Test-retest reliability of the Welfare Quality Assessment protocol for pigs applied to sows and piglets. Part 2. Assessment of the principles good feeding, good housing, and good health1.

The present study aimed at testing the feasibility and on-farm test-retest reliability of the Welfare Quality Assessment protocol for pigs applied to sows and piglets. The study was conducted on 13 farms in Northern Germany, which were visited 5 times by the same observer, and included 2 experimental setups: first, the complete Welfare Quality Assessment protocol for sows and piglets was applied to the farms. Second, additional assessments in the gestation unit considered all sows. The complete protocol assessments were used to evaluate the feasibility of the protocol. Furthermore, the data were analyzed with regard to on-farm test-retest reliability. The present publication focuses on the Welfare Quality principles good feeding, good housing, and good health, which are based on individual indicators (IN). The second experimental setup was utilized to verify the test-retest reliability of IN in the gestation unit with an increased number of animals under assessment. The test-retest reliability was calculated using Spearman's rank correlation coefficient (RS), intraclass correlation coefficient (ICC), smallest detectable change (SDC), and limits of agreement (LoA). Farm visit 1 (F1; day 0) was set as a reference and compared with farm visits 2 to 5 (F2 to 5; day 3, week 7, month 5, month 10). The IN of the above-named Welfare Quality principles achieved mostly acceptable test-retest reliability (e.g., wounds on the body F1 to F4: RS 0.34 to 0.57, ICC 0.40 to 0.41, SDC 0.02 to 0.12, LoA [-0.03; 0.02] to [-0.09; 0.14]) in terms of the on-farm test-retest reliability. Poor test-retest reliability was detected for body condition score concerning the principle good feeding, for bursitis and panting in sows and for huddling and panting in piglets within the principle good housing, and finally for vulva lesions, metritis, and local infections in sows and for scouring and lameness in piglets in the principle good health. Variations among the farm visits, which resulted in poor test-retest reliability, may be explained by seasonal effects (panting), moving animals (bursitis, lameness, huddling), rare occurrences of diseases (metritis, local infections, scouring), and differently conditioned sow groups (body condition score). The second experimental setup confirmed the results for IN in the gestation unit. Thus, the reported test-retest reliability determines the Welfare Quality Assessment protocol for sows and piglets to be a reliable approach to assess welfare in sows and piglets.