Identifying factors to Predict Sows Longevity in Commercial Farms

Abstract The aim of the present observational study was to identify factors associated with gilt longevity. The objective was to develop an on-farm, multi-criteria predictive model to classify gilts according to their maternal success and survival potential. A total of 512 sows (Hermitage) from a single commercial farm comprising 8,500 sows were monitored from birth until removal from the herd. At first service, age, backfat thickness, and loin depth were recorded. Sows were subsequently monitored until herd removal. All causes of removal were recorded and treated as equivalent for the purpose of this analysis. A regression tree analysis was performed (RStudio version 4.3.2; R Core Team, 2024) to identify which variables should be prioritized for inclusion in future predictive models and to determine optimal thresholds for categorizing numerical variables. Cox proportional hazards models were subsequently applied to these selected variables and thresholds to evaluate their association with herd survival. Results from the regression tree indicated that age at first service (AFS) was the most relevant factor influencing sow longevity, followed by backfat thickness at first service (BF). Loin depth at first service did not contribute significantly to the model and was therefore excluded from the Cox analysis. AFS, analyzed as a continuous variable, significantly increased the risk of herd removal by 0.35% per additional day (p = 0.035). When analyzed categorically, gilts with an AFS greater than 309 days had a 28% higher risk of removal (p = 0.007). Sows with BF below 13 mm at first service showed a tendency toward an increased risk of removal by 16% (p = 0.13). When AFS (continuous) was combined with BF, sows with less than 13 mm of BF at first service showed a 25% higher risk of removal (p = 0.025), while AFS increased the risk by 0.45% per additional day. When both variables were analyzed categorically (BF 13 mm and AFS 309 days), the risk of herd removal increased significantly by 22% for sows with less than 13 mm of BF (p = 0.04) and by 33% for those with an AFS greater than 309 days (p = 0.0026). In conclusion, AFS was identified as the most influential factor affecting sow survival. Younger age at first service was associated with a greater likelihood of remaining in the herd for a longer period. Although backfat thickness (BF) at first service was not a determinant factor on its own, its interaction with AFS indicated that sows with less than 13 mm of BF at first mating had an increased risk of herd removal.