Laboratory data are the cornerstone in surveillance of infectious disease. We investigated whether changes in reported incidence of Campylobacter and Salmonella infection might be explained by changes in stool sampling rates. Data were extracted from a national database on 585 843 patient stool samples tested by microbiology laboratories in Wales between 1998 and 2008. Salmonella incidence fell from 43 to 19 episodes/100 000 population but Campylobacter incidence after declining from 111/100 000 in 1998 to 84/100 000 in 2003 rose to 119/100 000 in 2008. The proportion of the population sampled rose from 2·0% in 1998 to 2·8% in 2008, mostly due to increases in samples from hospital patients and older adults. The proportion of positive samples declined for both Salmonella and Campylobacter from 3·1% to 1·1% and from 8·9% to 7·5%, respectively. The decline in Salmonella incidence is so substantial that it is not masked even by increased stool sampling, but the recent rise in Campylobacter incidence may be a surveillance artefact largely due to the increase in stool sampling in older people.