The importance of variant data for clinicians and patients
Knowing if a patient is infected with a SARS-CoV-2 variant has significant clinical importance. For one, specific treatments for COVID-19 – such as monoclonal antibody infusions – may not work as well against certain variants. Additionally, some variants could be more virulent in patients with certain pre-existing conditions. Clinicians need to know what they’re dealing with in order to better treat and monitor their COVID-19 patients.
However, the current process for sequencing and reporting SARS-CoV-2 data (see Part III) often severs the link between the sequence and the patient. As a result, data associated with the patient’s record is unavailable to researchers studying the variants, and the variant information is not accessible to clinicians who are treating the patient.
Preparing the EHR for variant data
Currently, there are no diagnostic tests for SARS-CoV-2 variants, however, that may soon change as newly developed tests gain approval from the FDA. When that happens, electronic health records (EHRs) need to be ready to store and manage sequencing data in a way that makes it useful and readily understood by clinicians and patients alike.
SARS-CoV-2 sequencing data is highly complex, containing elements at multiple levels of specificity – such as phylogenetic lineage, relevant mutations, and spike protein substitutions. This is a wide range of metadata, which is used to convey important information about the strain. While this scientific information may be useful for virologists, in order to serve clinicians it needs to be connected to patient data in the EHR. It must also contain the right level of specificity and meaning to allow both clinicians and researchers to link variant status directly to patient outcomes.
As discussed in Part II, there are multiple terminologies being used to refer to SARS-CoV-2 variants. For clinical research, for example, PANGO lineage could allow researchers to link specific outcomes – such as breakthrough infections – to specific strains which share similar point mutations. On the other hand, for treatment, more basic terminology, like “Delta” or “variant of concern”, could be used to guide patient care – such as choosing which monoclonal antibody treatment to administer. Robust, clinical interface terminology solutions can help translate this highly technical variant data into the appropriate terminology so that it can be properly linked to patient data in systems such as the EHR.
Conclusion
The recent surge in COVID-19 cases across the country has been largely fueled by the highly infectious Delta variant. Given the devastating impact of Delta, and the possibility that new variants could be even more infectious and virulent, it is critical that treating clinicians know which strain their patient has. Once approved, new diagnostic tests that can identfy variants will be another important tool for providers, but to make this information truly valuable healthcare institutions must prepare their EHRs to handle variant data in the form of clinically relevant terminology.
In this blog series, we have aimed to demonstrate the importance of integrating viral sequencing information with clinical data. There are a variety of use cases for this data – such as monitoring, clinical research, and patient care – however, to be useful this data must be translated into the right format at the right level of specificity. This can be achieved using terminological harmonization and ontologies, which would map data elements to a terminology that is meaningful for each use case, potentially (and hopefully) unlocking embedded sequencing information to better prepare us in the fight against future SARS-CoV-2 variants.
