9.3 Assessing Stakeholder Learning Needs & Specialty Curricula
Key Takeaways
Investigate the actual performance gap before choosing training; workflow, access, staffing, and template problems may need other solutions.
Combine reviewed query, denial, reconciliation, audit, and learner evidence; none independently proves a diagnosis or the cause of a gap.
Prioritize using transparent sample definitions, counts, rates, severity, and feasibility rather than a presumed universal 80/20 distribution.
Tailor the task and examples to the audience while verifying clinical evidence and classification rules as separate source tracks.
Write observable learning objectives, design matching assessments, pilot the material, and evaluate accurate workplace application.
Assessing Learning Needs and Planning a Relevant Curriculum
An educational needs assessment asks what people need to do, what is happening now, and why a gap exists. It should precede the choice of a lecture, pocket card, or online module. A documentation problem may reflect missing knowledge, a confusing process, an inaccessible inbox, an incorrect template, or inadequate time and coverage. Education is useful when learning can address the cause; a class alone will not repair a broken delivery route.
CDC’s training needs guidance provides a practical planning framework: gather relevant information, identify causes with the affected participants, and determine whether training is an appropriate solution. Its Quality Training Standards apply as a CDC benchmark, not as a federal requirement that every hospital CDI class receive CDC approval. The following workflow applies those planning principles to CDI.
Define the Question and the Audience
Start with an observable problem. “Improve all physician documentation” is too broad to guide case selection or evaluation. “Determine why the discharge diagnosis remains inconsistent with the final consultant assessment in reviewed records” identifies a task that can be investigated. Specify the setting, staff roles, record types, review period, and evidence needed to establish the gap.
Include the people who perform the work. A resident, attending, advanced practice provider, coder, case manager, and CDI specialist may encounter different parts of the same problem. Ask where responsibility changes, what information is available at that point, and what would make the next step reliable. Learners’ practical access, schedules, existing knowledge, and preferred opportunities for practice inform the plan; job title alone does not establish what they need to learn.
Combine Evidence Rather Than Trust One Signal
| Evidence source | What it can help reveal | What it does not establish alone |
|---|---|---|
| Reviewed query records | Repeated terminology questions, routing failures, unclear responses | Every queried condition was clinically present |
| Denial and appeal records | Stated payer reasons and recurring documentation disputes | The payer’s clinical or coding interpretation was correct |
| Working/final coding reconciliation | Reasons for differences between provisional and final classification | The original working DRG was the correct target |
| Record audits | Whether the intended documentation behavior is actually present | A causal explanation without further investigation |
| Interviews or surveys | Workflow barriers and confusing instructions | Their frequency across all encounters without suitable data |
Use exact, reviewed examples. A denial described as “unsupported diagnosis” may involve conflicting documentation, a payer-specific policy, or an appealable interpretation. Verify the underlying facts and applicable authority before turning the denial letter into a clinical teaching rule. Similarly, an authenticated query response can support coding without repeated documentation elsewhere; do not teach duplication merely because an internal export failed to locate it.
Prioritize With Transparent Counts and Denominators
An illustrative review classifies 50 query-related issues as 28 acuity questions, 12 routing problems, six duplicate requests, and four other issues. Acuity questions account for 28 / 50 = 56% of this group. Rank the categories, review severity and consequences, and decide which causes can be addressed. A Pareto chart can display the distribution; it does not prove that 20% of conditions always cause 80% of queries.
These counts describe issues in this reviewed sample. They are not rates per discharge, and one encounter might have more than one issue unless the method prevents it. A service with more complex cases or more intensive review may generate more queries without having worse documentation. For comparisons, define the relevant eligible encounters and calculate a rate using the same method before and after an intervention.
Priorities should reflect both frequency and potential harm. A rare but important ambiguity in an operative report may deserve attention even when its count is below a common routing problem. Document the selection rationale and assign the technical fix to its responsible owner rather than adding every problem to the education agenda.
