This statement used to come up a lot, particularly referencing the lethality, and therefore suitability, of the .22 Long Rifle as a self-defense or caliber of last choice for self-defense.
On the surface, it doesn't appear to make sense; however, it has been repeated frequently so maybe there's a grain of truth to it, right?
There have been many articles on the internet debunking the supposed high-lethality of the .22LR when compared to other potential calibers. To their credit, I conclude that most of them are right. Ceteris Paribus, most larger calibers do tend to be more lethal than a .22LR. However, yes, there really is a germ of truth to the statement.
I would like to reference:
The Medium is the Message: Firearm Caliber as a Determinant of Death from Assault, by Franklin E. Zimring, Journal of Legal Studies. Berkeley Law, 1-1-1972
This study is, in fact, a study of deaths by caliber in Chicago in 1970. The intention of the study was to attempt to quantify lethality by caliber with the thesis that larger calibers are more lethal and to quantify homicide by motive (intentional or unintentional).
Table 5, indicates that of the 113 (known caliber) fatal attacks, about 1/3 of all of them were with the .22 caliber, inclusive of [.22, .25, .32, .38, >.38] calibers.
When including the .25 in that mix, the two (.22 & .25) account for around half of all fatalities, outnumbering by far any other individual caliber in the set. Table 7, "Estimated Death Rate from Gun Attacks by Wound Location, Number of Wounds, and Gun Caliber" further reinforces that trend.
The study then goes on the normalize against multiple hits and location of hit as well as a number of other factors, eventually concluding that (brace yourself), yes, caliber matters and that the relative "dangerousness" of the .38 (and, presumably, larger) is about 3 times greater, according to Figure 1 (pp106 journal, pp11 in PDF).
The gist of it is, that more people were indeed killed with the .22 caliber than any other individual caliber within the study sample (Chicago, 1970) but that appears to be because there were a lot more people shot with a .22 than any other weapon, apparently because the .22 was a far more common weapon, not because it is more lethal.
It has been speculated that these numbers would probably be applicable to the general U.S. as a whole during the 1960's and 1970's due to the affordable price of inexpensive .22LR and .25ACP "Saturday Night Special" handguns, both domestic and imports.