Empirical and mathematical models on the relationship between patient age and nosocomial infection

This paper proposes two models, one a purely empirical one and the other a mathematical one, which depict the relationship between patient age and nosocomial infection rate. The empirical model is based on the two age-specific phenomena, the acquisition of resistance to infection with age mainly in the early years of life and the deterioration of existing resistance mechanisms with ageing predominantly in the later years of life. The net effect of these two factors is shown to approximate into a quadratic relationship between age and nosocomial infection rate, like the type demonstrated in earlier experimental studies. The second mathematical model is derived from studies on cancer research and here the relationship between age and nosocomial infection rate for patients in the age group 30-70 years is represented by a log linear model. The model was tested against experimental data derived from large surveys on nosocomial infection and the resulting correlation coefficient was 0.98. The model was an extremely good fit when tested against postoperative wound infection rates as well as nasal carriage rate of antibiotic resistant Staph. aureus. Furthermore, when patients in the survey were subdivided into groups of male patients and female patients and into two groups based on the type of operative wound, the model was still found to be a very good fit to the experimental data. This confirmed the validity of the model even in the presence of other patient-related parameters. Finally, the model was tested against the results of a totally different experimental study conducted elsewhere and the resulting correlation coefficient was 0.999, which confirmed the validity of the model in a universal context.