Where MDS Assessment Time Really Goes: A Breakdown for Coordinators
The MDS Coordinator Time Crunch
Ask any MDS coordinator how long an assessment takes, and you'll hear a consistent answer: 3-4 hours. Some facilities report faster times for straightforward cases; complex residents with multiple comorbidities can take significantly longer.
For a coordinator responsible for 20-30 assessments per month, that's 60-120 hours of MDS work—often squeezed between meetings, interruptions, and the constant pressure of ARD deadlines.
The consequences are predictable:
- ARD deadline pressure leads to rushed assessments
- Documentation gaps result in under-coded PDPM components
- Staff burnout drives turnover in a role that's already hard to fill
- Compliance risk increases when there's no time for thorough review
Understanding where assessment time actually goes is the first step toward finding solutions—whether that's workflow improvements, additional staffing, or technology assistance.
Who This Guide Is For: MDS Coordinators, Directors of Nursing, Administrators, and anyone evaluating ways to improve MDS efficiency.
Where Assessment Time Actually Goes
MDS assessment time falls into five categories:
1. Documentation Review
Reading through clinical notes, physician orders, therapy evaluations, nursing assessments, and hospital records. For a new admission with a thick discharge packet, this alone can take an hour or more.
2. Data Extraction
Finding the specific clinical data needed for each MDS section. This means hunting through pages of notes to find the right diagnoses, functional scores, medications, and conditions.
3. Resident Interaction
Conducting required interviews: the BIMS for cognitive status, PHQ-9 for mood, pain assessments, and other sections that require direct resident contact. This time is relatively fixed—you can't automate a conversation.
4. Section Completion
Entering data into the MDS system, selecting appropriate codes, and ensuring consistency across sections.
5. Validation
Cross-checking entries, resolving discrepancies between sections, reviewing for errors, and ensuring the assessment will pass validation edits.
Key Insight: Categories 1 and 2—documentation review and data extraction—typically consume the largest portion of assessment time. These are also the tasks most amenable to technology assistance.
The Documentation-Heavy Sections
Not all MDS sections are created equal. Some require extensive documentation review; others have more predictable time requirements.
High Documentation Burden
Section GG (Functional Abilities and Goals): Under PDPM, Section GG directly drives PT and OT reimbursement. Accurate scoring requires reviewing therapy evaluations, nursing ADL documentation, and often multiple days of records to capture the resident's typical (not best) performance. This section alone often takes longer than several simpler sections combined.
Section I (Active Diagnoses): Mapping all relevant diagnoses to ICD-10 codes requires reviewing physician notes, hospital records, medication lists, and lab results. Missing diagnoses means missing NTA points—and potentially significant reimbursement.
Section J (Health Conditions): Pain assessments, fall history, and other health conditions require pulling data from multiple documentation sources and time periods.
Section N (Medications): Medication reconciliation requires reviewing pharmacy records, physician orders, and MAR documentation to capture all relevant medications received during the lookback period.
More Predictable Time Requirements
Section C (Cognitive Patterns): The BIMS interview has a standardized format—time is primarily interview time, not documentation review.
Section D (Mood): The PHQ-9 is a structured interview with consistent time requirements.
Section A (Identification): Administrative information that's typically quick to complete.
Reduce Documentation Review Time
MDS Genie™ extracts clinical data from your documentation and pre-populates Section GG, diagnoses, and more—designed to cut assessment time by up to 60%.
Learn About MDS GenieWhat AI Can and Cannot Do
AI-assisted MDS tools are designed to reduce the documentation burden—not replace clinical judgment. Understanding what AI handles well vs. what requires human expertise helps set appropriate expectations.
Where AI Can Help
- Documentation extraction: Finding specific clinical data buried in pages of notes, therapy evaluations, and physician orders
- Diagnosis mapping: Converting clinical terms to ICD-10 codes and identifying NTA-relevant conditions
- Consistency checking: Flagging potential discrepancies between sections or documentation
- Pre-population: Filling in items based on clear documentation for coordinator review
- Historical comparison: Identifying changes from prior assessments
What Still Requires Human Judgment
- Resident interviews: BIMS, PHQ-9, and other interview-based assessments require direct interaction
- Clinical interpretation: Complex cases where documentation is ambiguous or conflicting
- Care planning: Translating MDS findings into actionable care plans
- Final validation: Reviewing AI suggestions and making final coding decisions
- Appeals and audits: Defending coding decisions requires human expertise
The Right Mental Model: Think of AI as a research assistant that does the documentation review so you can focus on clinical judgment and resident care. It extracts and suggests; you review and confirm.
Evaluating MDS Automation
If you're considering MDS automation tools, here's what to look for:
Questions to Ask Vendors
- What sections does the tool address? Focus on whether it covers the documentation-heavy sections (GG, I, J, N) where time savings are greatest.
- How does it integrate with your workflow? Can it work alongside your existing MDS system, or does it require complex EHR integration?
- What's the review process? Good tools present suggestions for coordinator review—not automatic submission.
- How does it handle source citations? For audit purposes, you need to know where each data point came from.
- What's the security model? HIPAA compliance, BAA availability, and PHI handling policies matter.
Realistic Expectations
No tool eliminates MDS work entirely. Resident interviews still require face time. Complex cases still require clinical judgment. The goal is to reduce the documentation burden so coordinators can focus on what requires their expertise.
When evaluating time savings claims, ask how they were measured and whether they align with your facility's documentation practices and case mix.
See How MDS Genie Works
MDS Genie™ is designed to help MDS coordinators work faster without sacrificing accuracy. See how AI-assisted coding works with your documentation.
Schedule a DemoFrequently Asked Questions
How long does a typical MDS assessment take?
Most MDS coordinators report that a complete MDS 3.0 assessment takes 3-4 hours. This includes reviewing clinical documentation, conducting resident interviews, completing all sections, and validating the assessment. Complex cases with multiple comorbidities can take longer.
Which MDS sections take the most time?
Documentation-heavy sections typically take the longest: Section GG (Functional Abilities) requires cross-referencing therapy evaluations and nursing notes; Section I (Active Diagnoses) requires mapping conditions to ICD-10 codes; and Section J (Health Conditions) requires reviewing multiple sources. Interview-based sections like the BIMS and PHQ-9 have more predictable time requirements.
Can AI replace MDS coordinators?
No. AI can assist MDS coordinators by automating documentation review and data extraction, but clinical judgment remains essential. Resident interviews, complex case decisions, and final validation require experienced MDS professionals. AI is a tool to reduce administrative burden, not a replacement for expertise.
Where can AI help with MDS assessments?
AI is most effective at documentation-intensive tasks: extracting functional scores from therapy notes, mapping diagnoses to ICD-10 codes, identifying NTA-qualifying conditions, and cross-referencing information across multiple documents. These are tasks that require reading large amounts of text—exactly where manual processes are slowest.
What parts of MDS still require human judgment?
Resident interviews (BIMS, PHQ-9, pain assessments) must be conducted in person. Complex clinical interpretation where documentation is ambiguous requires professional judgment. Care planning, final validation of all entries, and audit defense all require experienced MDS coordinators.