Sexually transmitted disease surveillance.

cell count is of no value in such situations in evaluating the probability of acute bacterial infection, as in a proportion of patients with infectious mononucleosis the first change evi- dent is a neutrophil leucocytosis.'To put my point in the language of clinical epidemiology, the positive predictive value of a neutrophil leucocytosis in diagnosing bacterial throat infection is low when the prevalence of infectious mononucleosis is high.This has obvious implications for those who care for young populations.Secondly, since this study was presumably designed to emphasise the importance of rational therapy of sore throats, I would like to ponder on the possible importance of irrational therapy in the dramatic reduction in poststreptococcal glomerulonephritis and rheu- matic fever which has taken place in the last three decades.Did wholesale prescription of antibiotics when they were available have a valuable role to play in the virtual dis- appearance of these diseases ?Is it only now that we can afford the luxury of a sophisticated rational therapy ?