With the growing digitalization of manufacturing and the associated reliance on digital data, manufacturers, their supply chains, and their customers are exposed to many and varied digital threats. One such digital threat is compromised integrity of the data produced and consumed, the result of data tampering, either intentional (i.e., cyberattacks) or unintentional (e.g., human or system error). Because of the reliance on digital data, data tampering can lead to serious damage in the physical world, such as structurally weaker or functionally different parts. In the cybersecurity world, a key factor in the cost of a digital threat is the mean time to identify and contain that threat. Because of the complexity of today’s goods and the distributed nature of supply chains, designing, manufacturing, and distributing a product involves a significant number of heterogeneous information systems operating on both business and engineering data. These systems are integrated to consume data from each other in a high-volume and fast-paced environment where tampered data can be quickly propagated across many systems and organizations. To promptly identify the “infected” systems and contain the threat, an efficient data traceability strategy is required. Although we have previously developed traceability methods for file-based and agent-based integrations, this paper discusses the need for and benefits of traceability at the data field level, supported by model-based integration specifications (or data mappings). We then present a state of the art of existing formal integration definition methods. Finally, we conclude with an evaluation of these methods.