Sarcoptic mange in pig breeding and finishing farms is a major economic problem. The economic losses by mange are caused by the costs of treatment and damage to piggery fixtures through permanent rubbing, reduced feed conversion efficiency, increased return rates, and piglet mortalities (Davies, 1995). Horst (2004) estimated that the loss due to reduced feed conversion efficacy and increased piglet mortalities at €66 per mangy sow and year. Kirchner (1998) reported a mortality of 11.5% in piglet as a consequence of frequent unrest (pruritus) in untreated mangy sows, in contrast to 3.7% in treated sows. Damriyasa et al. (2004) examined 11 breeding farms with a total of 2754 sows and estimated the mean economic loss due to Sarcoptes scabiei infestation at €4200 per affected farm and year. Based on these economic losses and the animal welfare issues, there is a need to eradicate and prevent sarcoptic mange and to establish mange-free pig farms. To achieve a mange-free status and certification a reliable management and control program, and sensitive and specific diagnostic methods are required as complementary tools. The enzyme-linked immunosorbent assay (ELISA), skin scraping, clinical score, papular dermatitis score and scratching index are tools with differing specificities and sensitivities. Of these, the scratching index using different limiting values is often used and has been recommended by various authors as an additional method for the diagnosis of sarcoptic mange. The aim of the present study was to observe the scratching behaviour of growing pigs suffering from sarcoptic mange in a closed herd before and after treatment, to calculate scratching indices using different methods and to critically evaluate the results obtained employing recommended limiting values. In a closed pig breeding and finishing herd suffering from sarcoptic mange, two selected groups of pigs were filmed during a period of 10 days before and after treatment. The observation always commenced each hour and lasted for 15 min. Before treatment, observations was done round the clock, after treatment from 8:00 to 22:15. Before treatment the pens were stocked with 11 (pen A) and 10 (pen B) growing pigs (Large White x Landrace sows; 5 months old) with an average weight of ~70 kg examined for sarcoptic mange by skin scrapings and ELISA. The animals had never been treated with an acaricide or endectocide before. After 10 days, the pigs were treated twice (18 days interval) with Dectomax® 1% solution for pigs (Pfizer, Austria) at a dose of 0.3 mg Doramectin i.m./kg body weight. After treatment, seven pigs were observed in both pens. Most scratching actions both before (83.1%) and after (94.5%) treatment were of one to 10s. After treatment, the 10s scratching episodes decreased by 67.3% (from 21.2 to 6.9 mean SRE/pig), and the scratching actions of longer than 10s by 91.7% (from 4.3 to 0.4 mean SRE/pig), such that the latter could be observed only occasionally after treatment. A distinct increase in scratching activity both before and after treatment could be observed primarily between 10:00 and 15:00. Significant differences of scratching and rubbing activity between before and after treatment could also be seen at midday. Depending on the methods used and the limiting values set, 6.7–34.6% of the observations before and 2.0–17.3% of the observations after treatment revealed a ‘‘strong evidence of mange’’ or a ‘‘suspicion of mange’’. All other observations indicated that the pigs were free from mange.