Intelligent Healthcare – Part 1

Integration is more than the coding of application or data interfaces. When dealing with complex integration within or across enterprises, there must be sufficient discipline to achieve reproducible results. Furthermore, that discipline must be tailored to the unique requirements of the domain/s in question. Few domains are as complex as Healthcare. Even more important perhaps is that integration cannot be viewed outside of the context of the outcomes within the domains they are meant to serve. Technical success may not translate to process or performance improvement if the relationships between domain goals and enabling technologies aren’t properly understood. Some of the basic concepts associated with our IH include the following:

  • Semantic Correlation – This concept is focused upon the realization that every data structure is based upon Semantics. This includes every data schema, every exchange protocol, all medical terminology etc. Semantic Correlation is a mechanism to bridge various data structures with minimal integration. This can be applied to both systems integration as well Program Lifecycle Management.
  •  
  • Federated Lifecycle Management – Global organizations must manage dozens of projects simultaneously. Many of these projects fit within research or organizational themes or sets. These themes also often correspond to larger themes involving many other external organizations. The problem that has plagued organizations trying to facilitate initiatives across domains is how to manage projects that aren’t actually under their control. Federated Lifecycle Management is both a philosophy and automation approach that attacks the problem through improved collaboration and community defined guidelines. It builds upon a core set of shared ‘project’ and/or community Semantics to help define and manage requirements that it in turn provides the basis for more sophisticated support elements.
  • Patient Centric Performance – Our stakeholder community is made up of caregivers and information technology providers. However, the true stakeholder community which underlies all Healthcare activities is the patient community. While we are obliged to meet various service level obligations on a systems level, the ultimate objectives must always include a perspective on patient outcomes. This requires a somewhat different philosophy for performance metrics of Healthcare solutions. The importance of this change cannot be understated – until we define how the technology can or should facilitate specific outcomes, it will be difficult to achieve those outcomes with any assured regularity.

It is entirely likely that the amount of data that Healthcare providers will have to manage in five years will be exponentially larger than today’s set. This data growth will tax both our infrastructure and the practice of Healthcare itself. All medical outcomes are based upon access to accurate and readily available information. Issues related to system connectivity, response time and accuracy are likely to impact Healthcare provision negatively if new approaches to Healthcare information management are not developed quickly. These types of issues have already been experienced in several major DoD and Federal healthcare initiatives.

Our primary problem is increasing complexity. The solution must be squarely aimed at how that problem will evolve over the next decade, rather than focus solely upon mastery of the past decade’s technology. This “capability gap” can only be bridged through adoption of an evolutionary Healthcare integration philosophy.

Figure 1 illustrates that Intelligent Healthcare is focused up front on Cross Domain interoperability issues or capability. The reason for this is simple, even in individual cases, the need for Cross Domain care coordination may make the difference in achieving a successful patient outcomes. By the same token, Cross Domain coordination is even more important in managing Public Health, supporting National Defense (through activities such as Biosurveillance) and ensuring rapid dissemination of practice improvements (lessons learned). Healthcare requires community classification, dynamic correlation and identification of complex relationships and the ability to track or trace events across event lifecycles (connecting incidents instead of viewing them as isolated and disconnected). Cross Domain interoperability is much more than adherence to data exchange standards, it requires instead a shared understanding of best practices and understanding in advance of how resources and capabilities can or should be coordinated and orchestration to affect desired outcomes.

Cross Domain Interoperability
 
 
 

The core architectural paradigm behind the next generation of automated Healthcare solutions is likely to be semantic data fusion. The reason for this is simple – complexity. The key enabling mechanism within data fusion is reliance on metadata. Metadata has been managed historically within individual system stacks. What that means is that the metadata was usually stored within the same system or if in more complex configurations within the system or component close to the primary user interface (and in the case of EMR solutions, the Medical Chart UI). The problem with this for Healthcare is that the community of Medical knowledge is expanding exponentially. Any attempt to centrally manage all types of medical knowledge in one system’s metadata framework is bound to fail. Part of the true strength of any such solution ought to be the ability to accommodate information not originally anticipated.

The implications of this are important, as there is no consistent method of coordinating metadata across multiple metadata sources except through custom interface development or management in most of today’s EHRs. As the scope of custom integrations grows, interface management and integration become mores difficult to manage. In other words, those architectures simply are not scalable.

The reason for our focus on semantic fusion then is a result of the recognition that the limitations of data exchange-based integration and Health record focused solutions will ultimately lead to a rethinking of the overall integration paradigm. In other words, as providers and developers alike run into the brick wall of designing around static document-based views of patients and process, they will also see the potential for adding dynamic capability around the edges utilizing fine grained SOA-based applications. These smaller applications will exploit federated data caches, warehouses or marts filled with a variety of data from many sources.

The other part of the paradigm shift that is likely to occur is the move away from EHRs and PHRs as the primary Healthcare data management platforms. The EHRs and PHRs are still incident or case-focused and lack the ability to identify trends or link to larger trends (i.e. those occurring across a family, group, community, organization or geographical region). The foundational information that these platforms can track is helpful in providing an organized way to present data to the physician, but ultimately there is no way to flag for the patient (user) what may be troubling or not (and help them determine when to seek treatment or when to engage in preventative care).     

The true goal for Healthcare interoperability is not a messaging or UI platform solution per se, both of those perspectives are still more technology focused than patient-centric. The true goal for Healthcare interoperability lies in the ability to expose the entire set of Healthcare data to the holistic practice of medicine – delivering data where and when it’s needed and at the correct level of granularity or detail. Once that data is delivered, it must also be placed in the proper context – this is the semantic fusion we spoke of before. This is not data fusion in the traditional sense in that all of the potential fusion scenarios cannot be designed in advance – it has to involve dynamic data aggregation, correlation and presentation.
Intelligent Healthcare
 
 
 
 
copyright 2010, Stephen Lahanas