Assessment of cattle production systems and herd health practices as drivers of artificial insemination adoption in selected districts in Uganda

Abstract

This study assessed cattle production systems, herd health practices, and their influence on artificial insemination (AI) adoption in the districts of Mbarara, Kabale, and Mpigi. A cross-sectional survey was conducted with 355 cattle keeping households selected via a multistage design combining purposive, stratified, probability proportional to size, and simple random sampling. Neyman allocation guided district samples (Mbarara = 120, Kabale = 132, Mpigi = 103). A structured, pre-tested questionnaire (pilot n=30; Cronbach’s α=0.87) was administered in face-to-face interviews during September to October 2024 using Kobo Collect. Data were cleaned in Microsoft Excel (2007) and analyzed in SPSS version 26. Descriptive statistics summarized demographics, production systems, herd health, and AI adoption. Pearson’s Chi-square tested group differences in the districts (α=0.05). A multiple regression model was used to show how demographic characteristics, cattle production systems, and health practices predict artificial insemination adoption. The study foundout that cattle production systems differ significantly by district (p<0.001 throughout). Kabale was high in zero-grazing and crossbred profile. Mpigi had more exotic breeds yet low milk yield was observed. Mbarara reported high use of extensive and semi-intensive system. Herd health practices were largely homogeneous, with only routine herd testing of diseases varying by district. AI access and use were uneven. Access was greater in Mbarara and Kabale and lowest in Mpigi. Information channels and perceived barriers were district-specific. In multivariable analysis, male and Christian religion were positively associated with AI use (p≤0.022), whereas keeping local, crossbred, or exotic pure breeds, fenced farming, higher milk yield per cow, reliance on home-grown feeds, and conducting deworming, tick control, or broader biosecurity were negatively associated (p≤0.045). These findings may support use of targeted and district-specific strategies that both expand reliable AI access in areas with low use and reframe the value scheme of AI for higher-performing, fenced, and bio secure farms. Furthermore, tailored communication through local information channels and inclusion-focused outreach can close demographic, production systems, and herd health management gaps in AI uptake.