A new method to evaluate the effects of bacterial dosage, infection route and Vibrio strain in experimental challenges of Litopenaeus vannamei, based on the Cox proportional hazard model

• Cox proportional hazards model was applied in Vibrio challenge test.
• The infection route had the highest effect on shrimp mortality, followed by dosage and strain.
• The prognostic index was constructed to predict population death time post Vibrio infection.

In the shrimp challenge test the Vibrio dosage, infection route, and strain are considered as risk factors that result in mortality. Assessment of Vibrio/shrimp interactions, and disease dynamics following infection by Vibrio, are useful techniques needed for detailed studies on the control of risk factors. In this paper we used an application of the Cox proportional hazard model to assess relative survival probability, estimate mortality risk, and construct a prognostic model to assess predictions of estimated time to death. Results indicate that infection route was the most important prognostic factor contributing to mortality in the challenge test (β = 3.698, P < 0.000). The shrimp infection rate following injection was found to be 40.4 times greater than that following immersion (hazard ratio (HR) = 40.4; p = 0.000). Our results also indicated that the HR resulting in shrimp mortality following a high dose of 108 cfu/shrimp was significantly greater (HR = 5.9, P < 0.000), than that following a baseline dosage of 107 cfu/shrimp. Strain Vh was found to be more virulent than Strain Vp (HR = 4.8; P < 0.000). The prognostic index also indicated that the infection route is the most important prognostic factor contributing to mortality in the challenge test.