AI Hospice Coding: ICD-10 Codes Suggested From the Chart
Lime reads the hospice chart and suggests the principal diagnosis and related conditions, with the evidence behind each code. It flags codes the record does not support. Your coders decide.
The Short Answer
AI hospice coding suggests ICD-10-CM codes from the hospice chart for a coder to approve. In hospice the codes do not set the payment rate. They support eligibility: the principal diagnosis and related conditions have to match the terminal prognosis in the certification. Lime drafts the code set, shows the evidence, and flags the gaps.
What Hospice Coding Has to Get Right
- Principal diagnosis. The condition contributing most to the terminal prognosis. It should agree with the physician's certification narrative.
- Related conditions. CMS expects the claim to list the other diagnoses related to the terminal illness, not the principal diagnosis alone.
- Codes Medicare returns. Since October 1, 2014, claims with debility (R54) or adult failure to thrive (R62.7) as the principal diagnosis come back to the provider.
- Dementia sequencing. Manifestation codes such as F02.80 fail as principal diagnosis. The underlying condition, such as Alzheimer's disease, goes first.
- Specificity. Each code at the highest level of detail the record supports.
- Annual updates. The ICD-10-CM code set changes every October 1.
A wrong code rarely changes the hospice daily rate. It weakens the eligibility case an auditor reads, and it sends claims back.
How Lime Helps Your Hospice Coders
- AI-suggested codes. Lime reads the clinical documentation and suggests ICD-10 codes, with the chart evidence behind each one.
- Flags before submission. The AI flags missing or potentially incorrect codes, so your coders fix them first.
- Sequencing support. The AI proposes a sequence, and your coders confirm it.
- Documentation prompts. When the chart lacks detail for a specific code, the AI prompts the clinician while the visit is fresh.
- Your team decides. Lime suggests and flags. Your coders make every final decision.
Coding Starts With the Documentation
Weak documentation produces weak codes. Hospice Scribe drafts the visit note from the patient conversation, so your coders start from a fuller record. Sentinel QA reviews every submitted note against the full chart and checks certifications, face-to-face encounters, and IDG notes. Use one product or all three. See the hospice page.
Related
FAQ
AI Hospice Coding, Answered
What hospice coding and compliance leaders ask.
AI hospice coding is software reading the hospice chart and suggesting ICD-10-CM diagnosis codes: the principal diagnosis and the related conditions. Lime shows the chart evidence behind each suggestion and flags codes the documentation does not support. Your coders review every suggestion and make the final call.
Home health coding sets the PDGM clinical group and comorbidity adjustment, so the codes change payment for each 30-day period. Hospice is paid a daily rate by level of care, so the codes do not set the rate. They support eligibility instead: the principal diagnosis and related conditions have to match the terminal prognosis the physician certified.
Since October 1, 2014, Medicare returns hospice claims listing debility (R54) or adult failure to thrive (R62.7) as the principal diagnosis. Manifestation codes, including many dementia codes such as F02.80, also fail as principal diagnosis. The underlying condition goes first.
No. The AI drafts the code set and points to the evidence. Your coders confirm the principal diagnosis, the related conditions, and the sequence, and query the clinician or physician when the record is unclear.
Lime's AI coding is software for your own coders. It does not include a guaranteed human review by Lime. Agencies short on coding capacity ask about Lime's virtual coding staff as a separate service.
Lime uses per-chart volume pricing, so the rate depends on your chart volume. Pilots are available.
See AI-suggested ICD-10 codes on your own hospice charts.
Request a Demo