Everyone in healthcare knows that interoperability, or more specifically Clinical Data Exchange (CDEx) is inadequate to meet stakeholder needs. The major stakeholders are Patients, Providers (clinicians and delivery systems) and Payers, including government agencies. 

Until now, we all have tried to improve on sourcing clinical data, by tapping into custom data feeds, getting standardized documents such as CCDs from aggregators and vendors or by even increasing the amount of chart reviews. In almost every case, those efforts are very heterogeneous, costly and still do not deliver the clinical data when and where it is needed. Often, we don’t even have a grasp on what constitutes having optimal clinical data.

The Velox team has spent the last years developing a solution to this increasingly important area we call Clinical Data Operations. Our Payer Enablement Platform (PEP) provides answers to all the pertinent questions about clinical data operations. Of course, before we could do that, we had to come up with a comprehensive, standardized framework that allows for inventorying, scoring and benchmarking both the status quo (clinical data operations as they exist right now) as well as clearly defining the desired state where all clinical data are available when and where they are needed at a cost that has a positive ROI. We have named this framework the CDEx Scoring Model.

Data Channels

The foundation of every standardized framework is to have a universal, easy to understand base unit that can be captured and scored. For the CDEx Scoring Model, we are using Data Channels. A data channel is defined as a data feed with a defined format (e.g. custom, CCD, FHIR) from a specific data source – more specifically from an authoritative source, i.e. where the data originated (provider system, lab).

Internal and External

The scoring model profiles both the external – constrained by what is available from data partners and how it gets to the requestor – and internal aspects (we sometimes refer to this as Clinical Data Operations or CDOps) – how those data are processed once they are in the control of a payers data operations – of each data channel. The latter is entirely in the control of the requestor and scores against best practices, speed and cost optimization.

Complexity Scores

Each channel has an overall score, which is a composite of the internal and external scores. The model captures a comprehensive set of standardized parameters. Some of those are:

• • Data format: format has implications for how difficult and complex processing the data is. The more standardized, especially when code sets are enforced, the better. The model scores accordingly.

• • Time-to-Data: the external parameters capture how often and how delayed data arrives at the payer. Depending on the use case, the timeliness of the data can be of moderate value or even critical. The internal parameters capture similar metrics.

• • Handoffs and transformations: data that can be captured in its native format directly from the source is less susceptible to errors and omissions that occur during handoffs and transformations (mapping) that can and often do happen when there are intermediaries. Therefore, the model considers fewer transformations and hops (handoffs) as better, both externally and internally.

Complexity scores are normalized on a scale from zero to ten. While a real-time FHIR feed can score as low as 1.0, a custom file that only becomes available every quarter and is provided by a third-party aggregator or vendor can score as high as 10.0.

The model also includes Weighted Complexity Scores. The weight is based on the number of members that have data flowing through the channel. The complexity of a channel that provides a large number of member data should weigh more heavily than one that only has a small number. The weighted score captures that in a standardized way.

Capturing the parameters that inform the complexity scoring is quick and typically uses existing data source inventory information (e.g. from HEDIS audit roadmaps). The model provides a compelling assessment of individual channels. It also rolls up scores and other KPIs across plans, markets and enables benchmarking against peers.

With the complexity scores, the CDEx Scoring Model provides a solid baseline.

The next steps are to determine what data is missing and how to best get to ‘complete’ clinical data. We will cover those topics in a future blog.