Build Specialty Content Around Verified Tasks
Tailor examples to the work while preserving the same standards of neutrality and evidence. Specialty relevance does not mean teaching every member of a discipline the same diagnosis list or guaranteeing that one format is superior.
| Audience and example task | Useful curriculum focus | Source and interpretation safeguard |
|---|---|---|
| Hospital medicine: describe a diagnostic conclusion and its uncertainty | Distinguish symptoms, established diagnoses, unresolved causes, acuity, and relevant relationships | Verify current clinical evidence separately from coding rules; a treatment or test value does not independently assign a diagnosis |
| Critical care: explain a procedure and its timing | Capture the actual service, anatomy, approach, device, and relevant start/stop events | Verify the applicable procedure classification and payment definition rather than assuming all duration thresholds coincide |
| Surgery: explain an event during or after care | Clarify the condition, relationship to care, and clinical significance | I.B.16 does not require the literal word “complication”; classification instructions and the record govern |
| Cardiology: reconcile evolving descriptions | Communicate the provider’s final interpretation and meaningful specificity | Do not infer a diagnosis or phenotype solely from a number or medication |
| Trainees and rotating staff: maintain an accurate record | Reconcile copied statements, current assessments, unresolved findings, and discharge communication | Teach verification of carried-forward text and the actual local responsibility for updates |
Clinical criteria belong in a reviewed clinical lesson with the framework’s population, date, limitations, and purpose stated. A payer’s preferred threshold is not automatically a universal diagnostic definition. Coding conventions answer a different question: how to classify supported documentation. Keep the two source tracks visible when preparing a lesson on respiratory failure, sepsis, kidney injury, malnutrition, or another diagnosis.
Turn the Gap Into a Measurable Objective
A useful objective names the learner, the task, the conditions, the expected performance, and the timeframe. For example: “By the end of the workshop, given three fictional discharge records and the applicable guidance, participants will identify the unresolved conflict and select an appropriate clarification step in all three.” The numbers describe this illustrative course, not a national passing standard.
Build the exercise and scoring criteria at the same time. If the objective concerns recognizing a conflict, assess that skill rather than recall of a reimbursement weight. Include cases where no query is needed and cases where a proposed query should be corrected. A curriculum that rewards only adding diagnoses can reinforce the wrong behavior even when knowledge scores rise.
Test the Plan Before Expanding It
Ask a clinical subject matter expert and the relevant coding or compliance specialist to review the material within their areas of expertise. Check source applicability, edition, examples, and answer keys. A clinician’s review of pathophysiology does not substitute for checking an official coding instruction. Pilot the material with intended learners and revise confusing wording or inaccessible delivery.
Record who will maintain the lesson and what changes should trigger another review. Evaluate both whether learners can perform the task and whether later records show the intended behavior. A decrease in queries may be useful evidence, but it can also reflect fewer reviews or an altered query threshold. Interpret it alongside consistent record sampling and measures of accuracy. The outcome of needs assessment is a defensible plan with a clear task and evaluation method, not merely a calendar of presentations.
Source anchors: CDC training needs analysis; CDC learning objectives; FY 2027 ICD-10-CM guidelines, I.A.19, I.B.14, and I.B.16; 2026 query practice brief.
A needs assessment finds that most delayed queries for one group were delivered to an inbox the covering clinicians cannot access. Which initial response best addresses the evidence?
Require a lecture on diagnostic specificity and leave delivery unchanged so knowledge improves before the workflow is altered.
Classify the delays as unsupported clinical diagnoses and prioritize training on severe conditions.
Repair the delivery and coverage process with its responsible owners, and assess separately whether a knowledge or skill gap also needs education.
Judge education effective if agreement rates rise, without checking whether messages reach the appropriate clinician.
A program classifies 50 reviewed query issues, including 28 acuity questions. Which conclusion is supported?
The acuity-query rate is 56 per 100 discharges across the entire hospital.
The figures prove that 20% of all diagnoses generate 80% of all queries.
All 28 queried conditions were clinically present because they were included in a formal needs assessment.
Acuity questions form 56% of these 50 issues; comparison with other services requires appropriately defined populations and denominators.
A surgical documentation course states that no complication code can be reported unless the surgeon writes the literal word “complication.” What correction follows FY 2027 ICD-10-CM guideline I.B.16?
Teach the documented relationship and clinical significance, subject to classification instructions; the literal word is not required, and unclear relationships may require a query.
Retain the statement because a complication code always requires the same mandatory phrase regardless of the rest of the record.
Teach that every event after surgery is a complication if it increases length of stay.
Teach that the coder can determine the relationship solely from a transfusion or medication order.
Sections you finish are checked off in the contents.