under safer conditions; escorts, for example. Given the fact that we have no empirical information on which to base an estimate of the total number of prostitutes, or the mortality rates for all prostitutes, there are only three things we can say with reasonable confidence.

First, that the visible prostitution sector has a very high job related mortality rate; 229 per 100,000 per year for homicide, and 459 per 100,000 per year for all causes. Very roughly, that's sixteen times the mortality rate for the two most dangerous industries tracked by the BLS.

Secondly, the mortality rates for the hidden prostitution sector are unknown.

And thirdly, while prostitutes who work in the hidden sector are less exposed to criminal violence, they are still more exposed than women in the general population. Prostitutes in general can't go to the police for protection because what they do is illegal. And prostitutes in the hidden sector spend time alone with clients they don't know, or don't know very well. That's less of a risk for escorts than it is for streetwalkers because escorts can screen customers and think about the information for a while before they make an irrevocable decision. Streetwalkers have to make instantaneous decisions with almost no information about the customer. But the risk for escorts still exists, and the real rates of both homocide and non-lethal violent attacks are almost certainly higher for prostitutes in the hidden sector than for women in the general population. We just have no way of estimating how much higher.

However, there's another problem with an overall estimate for prostitution mortality rates. The difference between mortality rates for visible and hidden prostitution is probably so high that, even if statistics for the hidden sector were available, averaging the rates across all types of prostitution would produce statistics that were too high for the hidden sector, and much too low for the visible sector. In other words, average mortality rates for all prostitutes wouldn't represent the actual mortality rates for any specific group of prostitutes, or the risk faced by any individual prostitute. Given the wide variety of conditions under which different groups of prostitutes work, meaningful statistics have to target specific groups.