Most electronic health record (EHR) onboarding is built around the system: here is the login screen, here is the menu, here is the template for every note type. The new starter leaves with a certificate. The documentation problems start on the ward the following week.
This guide is for hospital learning and development (L&D) directors and informatics or compliance leads who want a different starting point: the workflow the person must complete, and the friction they will meet while completing it. It shows how to diagnose whether a problem is a training gap or a system defect, how to design preparation and practice around a real task, and how to assess whether the new starter can complete it, with documentation that is complete and compliant.
A situation worth designing for
The scenario below is illustrative, not a documented case.
A new ward clerk joins on a Monday. By Thursday they are asked to complete a routine patient record update after an admission. The task is simple on paper: confirm demographics, record the referral source, attach a consent form and finalise the entry. In practice they meet four obstacles. The consent form is attached on a different screen from the one the classroom session showed. Two fields look identical but feed different reports. A mandatory field rejects a valid entry with an unhelpful message. And the local procedure for a missing referral source is written in a document nobody mentioned.
The clerk completes the entry, but leaves one field blank and uses a free-text workaround for another. The record is incomplete, and the next person to rely on it inherits the gap.
Is this a training problem? Partly. Only a training problem? Almost certainly not. That distinction is the heart of the design work.
Start with the task, not the system
Write down what the role must be able to do, in one sentence, before you open any training content. For this example: complete a patient record workflow end to end, and identify where the documentation process creates friction. Two verbs matter here, and they are different.
Completing the workflow is an sich bewerben task. The learner uses a known procedure in a situation they have not rehearsed word for word. Identifying friction is an analysieren or bewerten task: the learner has to notice that something is wrong, judge whether it is their error or the system's, and decide what to do about it. These cognitive demands come from Bloom's revised taxonomy, and they describe what the task requires of the person, not how you must deliver it.
This matters because of alignment. Carnegie Mellon's Eberly Center makes the point plainly: if the objective is to apply skills but the assessment only measures recall, learners practise one thing and are judged on another, which undermines both motivation and learning (Eberly Center, alignment of objectives, assessments and instructional strategies). A multiple-choice test on menu names tells you little about whether someone can finish the admission update above.
Diagnose before you prescribe
Before building anything, compare the task as it is really performed with what the current onboarding teaches and tests. Four checks tend to separate training gaps from other causes:
- Observe the task. Watch several starters or experienced staff complete the workflow, and record errors, workarounds and hesitation.
- Compare with the training. Does the current programme practise this exact sequence, or the system in general? Does any assessment require the learner to complete it?
- Überprüfen Sie die Umgebung. Are the tools, access permissions, staffing levels and local procedures what the training assumed? A new starter without the right access level cannot practise the task, whatever the classroom covered.
- Look for system defects. If experienced staff also struggle, or use the same workaround, the cause is probably the design of the screen, the field or the rule. More training will not fix it. Raise it with informatics as a change request.
This is the pitfall to name openly with your stakeholders: usability defects may need system changes, not more training. If training is used to paper over a defect, the cost returns in every cohort. Treat your current method and the ceiling of your current assessment as hypotheses to test, not facts.
Design preparation and practice around the workflow
Once you know which gaps training can close, match each stage of the programme to the task.
Vorbereitung. Check prerequisites with a short diagnostic before a person reaches practice. Confirm they know the terms they need, such as the difference between a registration field and a clinical field, and the regulatory rules that bound the task, such as what must be recorded and who may record it. Self-paced study suits this stage. A recall check here confirms prerequisites only. It is not evidence that the person can do the work.
Workflow mapping. Map the real workflow screen by screen, including the steps people skip and the points where the system and the local procedure disagree. Use the map to build the practice case and the answer key, and give it to your informatics or compliance lead to validate. The same map becomes a friction audit: every point where observers saw errors or workarounds is a candidate for a system fix, a clearer job aid or a practice case.
Praxis. Give learners the whole task, with a realistic case, and ask them to explain their decisions as well as perform them. Feedback should link to the decision, for example: "You attached the consent form to the wrong record type. Here is the evidence you could have checked, and here is what to do when the form type is unclear." Feedback on speed and completeness works best when it is specific to the step.
Live or asynchronous? Live instructor input adds most value when a choice is ambiguous: when two defensible actions exist, or when the procedure and the system disagree. A discussion led by someone who knows the local practice resolves what a written answer key cannot. Asynchronous review is enough for confirming rules, checking recall and giving feedback on a clear-cut case.
A worked example: the admission update
Illustrative. A practitioner should validate the case and the answer criteria before use.
