The impressive growth of secrecy-sensitive wireless applications calls for methods to complement traditional cryptography. In particular, physical-layer security is attractive to enhance communication confidentiality by exploiting the characteristics of the wireless environment. Recent works on information-theoretic wireless network secrecy have shown how to exploit interference to enhance the level of information confidentiality for node spatial distribution according to a homogeneous Poisson point process. We propose a framework for design and analysis of inhomogeneous Poisson networks with maximum secrecy rate. We determine the effects of the inhomogeneous node spatial distribution on the secrecy metrics, which provide guidelines for network design.