Why the Continuity of Care Document (CCD) Is Not Effective For Transitions of Care

The Continuity of Care Document (CCD) is a summary that contains data about patient’s clinical information. It is used to exchange information between providers, but doctors don’t find it valuable as information is not contextualized nor prioritized by importance.

Ten years ago, the United States poured billions of dollars into health information exchanges (HIEs) across the country in an effort to fix interoperability between Electronic Medical Records (EMRs). A lot of research has shown that reviewing a patient’s health record before a visit helps improve health outcomes; HIEs are very effective in that regard when providers take the time to review the data. And while overall interconnectedness of HIEs has helped connect doctors in the US, most providers do not take the time to review a patient’s historical record before a visit. So the data is there, but most have not found much value when using HIEs.

The need of contextual information

A common complaint that I have heard from doctors is that the Continuity of Care Document (CCD) generally lacks context and trust. The CCD is the healthcare summary that is shared between providers through HIEs to create a handoff for care (it is the most common document shared by HIEs). The summary contains data such as patient demographics, history, medications, allergies, and lab results. What the CCD generally does not include is unstructured data such as notes. So when a doctor is reading through a CCD, they do not understand why there is an allergy marked for penicillin; keep in mind that while 10% of people in the United States have a reported allergy marked in the EMRs for penicillin, fewer than 1% of the population actually does. An improperly marked penicillin allergy prevents people from getting the most effective medications for their treatments. Your health record will just state you are allergic to penicillin (there is no context to the information)- and doctors are supposed to just take it as an accepted fact that everyone else correctly confirmed all of the prior data. The reality though is that if one doctor accidentally enters an allergy into your record from 10 years ago — it will be copied forward for perpetuity and nobody has the time to confirm it. One cause of this issue is that the data in the CCD record is all weighted equally — if a team of specialists performed a litany of tests to determine you have an allergy to penicillin, it would carry the same weight as any other doctor entering the allergy. In reality, doctors really depend on that story and context to trust the data.

Doctors are supposed to trust a patient summary with no context and reasoning for how the data was decided and entered. In practice, doctors are barely reviewing that historical CCD data and will spend their limited time instead reviewing a recent clinical note in your chart, if available, from a high-fidelity source such as a transition of care note from a hospital or a progress note from a primary care doctor that they know.

The Future of CCDs

A number of startups are structuring the CCD data and historical information in a holistic manner. For example, Health Gorilla has a product called Patient360 that aggregates all the data into a dashboard. While this solves the problem of not needing to wade through 10 versions of the same CCDs, it still lacks providing meaningful context for how a patient got pneumonia or why there is an allergy for penicillin. For that context, you will still be expected to review or search through numerous clinical notes. Nobody is really summarizing yet the hundreds of pages of unstructured notes. This is what is needed for healthcare to create a more meaningful summary — using AI and NLP to create a summary with context that is missing with the CCD architecture.

Abstractive Health provides an automated narrative summary of the medical record as a software solution for healthcare. We use a natural language processing algorithm to summarize the clinical notes in the patient chart. We currently have a partnership with Weill Cornell where we are demonstrating the clinical quality of our automated hospital summaries compared to the hospital course section of the Discharge Summary.