Integrating Ethnography and Structural Equation Modelling to Assess Brucellosis Knowledge, Attitude, and Practices among Pastoralist Communities in Kenya

Background

Brucellosis is a neglected zoonotic disease endemic in pastoralist communities. It is driven by close human-animal interactions and culturally embedded practices such as consuming raw milk and blood, handling aborted materials, and skins and hides, and limited access to veterinary and healthcare services. Understanding how knowledge, practices, and cultural beliefs intersect structurally is essential for designing context specific One Health interventions for effective disease control.

Methods

We employed a mixed-methods approach to explore brucellosis-related knowledge, attitudes, and practices among pastoralist communities in Kajiado and Marsabit counties, in Kenya. Subsequently, we used partial least squares structural equation modeling (PSL-SEM) to estimate the direct and indirect effect of the latent variables (K, A, P) on the brucellosis transmission pathway.

Results

Brucellosis knowledge, attitude, and practices were markedly poor in both counties. Pastoralists in Marsabit demonstrated significantly better knowledge (54% vs 36%, p<0.001), and attitude (45% vs 23%, p<0.001) but worse practice scores (26% vs 52%, p<0.001), compared to Kajiado. Risky practices, such as raw milk consumption and unsafe disposal of animal birthing products, were widespread. PLS-SEM analysis showed a strong indirect effect of knowledge on practices mediated through attitudes: (Knowledge → Attitude: β = 0.921, p<0.001; Attitude → Practice: β = 0.580, p< 0.001), with minimal direct impact (Knowledge → Practice: β = 0.188). Cultural beliefs, such as resistance to boiling milk or using colostrum for healing – persisted regardless of awareness levels.

Discussion

This study reinforces the observation that knowledge alone is insufficient for behavior change. Cultural norms, traditional knowledge systems, and limited health infrastructure play a critical role in sustaining risk behaviors.

Conclusions

Effective brucellosis control must integrate culturally grounded, community-specific strategies that strengthen attitudes and promote safer practices within a One Health framework.

KEY MESSAGES

What is already known on this topic

• Brucellosis transmission is shaped by deeply embedded cultural practices and low levels of disease awareness among pastoralist communities.

• Persistent risky sanitary and dietary practices in pastoralist communities is driven by culture and survival necessity trade-offs.

• Previous studies on brucellosis transmission among pastoralist communities often rely on quantitative surveys that miss cultural and contextual nuances.

What this study adds

• Our study provides clear understanding of how knowledge, attitudes, and practices (KAP) interact structurally within pastoralist communities.

• The study bridges the gap between qualitative cultural insights and quantitative modeling to better inform zoonotic disease control interventions.

• The study demonstrates that combining ethnography with Structural Equation Modelling (SEM) provides a richer, more nuanced understanding of KAP dynamics around brucellosis.

• It reveals that cultural beliefs and attitudes significantly mediate the structural interaction between knowledge and practices, and confirms that knowledge alone isn’t adequate to influence safe dietary and sanitary practices.

• The SEM model identified key latent variables such as trust in veterinary services and perceived disease severity that influence preventive practices.

How this study might affect research, practice or policy

• The study findings advocate for culturally tailored health education that goes beyond information dissemination to address underlying beliefs and social structures.

• Policymakers, veterinary and public health practitioners can leverage the SEM model to target zoonotic disease control programs at the most influential behavioral drivers.

• To be effective future zoonotic disease control interventions must integrate traditional knowledge and local contexts into One Health frameworks to promote and sustain effective behavioral change.

• The study sets a precedent for integrating qualitative and quantitative methods in One Health research, promoting interdisciplinary collaboration.

• This integrated mixed-methods and structural modelling approach offers a replicable framework for studying zoonoses in other marginalized populations.