is also included and represents the amount of resources devoted to medical care (i.e., the
more resources the fewer people on the waiting lists).
The medical technology proxies all have the expected sign with two of them being
significant at the 95% level. We also examined in a separate model the effect of infant
mortality rate as a proxy. This variable was highly correlated with other variables in the
model. Our results did not indicate that this variable was statistically significant, though this
may be due to the confounding effect of other variables. Average length of inpatient stay
does have a relatively high correlation with the percent of the population aged 15-24 (0.84).
Omission of the correlated variable does not change the robustness of the coefficient value of
the variables of interest.
To further examine the impact of medical technology, models were estimated with
both serious injuries and slight injuries as the dependent variable. If medical technology is
reducing fatalities associated with traffic accidents, we would expect many of these fatalities
to now be classified as injuries, especially serious injuries. If this is the case we would
expect the medical technology proxy variables to either have no significant effect on the
number of injuries or to have the opposite sign from the fatality model. We find the latter
effect for our variables as shown in Table 2, models D and F. Reduced inpatient stays in the
hospital are associated with increases in serious and slight injuries. Increased per-capita NHS
staffing levels also are associated with increased serious and slight injuries. Reductions in
hospital waiting lists are associated with increased serious injuries but there is no effect on
slight injuries. These results suggest that these changes could be due to better treatment of
what would have formerly been fatalities. More importantly, the difference in the sign of the
coefficients suggests that there is an inherent difference between the association of these
variables with fatalities as opposed to injuries.
14