OncoEMR <> DeepScribe: Troubleshooting FAQs

No, you do not need to create the patient’s visit note in OncoEMR before beginning a recording in DeepScribe. However, you do need to create the note to start a session import, as the import will not be able to automatically create the new note.

Phone calls will interrupt recordings, even if that call is unanswered. If an interruption occurs, tap the red microphone to continue the recording. However, we recommend putting your phone on Do Not Disturb or enabling a Focus Schedule to avoid disruption during patient visits.

Yes, you can still type content into your visit note during a patient visit even while DeepScribe is recording in the background. When you import your AI-generated content from DeepScribe into OncoEMR, the import button will automatically save your progress before pulling in the DeepScribe content to ensure that nothing is lost.

This indicates that although the note is complete, the note has not been uploaded to the EHR yet. When this occurs, DeepScribe will be notified automatically, and will retry sending that note to OncoEMR until it is successfully uploaded. You also always have the option to copy and paste from DeepScribe into your OncoEMR Visit Note.

No, once you’ve hit “Scribe it” in DeepScribe to generate the visit note for a recording, you are not able to re-generate it.

On the backend, DeepScribe first identifies an OncoForm as a Visit Note if “Show in Visit Note list” is selected in the Advanced tab of the OncoForm editor. To use as context, DeepScribe will pull in the last visit note for that patient that was signed or finalized by the user.

First, confirm that the date of the Visit Note in OncoEMR matches the visit date of the DeepScribe session.

Second, confirm that you are using a visit note template with AI-enabled sections.

Next, confirm that the patient has session recordings in DeepScribe from the last 5 days that have not already been imported into OncoEMR.

Lastly, confirm that you have permission to edit Visit Notes in OncoEMR.

If you still do not see the import banner appearing after confirming these three things, please reach out to the OncoEMR Support team.

The integration logic will automatically import the right session from DeepScribe and populate note content into the appropriate visit note section when:

For more complex scenarios, the user performing the import in OncoEMR will be prompted to select from a list of available sessions to ensure the correct DeepScribe session is pulled in. These more complex scenarios include:

In the vast majority of cases, you will be able to import the AI-generated content directly into the corresponding sections of your visit note with the click of a button. You will only need to copy and paste the note content into OncoEMR if:

We do not currently support “undoing” an import, so you will need to manually delete the AI-generated content that you wish to remove. The content you imported should be highlighted light purple. If you have feedback to share about supporting this use case, please let us know.

A “partial import” may occur if your DeepScribe note configurations don’t match your OncoEMR note template. For example, if you included a field in your DeepScribe note configuration but the corresponding field is not included in your OncoEMR visit note, that AI-generated content will not be pulled in. You can leverage this Configuration Guide along with any additional practice-specific guidance from your EHR administrator to confirm that your Customization Studio settings are correct.

It is possible that you temporarily lost internet connectivity, so you can click the Import content button to try the import again.

If the visit note template in OncoEMR includes External field types (i.e. AI-enabled note fields) but none of the fields match the imported session, you will still see the error message. In this case:

Not at this time; each DeepScribe session has to be tied to a unique visit note so it cannot be associated with multiple notes. If you have feedback to share about supporting this use case, please let us know.

However, if DeepScribe regenerates a visit note on their end for any reason and you want to import the regenerated content, you can do so by superseding the note in OncoEMR.

Not at this time; if you have feedback to share about supporting this use case, please let us know.

Yes, you can create multiple recordings for the same patient encounter. However, when you click “Scribe it,” all recordings will be synthesized into a single visit note.

We don’t support multi-user workflows at this time, but DeepScribe is currently building functionality to support scenarios that would allow both MDs and their care team members to separately create recordings for the same patient. For now, the available workaround is for both clinicians to share the same recording device.

DeepScribe surfaces ICD-10 codes by analyzing the provider–patient conversation and identifying clinically relevant terms, symptoms, diagnoses, and impressions using proprietary natural language processing models. These concepts are then mapped to ICD-10 codes using the publicly available Clinical Tables API from the National Library of Medicine. In some cases, DeepScribe may also "pull forward" previously used or historically relevant ICD-10 codes based on context within the current conversation to ensure clinical continuity.

The diagnoses will be routed to either the Oncology/Hematology grid or the Other Medical Problems grid, depending on the type of diagnosis. Diagnosis types of primary, secondary, and history will be routed to the Oncology/Hematology grid. All other diagnosis types will be routed to the Other Medical Problems grid.

In addition to appearing in the visit note diagnoses grids, the new diagnoses will also appear in these same grids on the Patient Summary page. Information in these diagnoses grids are synced between the visit note and the summary page.

It is possible that you temporarily lost internet connectivity, so you can click the Import diagnoses button to try the import again.
​

The disease field is optional, it is not required to save or import your diagnosis. However, it is required for staging. If you already know you will be staging the diagnosis, it may save you time later to go ahead and add the disease field prior to importing. To enter stage details, please navigate to the Onc/Hem grid and follow your practice policies for adding details.

There are two options. First, you can click on the “x” button next to the diagnosis. This will remove it from the list of available diagnoses. Second, you can simply ignore any diagnosis you don’t wish to import. OE will only import diagnoses whose details have been completed and are in the “Ready to import” section.

You can do so in your DeepScribe settings. Turn off the diagnosis import feature by disabling the “ICD-10” toggle under the “Assessment” section of DeepScribe’s Customization Studio.

Not at this time. Currently, once you import diagnoses from DeepScribe, you cannot import any more. This helps prevent accidental duplication. In the diagnosis workflow, your progress is saved when you hit “save details” and see the diagnosis in the “Ready to import section.” All diagnoses in that “Ready to import” section will be imported to the visit note when you hit the “import diagnoses” button.

Certain fields—specifically, status and status date—are now required when adding a diagnosis as a part of a broader initiative to modernize the diagnosis and staging workflows (releasing more broadly in the second half of this year) to ensure that every diagnosis captures the most clinically relevant and up-to-date information.

Making these fields mandatory helps improve data quality, supports downstream workflows (like treatment planning and quality reporting), and ensures that clinicians and care teams have a clear, consistent understanding of a patient’s disease state at any point in time. This change was also made in response to user feedback highlighting the importance of having a discrete, up-to-date status for each diagnosis, and to address gaps in documentation completeness that previously made it difficult to get a full picture of a patient’s disease history or to run accurate reports. By making these fields required, we’re supporting better clinical decision-making and more reliable data for both patient care and practice operations.

As you create new AI-compatible templates, you should ensure that the datanames match across all note templates in order for carry forward to continue working across both AI- and non-AI compatible templates. For example, if you want the Plan content to carry forward to new notes, whether providers are using an AI-enabled or regular note template, you should ensure the dataname for each of the Plan components is the same across both templates.