[Bovine hypodermyiasis: a quasi sequential procedure for observation of live-stock for surveillance of recrudescence in eradicated zones].

A quasi-sequential sampling plan for observing cattle was undertaken in the framework of Hypoderma bovis epidemiological surveillance. This plan enabled us to test whether or not the prevalence of Hypoderma bovis is within a threshold p0 chosen for an eradication scheme or for qualifying a zone. The sampling cost was reduced but the estimation precision remained acceptable. The sensitivity of monthly controls with respect to the reference period (April, May, June, July) was studied from data observed over 6 consecutive years in Côtes d'Armor, France. The best ratio sensitivity/sampling cost was obtained with a unique counting plan in June where 48% (36 to 60%) of the herds found to be infested during the reference period were detected. The test procedure concerning the prevalence level was then adapted to take into account the sensitivity of the observation method. This is tantamount to reducing the threshold p0.