Modelling bodyweight to avoid anthelmintic underdosing of goats in resources-limited settings

Improved management of livestock in resource-limited settings can provide a means towards improved human nutrition and livelihoods. However, gastrointestinal nematodes (GIN) are a significant production-limiting factor. Anthelmintics play a role in GIN management; however, few anthelmintic classes are available in many low-middle-income countries. Utilising a limited range of classes may increase selection for anthelmintic resistance; therefore, strategies to reduce other selective pressures are of heightened importance. Avoiding anthelmintic underdosing is one such strategy, but it can be challenging without access to accurate bodyweight measurement. Many previous studies have used thoracic girth as a practical proxy for bodyweight in goats; however, they have rarely considered the potential impact of natural variation on therapeutic doses. Here, the relationship between bodyweight and thoracic girth was modelled using data from 820 goats from three Malawian biomes in two seasons, with the specific aim of avoiding underestimation of bodyweight. The internally cross-validated linear regression (∛Weight ~ 0.053 + 0.040*Girth, R² = 0.92, rounded up to the nearest 5 kg) was validated against data from an additional 352 Malawian goats (1.4% of goats allocated an underdose and 10.2% allocated a dose > 200% of bodyweight). The equation was further externally validated using an historical dataset of 150 goats from Assam, India (2.7% of goats were allocated to an underdose and 24.8% allocated to a > 200% of bodyweight). These results suggest that a more globally generalisable approach may be feasible, provided the accuracy of the estimate is considered alongside the therapeutic index of the pharmaceutical.