In Bohmian Mechanics, it assumes a universal wave field in which particle's motion can be calculated using Newton's law of motion:
\begin{equation} m\frac{d^{2}x}{dt^{2}} = - \nabla(V+U) \end{equation}
where $V$ is the classical potential and $U$ is the 'quantum potential'.
In this way, Bohm showed that his theory can recover all the prediction made by the Copenhagen interpretation. However, I feel a bit confused about his explanation to the EPR paradox. We know from Bell's inequality that the value of observable does not take a definite value prior to the measurement and an instantaneous interaction exists between two particles. In Bohm's view, particle behaves in a deterministic way. So I feel in Bohm's theory the particle should be in a certain state before measurement which shouldn't be right according to Bell's inequality. If so, in its explanation of the EPR paradox (Bohm 1952), where does the superposition of states come from?