Finding effective ways to monitor laying hen welfare is challenging as UK flock sizes can reach 16,000 birds. Eggs provide potential for welfare monitoring, as they are a daily output with previous evidence of links to stress. We explored the associations between stressors and eggs using two complementary studies. In Study 1, hens experienced social or heat stressors and eggs were scored daily for defects in shell characteristics. All eggs were scored on a three-point scale: 1 (no defect); 2 (minor defects); or 3 (unsuitable for whole egg sale in the UK). Texture defects were higher after stress treatments and were explored further as a promising proxy measure of welfare. In Study 2, eggshell texture from five commercial flocks was scored before versus at the onset of an avian influenza-enforced indoor housing, and scores were correlated with industry data for egg quality. Eggs were more likely to have texture defects after the enforced indoor housing, and manually scored texture correlated significantly with shell strength and shell colour during automated grading. Shell strength was weaker immediately after the enforced indoor housing and eggs were darker. We suggest that eggshell texture could be a useful addition to assessing changes or stresses in a hen’s environment for both research and commercial purposes, but further validation is needed to understand the generalisability of these results to other stressors. Additionally, data already collected in factories, such as shell strength and colour, may provide information on stress and could be valorised for understanding hen welfare.

The aim of this study was to assess the extent to which three non-invasive measures of welfare in laying hens (egg-shell quality, corticosteroid levels as measured from the birds' faeces, and behavioural preferences) were correlated over a period of five days in two groups of birds. One group had access to an enriched test area (bark chips on the floor and a tray of sprouted wheat); the other group had access to a comparably sized barren area (bare wire mesh floor). The measure of preference used was the amount of time hens spent in the test area as measured each day. It was predicted that birds with access to the less preferred environment would show higher levels of faecal corticosteroids and egg-shell anomalies. However, although the birds showed a preference for the enriched environment from Day 1, the other two measures did not follow the same pattern. Faecal corticosteroid metabolites showed an initial increase in both groups, which declined significantly by Day 4, with the ‘enriched’ birds in fact showing a trend for higher levels than the ‘barren’ birds. Shell thickness also showed a change over the five days, but with a different time course: declining to a minimal level on Day 3 and then rising again by Day 5. No measure of shell quality was significantly different between the two environments, but there was a trend for changes in shell thickness to be more pronounced in eggs from enriched birds. The results indicate the caution that needs to be exercised in using shell quality or corticosteroid measurements in isolation from assessments of what the animals themselves prefer.