An interactive computer-based guide (termed RESPITE) was developed to demonstrate to pig producers the expected level of pneumonia in pig herds which follow specified management practices known to influence the severity of pneumonia problems. A program was used to evaluate the combined effect of known risk factors present in the particular herd on the prevalence of lung lesions in slaughter pigs. The expected prevalence was estimated using formulae that were derived from observations and data in the published literature. Eleven risk factors considered are the number of pigs in the same room, all-in/all-out versus continuous flow of pigs, closed versus open herd, temperature fluctuation ratio, width of the buildings, space allotted per pig, the type of partitions separating adjacent pens, the presence or absence of diarrhoea as a clinical problem, liquid versus solid manure disposal, ascarid control efficiency, and the presence or absence of active Aujeszky's disease in the herd. Each of the factors can be assessed in the farrowing, nursery and grow-finish stages of production and the appropriate observations are entered. The computer uses the data to calculate the expected prevalence of pneumonia in animals leaving that stage of production. Of the pneumonia developing in the farrowing, nursery and grow-finish stages, 90 per cent, 75 per cent and 25 per cent respectively is considered to resolve before slaughter inspection at 100 kg. Adjusting for this degree of resolution, the program then estimates the expected final prevalence of pneumonic lesions in slaughter pigs. Comparison of predictions with slaughterhouse data from four herds indicated that the guide was realistic in its estimates.