The CDS Collaboratory Architecture (CCA) is designed to facilitate behavior change and improve health care outcomes for patients. This is a data-driven, goal-oriented objective, and consequently, the infrastructure is designed to request and persist clinical data from distributed data sources. It aggregates this clinical information to create a comprehensive operational dataset ideally suited for developing individual and population-based action plans. A design principle unique to our approach is that rule and workflow processing can be configured within a patient-specific context, each session dynamically instanced and provisioned with select knowledge bases and individualized preferences. This design, while resource intensive, ensures personalized, high-performance rule evaluations and workflow orchestration.

CCA manages the data structure and semantics required to aggregate heterogeneous, distributed data into a collection of normalized, canonical data. Furthermore, it recognizes that not all analyses of this operational data store are best approached with predicate logic; some require alternative inference techniques. To expand the analytic capabilities available, we extended the Drools rule engine with a Predictive Model Markup Language (PMML) interface, a standard used to represent predictive models such as support vector machines, neural networks, or cluster algorithms. By enabling the Drools engine to import and emulate the functionality of alternative inference models, CCA expands the reasoning techniques available to solve the problem at hand.

In addition to the core services required to deliver these functional capabilities, CCA includes several prototypical applications designed to facilitate knowledge management and promote usability in the clinical environment. In the coming months, the project will contribute the following applications and demonstrate how they can be layered upon existing health information systems.

Patient Portal: Based on the CareWeb framework, Patient Portal is a patient-focused application seeks to improve collaboration with healthcare professionals by enabling a) complete access to clinical data medical records (e.g. HL7 FHIR Server), b) secure messaging, and c) context sensitive educational material (e.g. HL7 Infobutton). The portal provides facilities for patients to receive recommendations and alerts related to their care, and complete tasks requested by their healthcare provider team or by an automated CDS system. The portal will support the SMART on FHIR initiative, allowing compliant third-party modules to be easily integrated into the portal framework.
Provider Portal: Provider Portal, a simple reconfiguration of the above Patient Portal, is focused on the healthcare professional needing additional tools designed to facilitate decision-making and workflow-centric tasks. Through the Provider Portal, personnel will have access to clinical analytics and diagnosis tools for conducting risk assessments, disease probability analyses, diagnostic workups, treatment planning, and transition planning. The portal complies with the HL7 Clinical Context Object Workgroup (CCOW) standard.
Universal Inbox: The Universal Inbox is the central messaging and workflow management tool in both Portals. It is similar to other well-known messaging clients (e.g., MS Outlook), but is additionally able to manage CDS alerts, with or without associated tasks, HL7 Clinical Document Architecture (CDA) document exchanges (request and receive) from other healthcare organizations and medical device data uploaded by patients from home.
Clinical Analytics Drawer: The Clinical Analytics Drawer displays information regarding a patient's relative risk for developing one or more diseases. This analysis is done automatically using patient information that CCA retrieves from source systems in the background. For example, when the patient checks into a clinic for an office visit, the system feeds relevant information to predictive models that assess the patient’s risk for specific conditions. By the time the provider sees the patient, the results of that analysis are available within the Clinical Analytics Drawer in a way that is sensitive to their workflow.
Diagnostic Guide: A high relative risk score in the Clinical Analytics Drawer should not be misinterpreted as diagnostic. To facilitate an appropriate workup for a condition in question, a provider is able to access a Diagnostic Guide that provides a graphical depiction of the steps and choices available to a provider while making decisions regarding a diagnostic workup. For each available option, the Guide calculates a utility score that reflects the potential contribution to making a diagnosis that a particular choice may offer. A utility score is a reflection of the benefit an intervention might have for achieving a particular clinical goal, for example, diagnostic or economic utility. When a provider selects one of the options presented, the system displays a summary describing the intervention, its potential relevance for the workup, and whether the choice is available at the local facility, in the local community, or both.
Resource Capacity Simulator: The Resource Capacity Simulator provides tools for the healthcare administrator to optimize the allocation and scheduling of resources. It helps health care administrators to predict the incidence of disease given specific what-if scenarios. The application can then determine the maximum number of patients that a facility can accommodate based on the resources available in the scenario. The Resource Capacity Simulator provides the optimal allocation of staff, hospital beds, equipment, etc., determining what additional resources may be required to care for the simulated demand.
CDS Workbench: The CDS Workbench is an early prototype of a graphical tool enabling non-technical Subject Matter Experts to author domain knowledge (rules, guidelines, and workflow). The Workbench simplifies the creation of complex content by representing various clinical tests, activities, and procedures as objects in a drag-and-drop graphical editing environment. The Workbench compiles their work for storage and later retrieval and use by the runtime system.

CCA is an ambitious attempt to deliver high quality, advanced clinical decision support using commodity, open source infrastructure. It is a platform for researching how information technology can help change patient / provider behavior to better achieve and sustain health.