Der Fall. The learner receives a simulated record for a fictitious patient. They are asked to complete the admission update. The case includes a scanned consent form, a referral letter with a missing reference number, and a field that rejects the entry format shown in the training manual.
The learner's choices. At the missing reference number, the learner can (a) leave the field blank and finalise, (b) type a placeholder, or (c) record the gap using the approved procedure and request the missing information. At the rejected field, they can (a) retry with a different format until it accepts, (b) put the value in a free-text note, or (c) log the issue as a suspected system fault and continue with the approved interim step.
Feedback bekommen. For each choice, the feedback explains why it is defensible or incomplete against the agreed criteria. A blank field is incomplete, because downstream users cannot tell whether the data is missing or was never needed. A placeholder is not defensible, because it records information that is not true. Recording the gap under the approved procedure is complete. In the second decision, trying formats until one is accepted may produce a valid entry, but it hides a defect the informatics team needs to see. Logging the fault is the response that protects both the record and the next cohort.
The retry. The learner attempts a second case with a different gap, such as a consent form that is attached but unsigned, and applies the same reasoning to it.
Across a cohort, the learners' friction reports also become a structured list of suspected defects for the informatics team.
Assess capability and plan for transfer
Pick measures that reflect the task. Two practical ones are task time und Vollständigkeit der Dokumentation. Define both before you start: what counts as complete, who judges it, which records are in the denominator, and when you measure.
A simple rubric keeps the judgement consistent. The one below is a proposal to adapt with your clinical or compliance specialist. It is not a validated readiness standard.
| Kriterium | What an observer looks for | Score |
|---|---|---|
| Aufgabenausführung | Completes the workflow and identifies documentation friction, recording the action and the evidence behind it | 0 bis 2 |
| Argumentation | Explains the relevant constraints, alternatives and uncertainty; checks task time and documentation completeness | 0 bis 2 |
| Boundaries | Uses approved procedures and asks for help when information or authority is missing | 0 bis 2 |
Anchors: 0 means absent or unsupported; 1 means partial, with relevant omissions; 2 means complete and justified against the agreed criteria. Define critical errors separately, for example recording information that is untrue, or bypassing a required consent check. An overall score must never hide a critical failure.
For transfer, assess a new, unseen case rather than the one used in practice, and follow up in the workplace after the learner has had enough exposure to real work. The Centers for Disease Control and Prevention (CDC) recommend evaluating both learning and learning transfer where possible, because a person who passes in training may still struggle to apply it on the job (CDC, Evaluate training: measuring effectiveness). A supervised work task or a review of a sample of the learner's real entries by a preceptor can serve as that follow-up.
If your onboarding involves interprofessional handovers, such as a registration team passing records to clinical staff, the Agency for Healthcare Research and Quality's TeamSTEPPS programme is a starting point for healthcare teamwork and communication resources. Have a clinical educator choose and adapt the relevant tools.
Was die Beweise zeigen können und was nicht.
Be clear about what you can claim. These are the questions your audit should answer.
What would show that learners can complete the workflow and identify friction? Rubric scores on an unseen case, task time and documentation completeness compared with a baseline, and the quality of the friction reports they submit. Compare like with like: the same population, the same denominator and the same timing.
Which constraints could remain even after performance improves? Several. Defective screens and rules remain until someone changes the system. Staffing, workload and access permissions shape how much time a person has to document properly. Local procedures that conflict with the system stay in conflict. And concurrent operational changes, such as a new template or a software upgrade, can move your measures independently of training. Examine them before attributing any change to the programme.
The sources above support the design principles of aligning objectives, practice and assessment and of measuring learning and transfer. They are not evidence that any particular programme, or any platform, reduces documentation burden. This article makes no outcome promises, and your own evaluation should not either.
Where AhaSlides can fit
AhaSlides supports adaptive and interactive learning, and lets you move between live and self-paced delivery. In an onboarding programme like the one above, that could mean a self-paced prerequisite check before practice, an interactive facilitated session for the ambiguous decision, and a self-paced review of a new case afterwards. Specific authoring, feedback and reporting functions should be checked against your own workflow in a demonstration before you plan around them. Any AI, simulation or EHR integration should be treated as external or unconfirmed until demonstrated.
Dein nächster Schritt
Start with the task rubric: take one high-volume workflow in your hospital, draft the three criteria above with your informatics or compliance lead, and observe two or three real attempts. The friction you record is your list of training gaps and system defects, in that order of separation.
When your design is ready, we can show you how a live and self-paced version of the practice and review stages would work. Request an AhaSlides demo and bring your workflow.
